{
"cells": [
{
"cell_type": "code",
"execution_count": 2,
"id": "ced6196b-109c-489a-ab48-e65829ac6796",
"metadata": {},
"outputs": [],
"source": [
"from transformers import AutoTokenizer, RobertaForTokenClassification\n",
"import torch\n",
"from transformers import AutoTokenizer\n",
"from json import JSONEncoder\n",
"from transformers import AutoTokenizer, AutoModelForTokenClassification"
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "89c25949-7688-45f5-be0b-5bfe88bdbc05",
"metadata": {},
"outputs": [],
"source": [
"text=\"\"\"So, if you're a NASA scientist, you should be able to tell me the whole story about the Face On Mars, which obviously is evidence that there is life on Mars, and that the face was created by aliens, correct?\" No, twenty five years ago, our Viking 1 spacecraft was circling the planet, snapping photos, when it spotted the shadowy likeness of a human face. Us scientists figured out that it was just another Martian mesa, common around Cydonia, only this one had shadows that made it look like an Egyption Pharaoh. Very few days later, we revealed the image for all to see, and we made sure to note that it was a huge rock formation that just resembled a human head and face, but all of it was formed by shadows. We only announced it because we thought it would be a good way to engage the public with NASA's findings, and atrract attention to Mars-- and it did.\n",
"\n",
"The face on Mars soon became a pop icon; shot in movies, appeared in books, magazines, radio talk shows, and haunted grocery store checkout lines for 25 years. Some people thought the natural landform was evidence of life on Mars, and that us scientists wanted to hide it, but really, the defenders of the NASA budget wish there was ancient civilization on Mars. We decided to take another shot just to make sure we weren't wrong, on April 5, 1998. Michael Malin and his Mars Orbiter camera team took a picture that was ten times sharper than the original Viking photos, revealing a natural landform, which meant no alien monument. \"But that picture wasn't very clear at all, which could mean alien markings were hidden by haze\" Well no, yes that rumor started, but to prove them wrong on April 8, 2001 we decided to take another picture, making sure it was a cloudless summer day. Malin's team captured an amazing photo using the camera's absolute maximum revolution. With this camera you can discern things in a digital image, 3 times bigger than the pixel size which means if there were any signs of life, you could easily see what they were. What the picture showed was the butte or mesa, which are landforms common around the American West.\"\"\""
]
},
{
"cell_type": "markdown",
"id": "58c6201e-6dd3-4b80-b2de-4d91ca409b50",
"metadata": {},
"source": [
"## Procesar json de salidas de modelos"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "442f07fd-dbaf-4f52-8d5c-a8cbd09efdf2",
"metadata": {},
"outputs": [],
"source": [
"import pandas as pd\n",
"def obtener_dataframe(data):\n",
" \n",
" \n",
" data_flattened = [flatten_json(class_info) for class_info in data]\n",
" df = pd.DataFrame(data_flattened)\n",
" \n",
" return df\n",
" ###\n",
" ### funcion \"flatten_json\" tomada de https://levelup.gitconnected.com/a-deep-dive-into-nested-json-to-data-frame-with-python-69bdabb41938 \n",
" ### Renu Khandelwal Jul 23, 2023\n",
"def flatten_json(y):\n",
" try:\n",
" out = {}\n",
" \n",
" def flatten(x, name=''):\n",
" if type(x) is dict:\n",
" for a in x:\n",
" flatten(x[a], name + a + '_')\n",
" elif type(x) is list:\n",
" i = 0\n",
" for a in x:\n",
" flatten(a, name + str(i) + '_')\n",
" i += 1\n",
" else:\n",
" out[name[:-1]] = x\n",
" \n",
" flatten(y)\n",
" return out\n",
" except json.JSONDecodeError:\n",
" print(\"Error: The JSON document could not be decoded.\")\n",
" except TypeError:\n",
" print(\"Error: Invalid operation or function argument type.\")\n",
" except KeyError:\n",
" print(\"Error: One or more keys do not exist.\")\n",
" except ValueError:\n",
" print(\"Error: Invalid value detected.\")\n",
" except Exception as e:\n",
" # Catch any other exceptions\n",
" print(f\"An unexpected error occurred: {str(e)}\") \n",
"\n"
]
},
{
"cell_type": "markdown",
"id": "6638783a-4b70-4ca4-8b63-fee3c38a090d",
"metadata": {},
"source": [
"## 1 FacebookAI/xlm-roberta-large-finetuned-conll03-english"
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "eafceadf-dfa5-4b5b-b03a-c63abf6e81d0",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Some weights of the model checkpoint at FacebookAI/xlm-roberta-large-finetuned-conll03-english were not used when initializing XLMRobertaForTokenClassification: ['roberta.pooler.dense.bias', 'roberta.pooler.dense.weight']\n",
"- This IS expected if you are initializing XLMRobertaForTokenClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
"- This IS NOT expected if you are initializing XLMRobertaForTokenClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"loss: 0.0\n"
]
}
],
"source": [
"tokenizer = AutoTokenizer.from_pretrained(\"FacebookAI/xlm-roberta-large-finetuned-conll03-english\")\n",
"tokens = tokenizer.tokenize(text)\n",
"ids = tokenizer.convert_tokens_to_ids(tokens)\n",
"input_ids = torch.tensor([ids])\n",
"model_args={\"trust_remote_code\": True}\n",
"model = AutoModelForTokenClassification.from_pretrained(\"FacebookAI/xlm-roberta-large-finetuned-conll03-english\", **model_args)\n",
"with torch.no_grad():\n",
" logits = model(input_ids).logits\n",
"predicted_token_class_ids = logits.argmax(-1)\n",
"\n",
"predicted_tokens_classes = [model.config.id2label[t.item()] for t in predicted_token_class_ids[0]]\n",
"\n",
"labels = predicted_token_class_ids\n",
"loss = model(input_ids, labels=labels).loss\n",
"print('loss:',round(loss.item(), 2))\n",
"\n"
]
},
{
"cell_type": "markdown",
"id": "9efe9516-d6c8-4242-bb2c-aeadd7f41c1d",
"metadata": {},
"source": [
"## Metricas FacebookAI/xlm-roberta-large-finetuned-conll03-english"
]
},
{
"cell_type": "code",
"execution_count": 46,
"id": "b9c3289c-08e9-4546-81a1-324d29bb1989",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Some weights of the model checkpoint at FacebookAI/xlm-roberta-large-finetuned-conll03-english were not used when initializing XLMRobertaForTokenClassification: ['roberta.pooler.dense.bias', 'roberta.pooler.dense.weight']\n",
"- This IS expected if you are initializing XLMRobertaForTokenClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
"- This IS NOT expected if you are initializing XLMRobertaForTokenClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[{'entity': 'I-ORG', 'score': 0.9999913, 'index': 8, 'word': '▁NASA', 'start': 16, 'end': 20}, {'entity': 'I-MISC', 'score': 0.9999641, 'index': 23, 'word': '▁Face', 'start': 88, 'end': 92}, {'entity': 'I-MISC', 'score': 0.99989665, 'index': 24, 'word': '▁On', 'start': 93, 'end': 95}, {'entity': 'I-MISC', 'score': 0.97350365, 'index': 25, 'word': '▁Mars', 'start': 96, 'end': 100}, {'entity': 'I-LOC', 'score': 0.9999362, 'index': 36, 'word': '▁Mars', 'start': 152, 'end': 156}, {'entity': 'I-MISC', 'score': 0.9992086, 'index': 58, 'word': '▁Viking', 'start': 240, 'end': 246}, {'entity': 'I-MISC', 'score': 0.9989502, 'index': 59, 'word': '▁1', 'start': 247, 'end': 248}, {'entity': 'I-MISC', 'score': 0.999977, 'index': 97, 'word': '▁Marti', 'start': 407, 'end': 412}, {'entity': 'I-MISC', 'score': 0.99619055, 'index': 98, 'word': 'an', 'start': 412, 'end': 414}, {'entity': 'I-LOC', 'score': 0.9999354, 'index': 103, 'word': '▁Cy', 'start': 435, 'end': 437}, {'entity': 'I-LOC', 'score': 0.99994576, 'index': 104, 'word': 'do', 'start': 437, 'end': 439}, {'entity': 'I-LOC', 'score': 0.99992585, 'index': 105, 'word': 'nia', 'start': 439, 'end': 442}, {'entity': 'I-MISC', 'score': 0.9999789, 'index': 119, 'word': '▁Egypt', 'start': 496, 'end': 501}, {'entity': 'I-MISC', 'score': 0.9614088, 'index': 120, 'word': 'ion', 'start': 501, 'end': 504}, {'entity': 'I-ORG', 'score': 0.99997246, 'index': 193, 'word': '▁NASA', 'start': 801, 'end': 805}, {'entity': 'I-LOC', 'score': 0.99979633, 'index': 205, 'word': '▁Mars', 'start': 843, 'end': 847}, {'entity': 'I-LOC', 'score': 0.9998061, 'index': 215, 'word': '▁Mars', 'start': 875, 'end': 879}, {'entity': 'I-LOC', 'score': 0.99984956, 'index': 264, 'word': '▁Mars', 'start': 1088, 'end': 1092}, {'entity': 'I-ORG', 'score': 0.99996305, 'index': 285, 'word': '▁NASA', 'start': 1169, 'end': 1173}, {'entity': 'I-LOC', 'score': 0.9998203, 'index': 295, 'word': '▁Mars', 'start': 1220, 'end': 1224}, {'entity': 'I-PER', 'score': 0.9999932, 'index': 319, 'word': '▁Michael', 'start': 1312, 'end': 1319}, {'entity': 'I-PER', 'score': 0.99999106, 'index': 320, 'word': '▁Malin', 'start': 1320, 'end': 1325}, {'entity': 'I-MISC', 'score': 0.94105357, 'index': 323, 'word': '▁Mars', 'start': 1334, 'end': 1338}, {'entity': 'I-MISC', 'score': 0.9839579, 'index': 324, 'word': '▁Or', 'start': 1339, 'end': 1341}, {'entity': 'I-MISC', 'score': 0.9913346, 'index': 325, 'word': 'bit', 'start': 1341, 'end': 1344}, {'entity': 'I-MISC', 'score': 0.9759228, 'index': 326, 'word': 'er', 'start': 1344, 'end': 1346}, {'entity': 'I-MISC', 'score': 0.999749, 'index': 341, 'word': '▁Viking', 'start': 1419, 'end': 1425}, {'entity': 'I-PER', 'score': 0.9999914, 'index': 416, 'word': '▁Malin', 'start': 1745, 'end': 1750}, {'entity': 'I-MISC', 'score': 0.92417294, 'index': 491, 'word': '▁American', 'start': 2094, 'end': 2102}, {'entity': 'I-LOC', 'score': 0.99954396, 'index': 492, 'word': '▁West', 'start': 2103, 'end': 2107}]\n"
]
}
],
"source": [
"import json\n",
"from transformers import pipeline\n",
"tokenizer = AutoTokenizer.from_pretrained(\"FacebookAI/xlm-roberta-large-finetuned-conll03-english\")\n",
"model = AutoModelForTokenClassification.from_pretrained(\"FacebookAI/xlm-roberta-large-finetuned-conll03-english\")\n",
"classifier = pipeline(\"ner\", model=model, tokenizer=tokenizer)\n",
"salida=classifier(text)\n",
"print(salida)\n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": 47,
"id": "ef6050de-0b46-4477-be36-3e6132143cb2",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"entity\n",
"I-LOC 9\n",
"I-MISC 15\n",
"I-ORG 3\n",
"I-PER 3\n",
"dtype: int64\n"
]
},
{
"data": {
"text/plain": [
"entity word \n",
"I-LOC do 1\n",
" nia 1\n",
" ▁Cy 1\n",
" ▁Mars 5\n",
" ▁West 1\n",
"I-MISC an 1\n",
" bit 1\n",
" er 1\n",
" ion 1\n",
" ▁1 1\n",
" ▁American 1\n",
" ▁Egypt 1\n",
" ▁Face 1\n",
" ▁Mars 2\n",
" ▁Marti 1\n",
" ▁On 1\n",
" ▁Or 1\n",
" ▁Viking 2\n",
"I-ORG ▁NASA 3\n",
"I-PER ▁Malin 2\n",
" ▁Michael 1\n",
"dtype: int64"
]
},
"execution_count": 47,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"\n",
"# Create DataFrame from flattened JSON\n",
"with open(\"metricas_Facebook.json\", encoding='utf-8') as f:\n",
" data = json.load(f)\n",
"aux=obtener_dataframe(data)\n",
"aux\n",
"print(aux.groupby(['entity']).size())\n",
"aux.groupby(['entity', 'word']).size()"
]
},
{
"cell_type": "code",
"execution_count": 29,
"id": "9b975064-5919-4fe1-b266-8b9632d10e51",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"{'LOC': {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 7},\n",
" 'MISC': {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 7},\n",
" 'ORG': {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 3},\n",
" 'PER': {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 2},\n",
" 'overall_precision': 1.0,\n",
" 'overall_recall': 1.0,\n",
" 'overall_f1': 1.0,\n",
" 'overall_accuracy': 1.0}"
]
},
"execution_count": 29,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"#pip install seqeval\n",
"from datasets import load_dataset, load_metric\n",
"metric = load_metric(\"seqeval\",**model_args)\n",
"metric.compute(predictions=[predicted_tokens_classes], references=[predicted_tokens_classes])"
]
},
{
"cell_type": "markdown",
"id": "188bdc60-76fe-44a4-9424-b69cc0044b2f",
"metadata": {},
"source": [
"## 2 manu/lilt-infoxlm-base"
]
},
{
"cell_type": "code",
"execution_count": 52,
"id": "8d891893-0445-47d1-aa06-8674ce5d8cce",
"metadata": {},
"outputs": [
{
"ename": "ValueError",
"evalue": "The checkpoint you are trying to load has model type `liltrobertalike` but Transformers does not recognize this architecture. This could be because of an issue with the checkpoint, or because your version of Transformers is out of date.",
"output_type": "error",
"traceback": [
"\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[1;31mKeyError\u001b[0m Traceback (most recent call last)",
"File \u001b[1;32m~\\anaconda3\\Lib\\site-packages\\transformers\\models\\auto\\configuration_auto.py:951\u001b[0m, in \u001b[0;36mAutoConfig.from_pretrained\u001b[1;34m(cls, pretrained_model_name_or_path, **kwargs)\u001b[0m\n\u001b[0;32m 950\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m--> 951\u001b[0m config_class \u001b[38;5;241m=\u001b[39m CONFIG_MAPPING[config_dict[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmodel_type\u001b[39m\u001b[38;5;124m\"\u001b[39m]]\n\u001b[0;32m 952\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m:\n",
"File \u001b[1;32m~\\anaconda3\\Lib\\site-packages\\transformers\\models\\auto\\configuration_auto.py:653\u001b[0m, in \u001b[0;36m_LazyConfigMapping.__getitem__\u001b[1;34m(self, key)\u001b[0m\n\u001b[0;32m 652\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m key \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_mapping:\n\u001b[1;32m--> 653\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m(key)\n\u001b[0;32m 654\u001b[0m value \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_mapping[key]\n",
"\u001b[1;31mKeyError\u001b[0m: 'liltrobertalike'",
"\nDuring handling of the above exception, another exception occurred:\n",
"\u001b[1;31mValueError\u001b[0m Traceback (most recent call last)",
"Cell \u001b[1;32mIn[52], line 2\u001b[0m\n\u001b[0;32m 1\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtransformers\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m AutoModelForMaskedLM\n\u001b[1;32m----> 2\u001b[0m model \u001b[38;5;241m=\u001b[39m AutoModelForMaskedLM\u001b[38;5;241m.\u001b[39mfrom_pretrained(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmanu/lilt-infoxlm-base\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n",
"File \u001b[1;32m~\\anaconda3\\Lib\\site-packages\\transformers\\models\\auto\\auto_factory.py:523\u001b[0m, in \u001b[0;36m_BaseAutoModelClass.from_pretrained\u001b[1;34m(cls, pretrained_model_name_or_path, *model_args, **kwargs)\u001b[0m\n\u001b[0;32m 520\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m kwargs\u001b[38;5;241m.\u001b[39mget(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mquantization_config\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;28;01mNone\u001b[39;00m) \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m 521\u001b[0m _ \u001b[38;5;241m=\u001b[39m kwargs\u001b[38;5;241m.\u001b[39mpop(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mquantization_config\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m--> 523\u001b[0m config, kwargs \u001b[38;5;241m=\u001b[39m AutoConfig\u001b[38;5;241m.\u001b[39mfrom_pretrained(\n\u001b[0;32m 524\u001b[0m pretrained_model_name_or_path,\n\u001b[0;32m 525\u001b[0m return_unused_kwargs\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mTrue\u001b[39;00m,\n\u001b[0;32m 526\u001b[0m trust_remote_code\u001b[38;5;241m=\u001b[39mtrust_remote_code,\n\u001b[0;32m 527\u001b[0m code_revision\u001b[38;5;241m=\u001b[39mcode_revision,\n\u001b[0;32m 528\u001b[0m _commit_hash\u001b[38;5;241m=\u001b[39mcommit_hash,\n\u001b[0;32m 529\u001b[0m \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mhub_kwargs,\n\u001b[0;32m 530\u001b[0m \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs,\n\u001b[0;32m 531\u001b[0m )\n\u001b[0;32m 533\u001b[0m \u001b[38;5;66;03m# if torch_dtype=auto was passed here, ensure to pass it on\u001b[39;00m\n\u001b[0;32m 534\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m kwargs_orig\u001b[38;5;241m.\u001b[39mget(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mtorch_dtype\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;28;01mNone\u001b[39;00m) \u001b[38;5;241m==\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mauto\u001b[39m\u001b[38;5;124m\"\u001b[39m:\n",
"File \u001b[1;32m~\\anaconda3\\Lib\\site-packages\\transformers\\models\\auto\\configuration_auto.py:953\u001b[0m, in \u001b[0;36mAutoConfig.from_pretrained\u001b[1;34m(cls, pretrained_model_name_or_path, **kwargs)\u001b[0m\n\u001b[0;32m 951\u001b[0m config_class \u001b[38;5;241m=\u001b[39m CONFIG_MAPPING[config_dict[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmodel_type\u001b[39m\u001b[38;5;124m\"\u001b[39m]]\n\u001b[0;32m 952\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m:\n\u001b[1;32m--> 953\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\n\u001b[0;32m 954\u001b[0m \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mThe checkpoint you are trying to load has model type `\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mconfig_dict[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mmodel_type\u001b[39m\u001b[38;5;124m'\u001b[39m]\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m` \u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m 955\u001b[0m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mbut Transformers does not recognize this architecture. This could be because of an \u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m 956\u001b[0m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124missue with the checkpoint, or because your version of Transformers is out of date.\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m 957\u001b[0m )\n\u001b[0;32m 958\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m config_class\u001b[38;5;241m.\u001b[39mfrom_dict(config_dict, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39munused_kwargs)\n\u001b[0;32m 959\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m 960\u001b[0m \u001b[38;5;66;03m# Fallback: use pattern matching on the string.\u001b[39;00m\n\u001b[0;32m 961\u001b[0m \u001b[38;5;66;03m# We go from longer names to shorter names to catch roberta before bert (for instance)\u001b[39;00m\n",
"\u001b[1;31mValueError\u001b[0m: The checkpoint you are trying to load has model type `liltrobertalike` but Transformers does not recognize this architecture. This could be because of an issue with the checkpoint, or because your version of Transformers is out of date."
]
}
],
"source": [
"from transformers import AutoModelForMaskedLM\n",
"model = AutoModelForMaskedLM.from_pretrained(\"manu/lilt-infoxlm-base\")"
]
},
{
"cell_type": "markdown",
"id": "2a918756-380f-438f-a544-a4a6767e6eee",
"metadata": {},
"source": [
"## 3 projecte-aina/DEBERTA_CIEL"
]
},
{
"cell_type": "code",
"execution_count": 54,
"id": "00dc1b6a-7538-4a3e-abff-49b5d72321a5",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Asking to truncate to max_length but no maximum length is provided and the model has no predefined maximum length. Default to no truncation.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[{'entity_group': 'GPE', 'score': 0.6340015, 'word': 'Cy', 'start': 434, 'end': 437}, {'entity_group': 'location-other', 'score': 0.20539406, 'word': 'donia', 'start': 437, 'end': 442}, {'entity_group': 'organization-other', 'score': 0.3824667, 'word': 'NASA', 'start': 1168, 'end': 1173}, {'entity_group': 'person-other', 'score': 0.62630403, 'word': 'Michael Malin', 'start': 1311, 'end': 1325}, {'entity_group': 'organization-other', 'score': 0.28832194, 'word': 'Mars', 'start': 1333, 'end': 1338}, {'entity_group': 'organization-privatecompany', 'score': 0.13149069, 'word': 'Orbiter', 'start': 1338, 'end': 1346}, {'entity_group': 'person-other', 'score': 0.27721033, 'word': 'Malin', 'start': 1744, 'end': 1750}, {'entity_group': 'location-other', 'score': 0.44634995, 'word': 'American', 'start': 2093, 'end': 2102}, {'entity_group': 'location-other', 'score': 0.18410492, 'word': 'West', 'start': 2102, 'end': 2107}]\n"
]
}
],
"source": [
"\n",
"from transformers import pipeline\n",
"\n",
"pipe = pipeline(\"ner\", model=\"projecte-aina/DEBERTA_CIEL\")\n",
"ner_entity_results = pipe(text, aggregation_strategy=\"simple\")\n",
"print(ner_entity_results)"
]
},
{
"cell_type": "code",
"execution_count": 56,
"id": "183cc325-3f42-4207-9d8c-18cf79db19ad",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"entity_group\n",
"GPE 1\n",
"location-other 3\n",
"organization-other 2\n",
"organization-privatecompany 1\n",
"person-other 2\n",
"dtype: int64\n"
]
},
{
"data": {
"text/plain": [
"entity_group word \n",
"GPE Cy 1\n",
"location-other American 1\n",
" West 1\n",
" donia 1\n",
"organization-other Mars 1\n",
" NASA 1\n",
"organization-privatecompany Orbiter 1\n",
"person-other Malin 1\n",
" Michael Malin 1\n",
"dtype: int64"
]
},
"execution_count": 56,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Create DataFrame from flattened JSON\n",
"with open(\"3 metricas projecte-ainaDEBERTA_CIEL.json\", encoding='utf-8') as f:\n",
" data = json.load(f)\n",
"aux=obtener_dataframe(data)\n",
"aux\n",
"print(aux.groupby(['entity_group']).size())\n",
"aux.groupby(['entity_group', 'word']).size()"
]
},
{
"cell_type": "markdown",
"id": "6ee85c10-af5e-475b-95f5-9b822eca4c04",
"metadata": {},
"source": [
"## 4 gunghio/distilbert-base-multilingual-cased-finetuned-conll2003-ner"
]
},
{
"cell_type": "code",
"execution_count": 58,
"id": "3d3e7e83-4ed1-41b8-832c-9a544ef596e7",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"[{'entity_group': 'ORG', 'score': 0.98227155, 'word': 'NASA', 'start': 16, 'end': 20}, {'entity_group': 'MISC', 'score': 0.7845409, 'word': 'Face On Mars', 'start': 88, 'end': 100}, {'entity_group': 'LOC', 'score': 0.97792244, 'word': 'Mars', 'start': 152, 'end': 156}, {'entity_group': 'MISC', 'score': 0.9694425, 'word': 'Viking 1', 'start': 240, 'end': 248}, {'entity_group': 'MISC', 'score': 0.95397955, 'word': 'Martian', 'start': 407, 'end': 414}, {'entity_group': 'LOC', 'score': 0.92743087, 'word': 'Cydonia', 'start': 435, 'end': 442}, {'entity_group': 'MISC', 'score': 0.6778414, 'word': 'Egyption Ph', 'start': 496, 'end': 507}, {'entity_group': 'MISC', 'score': 0.6644676, 'word': '##oh', 'start': 510, 'end': 512}, {'entity_group': 'ORG', 'score': 0.9827271, 'word': 'NASA', 'start': 801, 'end': 805}, {'entity_group': 'LOC', 'score': 0.9836016, 'word': 'Mars', 'start': 843, 'end': 847}, {'entity_group': 'LOC', 'score': 0.97961295, 'word': 'Mars', 'start': 875, 'end': 879}, {'entity_group': 'LOC', 'score': 0.976, 'word': 'Mars', 'start': 1088, 'end': 1092}, {'entity_group': 'ORG', 'score': 0.9822379, 'word': 'NASA', 'start': 1169, 'end': 1173}, {'entity_group': 'LOC', 'score': 0.9719374, 'word': 'Mars', 'start': 1220, 'end': 1224}, {'entity_group': 'PER', 'score': 0.99495125, 'word': 'Michael Malin', 'start': 1312, 'end': 1325}, {'entity_group': 'ORG', 'score': 0.74125415, 'word': 'Mars Orbiter', 'start': 1334, 'end': 1346}, {'entity_group': 'MISC', 'score': 0.9728442, 'word': 'Viking', 'start': 1419, 'end': 1425}, {'entity_group': 'PER', 'score': 0.8149092, 'word': 'Malin', 'start': 1745, 'end': 1750}, {'entity_group': 'LOC', 'score': 0.9479703, 'word': 'American West', 'start': 2094, 'end': 2107}]\n"
]
}
],
"source": [
"from transformers import pipeline\n",
"\n",
"pipe = pipeline(\"token-classification\", model=\"gunghio/distilbert-base-multilingual-cased-finetuned-conll2003-ner\")\n",
"#pipe = pipeline(\"ner\", model=\"projecte-aina/DEBERTA_CIEL\")\n",
"ner_entity_results = pipe(text, aggregation_strategy=\"simple\")\n",
"print(ner_entity_results)"
]
},
{
"cell_type": "code",
"execution_count": 60,
"id": "ba56217a-23f9-482b-bcb8-9a5295972506",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"entity_group\n",
"LOC 7\n",
"MISC 6\n",
"ORG 4\n",
"PER 2\n",
"dtype: int64\n"
]
},
{
"data": {
"text/plain": [
"entity_group word \n",
"LOC American West 1\n",
" Cydonia 1\n",
" Mars 5\n",
"MISC ##oh 1\n",
" Egyption Ph 1\n",
" Face On Mars 1\n",
" Martian 1\n",
" Viking 1\n",
" Viking 1 1\n",
"ORG Mars Orbiter 1\n",
" NASA 3\n",
"PER Malin 1\n",
" Michael Malin 1\n",
"dtype: int64"
]
},
"execution_count": 60,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"\n",
"with open(\"4 metricas gunghio distilbert-base-multilingual-cased-finetuned-conll2003-ner.json\", encoding='utf-8') as f:\n",
" data = json.load(f)\n",
"aux=obtener_dataframe(data)\n",
"aux\n",
"print(aux.groupby(['entity_group']).size())\n",
"aux.groupby(['entity_group', 'word']).size()"
]
},
{
"cell_type": "markdown",
"id": "2df8815c-41d5-412e-a650-8892c4fb3bdb",
"metadata": {},
"source": [
"## 5 mrm8488/distilbert-base-multi-cased-finetuned-typo-detection"
]
},
{
"cell_type": "code",
"execution_count": 62,
"id": "a2162ff8-e760-433c-8266-9088bfe7764a",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[{'entity': 'ok',\n",
" 'score': 0.9418772,\n",
" 'index': 2,\n",
" 'word': ',',\n",
" 'start': 2,\n",
" 'end': 3},\n",
" {'entity': 'ok',\n",
" 'score': 0.9061194,\n",
" 'index': 3,\n",
" 'word': 'if',\n",
" 'start': 4,\n",
" 'end': 6},\n",
" {'entity': 'ok',\n",
" 'score': 0.9935272,\n",
" 'index': 4,\n",
" 'word': 'you',\n",
" 'start': 7,\n",
" 'end': 10},\n",
" {'entity': 'ok',\n",
" 'score': 0.9791702,\n",
" 'index': 5,\n",
" 'word': \"'\",\n",
" 'start': 10,\n",
" 'end': 11},\n",
" {'entity': 'ok',\n",
" 'score': 0.99081236,\n",
" 'index': 6,\n",
" 'word': 're',\n",
" 'start': 11,\n",
" 'end': 13},\n",
" {'entity': 'ok',\n",
" 'score': 0.979658,\n",
" 'index': 7,\n",
" 'word': 'a',\n",
" 'start': 14,\n",
" 'end': 15},\n",
" {'entity': 'ok',\n",
" 'score': 0.7008045,\n",
" 'index': 8,\n",
" 'word': 'NASA',\n",
" 'start': 16,\n",
" 'end': 20},\n",
" {'entity': 'ok',\n",
" 'score': 0.9583886,\n",
" 'index': 9,\n",
" 'word': 'scientist',\n",
" 'start': 21,\n",
" 'end': 30},\n",
" {'entity': 'ok',\n",
" 'score': 0.9387937,\n",
" 'index': 10,\n",
" 'word': ',',\n",
" 'start': 30,\n",
" 'end': 31},\n",
" {'entity': 'typo',\n",
" 'score': 0.955443,\n",
" 'index': 11,\n",
" 'word': 'you',\n",
" 'start': 32,\n",
" 'end': 35},\n",
" {'entity': 'typo',\n",
" 'score': 0.5262742,\n",
" 'index': 12,\n",
" 'word': 'should',\n",
" 'start': 36,\n",
" 'end': 42},\n",
" {'entity': 'ok',\n",
" 'score': 0.943428,\n",
" 'index': 13,\n",
" 'word': 'be',\n",
" 'start': 43,\n",
" 'end': 45},\n",
" {'entity': 'ok',\n",
" 'score': 0.9880336,\n",
" 'index': 14,\n",
" 'word': 'able',\n",
" 'start': 46,\n",
" 'end': 50},\n",
" {'entity': 'ok',\n",
" 'score': 0.99831665,\n",
" 'index': 15,\n",
" 'word': 'to',\n",
" 'start': 51,\n",
" 'end': 53},\n",
" {'entity': 'ok',\n",
" 'score': 0.57461846,\n",
" 'index': 16,\n",
" 'word': 'tell',\n",
" 'start': 54,\n",
" 'end': 58},\n",
" {'entity': 'ok',\n",
" 'score': 0.9945175,\n",
" 'index': 17,\n",
" 'word': 'me',\n",
" 'start': 59,\n",
" 'end': 61},\n",
" {'entity': 'ok',\n",
" 'score': 0.810129,\n",
" 'index': 18,\n",
" 'word': 'the',\n",
" 'start': 62,\n",
" 'end': 65},\n",
" {'entity': 'ok',\n",
" 'score': 0.5944688,\n",
" 'index': 19,\n",
" 'word': 'whole',\n",
" 'start': 66,\n",
" 'end': 71},\n",
" {'entity': 'ok',\n",
" 'score': 0.9751041,\n",
" 'index': 20,\n",
" 'word': 'story',\n",
" 'start': 72,\n",
" 'end': 77},\n",
" {'entity': 'ok',\n",
" 'score': 0.8432575,\n",
" 'index': 21,\n",
" 'word': 'about',\n",
" 'start': 78,\n",
" 'end': 83},\n",
" {'entity': 'typo',\n",
" 'score': 0.93727547,\n",
" 'index': 22,\n",
" 'word': 'the',\n",
" 'start': 84,\n",
" 'end': 87},\n",
" {'entity': 'typo',\n",
" 'score': 0.9830071,\n",
" 'index': 23,\n",
" 'word': 'Face',\n",
" 'start': 88,\n",
" 'end': 92},\n",
" {'entity': 'ok',\n",
" 'score': 0.9957579,\n",
" 'index': 24,\n",
" 'word': 'On',\n",
" 'start': 93,\n",
" 'end': 95},\n",
" {'entity': 'ok',\n",
" 'score': 0.98799103,\n",
" 'index': 25,\n",
" 'word': 'Mars',\n",
" 'start': 96,\n",
" 'end': 100},\n",
" {'entity': 'typo',\n",
" 'score': 0.951169,\n",
" 'index': 26,\n",
" 'word': ',',\n",
" 'start': 100,\n",
" 'end': 101},\n",
" {'entity': 'typo',\n",
" 'score': 0.9631144,\n",
" 'index': 27,\n",
" 'word': 'which',\n",
" 'start': 102,\n",
" 'end': 107},\n",
" {'entity': 'typo',\n",
" 'score': 0.99146336,\n",
" 'index': 28,\n",
" 'word': 'obvious',\n",
" 'start': 108,\n",
" 'end': 115},\n",
" {'entity': 'ok',\n",
" 'score': 0.9865329,\n",
" 'index': 29,\n",
" 'word': '##ly',\n",
" 'start': 115,\n",
" 'end': 117},\n",
" {'entity': 'ok',\n",
" 'score': 0.99190086,\n",
" 'index': 30,\n",
" 'word': 'is',\n",
" 'start': 118,\n",
" 'end': 120},\n",
" {'entity': 'ok',\n",
" 'score': 0.87074775,\n",
" 'index': 31,\n",
" 'word': 'evidence',\n",
" 'start': 121,\n",
" 'end': 129},\n",
" {'entity': 'ok',\n",
" 'score': 0.99500775,\n",
" 'index': 32,\n",
" 'word': 'that',\n",
" 'start': 130,\n",
" 'end': 134},\n",
" {'entity': 'typo',\n",
" 'score': 0.94334555,\n",
" 'index': 33,\n",
" 'word': 'there',\n",
" 'start': 135,\n",
" 'end': 140},\n",
" {'entity': 'ok',\n",
" 'score': 0.8655428,\n",
" 'index': 34,\n",
" 'word': 'is',\n",
" 'start': 141,\n",
" 'end': 143},\n",
" {'entity': 'typo',\n",
" 'score': 0.5147766,\n",
" 'index': 35,\n",
" 'word': 'life',\n",
" 'start': 144,\n",
" 'end': 148},\n",
" {'entity': 'ok',\n",
" 'score': 0.9896828,\n",
" 'index': 36,\n",
" 'word': 'on',\n",
" 'start': 149,\n",
" 'end': 151},\n",
" {'entity': 'ok',\n",
" 'score': 0.94172686,\n",
" 'index': 37,\n",
" 'word': 'Mars',\n",
" 'start': 152,\n",
" 'end': 156},\n",
" {'entity': 'typo',\n",
" 'score': 0.92494434,\n",
" 'index': 38,\n",
" 'word': ',',\n",
" 'start': 156,\n",
" 'end': 157},\n",
" {'entity': 'ok',\n",
" 'score': 0.9805345,\n",
" 'index': 39,\n",
" 'word': 'and',\n",
" 'start': 158,\n",
" 'end': 161},\n",
" {'entity': 'ok',\n",
" 'score': 0.8247318,\n",
" 'index': 40,\n",
" 'word': 'that',\n",
" 'start': 162,\n",
" 'end': 166},\n",
" {'entity': 'typo',\n",
" 'score': 0.8335082,\n",
" 'index': 41,\n",
" 'word': 'the',\n",
" 'start': 167,\n",
" 'end': 170},\n",
" {'entity': 'typo',\n",
" 'score': 0.984109,\n",
" 'index': 42,\n",
" 'word': 'face',\n",
" 'start': 171,\n",
" 'end': 175},\n",
" {'entity': 'ok',\n",
" 'score': 0.9346752,\n",
" 'index': 43,\n",
" 'word': 'was',\n",
" 'start': 176,\n",
" 'end': 179},\n",
" {'entity': 'typo',\n",
" 'score': 0.61687183,\n",
" 'index': 44,\n",
" 'word': 'created',\n",
" 'start': 180,\n",
" 'end': 187},\n",
" {'entity': 'ok',\n",
" 'score': 0.9703255,\n",
" 'index': 45,\n",
" 'word': 'by',\n",
" 'start': 188,\n",
" 'end': 190},\n",
" {'entity': 'typo',\n",
" 'score': 0.6155792,\n",
" 'index': 46,\n",
" 'word': 'alien',\n",
" 'start': 191,\n",
" 'end': 196},\n",
" {'entity': 'ok',\n",
" 'score': 0.93030524,\n",
" 'index': 47,\n",
" 'word': '##s',\n",
" 'start': 196,\n",
" 'end': 197},\n",
" {'entity': 'typo',\n",
" 'score': 0.8034546,\n",
" 'index': 48,\n",
" 'word': ',',\n",
" 'start': 197,\n",
" 'end': 198},\n",
" {'entity': 'ok',\n",
" 'score': 0.8684226,\n",
" 'index': 49,\n",
" 'word': 'correct',\n",
" 'start': 199,\n",
" 'end': 206},\n",
" {'entity': 'ok',\n",
" 'score': 0.9899316,\n",
" 'index': 50,\n",
" 'word': '?',\n",
" 'start': 206,\n",
" 'end': 207},\n",
" {'entity': 'typo',\n",
" 'score': 0.68965435,\n",
" 'index': 51,\n",
" 'word': '\"',\n",
" 'start': 207,\n",
" 'end': 208},\n",
" {'entity': 'typo',\n",
" 'score': 0.5541892,\n",
" 'index': 52,\n",
" 'word': 'No',\n",
" 'start': 209,\n",
" 'end': 211},\n",
" {'entity': 'ok',\n",
" 'score': 0.99887687,\n",
" 'index': 53,\n",
" 'word': ',',\n",
" 'start': 211,\n",
" 'end': 212},\n",
" {'entity': 'ok',\n",
" 'score': 0.9993892,\n",
" 'index': 54,\n",
" 'word': 'twenty',\n",
" 'start': 213,\n",
" 'end': 219},\n",
" {'entity': 'ok',\n",
" 'score': 0.9983181,\n",
" 'index': 55,\n",
" 'word': 'five',\n",
" 'start': 220,\n",
" 'end': 224},\n",
" {'entity': 'ok',\n",
" 'score': 0.9524137,\n",
" 'index': 56,\n",
" 'word': 'years',\n",
" 'start': 225,\n",
" 'end': 230},\n",
" {'entity': 'ok',\n",
" 'score': 0.98393893,\n",
" 'index': 57,\n",
" 'word': 'ago',\n",
" 'start': 231,\n",
" 'end': 234},\n",
" {'entity': 'ok',\n",
" 'score': 0.877359,\n",
" 'index': 58,\n",
" 'word': ',',\n",
" 'start': 234,\n",
" 'end': 235},\n",
" {'entity': 'typo',\n",
" 'score': 0.8272593,\n",
" 'index': 59,\n",
" 'word': 'our',\n",
" 'start': 236,\n",
" 'end': 239},\n",
" {'entity': 'ok',\n",
" 'score': 0.88372874,\n",
" 'index': 60,\n",
" 'word': 'Viking',\n",
" 'start': 240,\n",
" 'end': 246},\n",
" {'entity': 'ok',\n",
" 'score': 0.99642414,\n",
" 'index': 61,\n",
" 'word': '1',\n",
" 'start': 247,\n",
" 'end': 248},\n",
" {'entity': 'ok',\n",
" 'score': 0.9792823,\n",
" 'index': 62,\n",
" 'word': 'spacecraft',\n",
" 'start': 249,\n",
" 'end': 259},\n",
" {'entity': 'ok',\n",
" 'score': 0.7188466,\n",
" 'index': 63,\n",
" 'word': 'was',\n",
" 'start': 260,\n",
" 'end': 263},\n",
" {'entity': 'ok',\n",
" 'score': 0.6053355,\n",
" 'index': 64,\n",
" 'word': 'ci',\n",
" 'start': 264,\n",
" 'end': 266},\n",
" {'entity': 'ok',\n",
" 'score': 0.98061955,\n",
" 'index': 65,\n",
" 'word': '##rc',\n",
" 'start': 266,\n",
" 'end': 268},\n",
" {'entity': 'ok',\n",
" 'score': 0.9918943,\n",
" 'index': 66,\n",
" 'word': '##ling',\n",
" 'start': 268,\n",
" 'end': 272},\n",
" {'entity': 'ok',\n",
" 'score': 0.9991246,\n",
" 'index': 67,\n",
" 'word': 'the',\n",
" 'start': 273,\n",
" 'end': 276},\n",
" {'entity': 'ok',\n",
" 'score': 0.99520606,\n",
" 'index': 68,\n",
" 'word': 'planet',\n",
" 'start': 277,\n",
" 'end': 283},\n",
" {'entity': 'typo',\n",
" 'score': 0.9603083,\n",
" 'index': 69,\n",
" 'word': ',',\n",
" 'start': 283,\n",
" 'end': 284},\n",
" {'entity': 'typo',\n",
" 'score': 0.9765087,\n",
" 'index': 70,\n",
" 'word': 'sna',\n",
" 'start': 285,\n",
" 'end': 288},\n",
" {'entity': 'ok',\n",
" 'score': 0.9886219,\n",
" 'index': 71,\n",
" 'word': '##pping',\n",
" 'start': 288,\n",
" 'end': 293},\n",
" {'entity': 'ok',\n",
" 'score': 0.99900526,\n",
" 'index': 72,\n",
" 'word': 'photos',\n",
" 'start': 294,\n",
" 'end': 300},\n",
" {'entity': 'typo',\n",
" 'score': 0.8964089,\n",
" 'index': 73,\n",
" 'word': ',',\n",
" 'start': 300,\n",
" 'end': 301},\n",
" {'entity': 'typo',\n",
" 'score': 0.8211978,\n",
" 'index': 74,\n",
" 'word': 'when',\n",
" 'start': 302,\n",
" 'end': 306},\n",
" {'entity': 'typo',\n",
" 'score': 0.9168602,\n",
" 'index': 75,\n",
" 'word': 'it',\n",
" 'start': 307,\n",
" 'end': 309},\n",
" {'entity': 'ok',\n",
" 'score': 0.9347477,\n",
" 'index': 76,\n",
" 'word': 'spotted',\n",
" 'start': 310,\n",
" 'end': 317},\n",
" {'entity': 'ok',\n",
" 'score': 0.9530431,\n",
" 'index': 77,\n",
" 'word': 'the',\n",
" 'start': 318,\n",
" 'end': 321},\n",
" {'entity': 'typo',\n",
" 'score': 0.52226573,\n",
" 'index': 78,\n",
" 'word': 'sh',\n",
" 'start': 322,\n",
" 'end': 324},\n",
" {'entity': 'ok',\n",
" 'score': 0.98722374,\n",
" 'index': 79,\n",
" 'word': '##adow',\n",
" 'start': 324,\n",
" 'end': 328},\n",
" {'entity': 'ok',\n",
" 'score': 0.9884067,\n",
" 'index': 80,\n",
" 'word': '##y',\n",
" 'start': 328,\n",
" 'end': 329},\n",
" {'entity': 'ok',\n",
" 'score': 0.99751353,\n",
" 'index': 81,\n",
" 'word': 'like',\n",
" 'start': 330,\n",
" 'end': 334},\n",
" {'entity': 'ok',\n",
" 'score': 0.98541266,\n",
" 'index': 82,\n",
" 'word': '##ness',\n",
" 'start': 334,\n",
" 'end': 338},\n",
" {'entity': 'ok',\n",
" 'score': 0.99096996,\n",
" 'index': 83,\n",
" 'word': 'of',\n",
" 'start': 339,\n",
" 'end': 341},\n",
" {'entity': 'ok',\n",
" 'score': 0.9794129,\n",
" 'index': 84,\n",
" 'word': 'a',\n",
" 'start': 342,\n",
" 'end': 343},\n",
" {'entity': 'ok',\n",
" 'score': 0.9906974,\n",
" 'index': 85,\n",
" 'word': 'human',\n",
" 'start': 344,\n",
" 'end': 349},\n",
" {'entity': 'ok',\n",
" 'score': 0.99601525,\n",
" 'index': 86,\n",
" 'word': 'face',\n",
" 'start': 350,\n",
" 'end': 354},\n",
" {'entity': 'typo',\n",
" 'score': 0.80661726,\n",
" 'index': 87,\n",
" 'word': '.',\n",
" 'start': 354,\n",
" 'end': 355},\n",
" {'entity': 'typo',\n",
" 'score': 0.8332319,\n",
" 'index': 88,\n",
" 'word': 'Us',\n",
" 'start': 356,\n",
" 'end': 358},\n",
" {'entity': 'ok',\n",
" 'score': 0.9995962,\n",
" 'index': 89,\n",
" 'word': 'scientists',\n",
" 'start': 359,\n",
" 'end': 369},\n",
" {'entity': 'typo',\n",
" 'score': 0.83959967,\n",
" 'index': 90,\n",
" 'word': 'figure',\n",
" 'start': 370,\n",
" 'end': 376},\n",
" {'entity': 'ok',\n",
" 'score': 0.99624974,\n",
" 'index': 91,\n",
" 'word': '##d',\n",
" 'start': 376,\n",
" 'end': 377},\n",
" {'entity': 'ok',\n",
" 'score': 0.9980217,\n",
" 'index': 92,\n",
" 'word': 'out',\n",
" 'start': 378,\n",
" 'end': 381},\n",
" {'entity': 'ok',\n",
" 'score': 0.5464159,\n",
" 'index': 93,\n",
" 'word': 'that',\n",
" 'start': 382,\n",
" 'end': 386},\n",
" {'entity': 'typo',\n",
" 'score': 0.95522094,\n",
" 'index': 94,\n",
" 'word': 'it',\n",
" 'start': 387,\n",
" 'end': 389},\n",
" {'entity': 'ok',\n",
" 'score': 0.53848577,\n",
" 'index': 95,\n",
" 'word': 'was',\n",
" 'start': 390,\n",
" 'end': 393},\n",
" {'entity': 'typo',\n",
" 'score': 0.9315185,\n",
" 'index': 96,\n",
" 'word': 'just',\n",
" 'start': 394,\n",
" 'end': 398},\n",
" {'entity': 'ok',\n",
" 'score': 0.92040026,\n",
" 'index': 97,\n",
" 'word': 'another',\n",
" 'start': 399,\n",
" 'end': 406},\n",
" {'entity': 'typo',\n",
" 'score': 0.5492458,\n",
" 'index': 98,\n",
" 'word': 'Mart',\n",
" 'start': 407,\n",
" 'end': 411},\n",
" {'entity': 'ok',\n",
" 'score': 0.91541255,\n",
" 'index': 99,\n",
" 'word': '##ian',\n",
" 'start': 411,\n",
" 'end': 414},\n",
" {'entity': 'typo',\n",
" 'score': 0.9870064,\n",
" 'index': 100,\n",
" 'word': 'mesa',\n",
" 'start': 415,\n",
" 'end': 419},\n",
" {'entity': 'typo',\n",
" 'score': 0.9882908,\n",
" 'index': 101,\n",
" 'word': ',',\n",
" 'start': 419,\n",
" 'end': 420},\n",
" {'entity': 'typo',\n",
" 'score': 0.94008124,\n",
" 'index': 102,\n",
" 'word': 'common',\n",
" 'start': 421,\n",
" 'end': 427},\n",
" {'entity': 'ok',\n",
" 'score': 0.99831474,\n",
" 'index': 103,\n",
" 'word': 'around',\n",
" 'start': 428,\n",
" 'end': 434},\n",
" {'entity': 'ok',\n",
" 'score': 0.9985098,\n",
" 'index': 104,\n",
" 'word': 'C',\n",
" 'start': 435,\n",
" 'end': 436},\n",
" {'entity': 'ok',\n",
" 'score': 0.95086056,\n",
" 'index': 105,\n",
" 'word': '##yd',\n",
" 'start': 436,\n",
" 'end': 438},\n",
" {'entity': 'ok',\n",
" 'score': 0.99512345,\n",
" 'index': 106,\n",
" 'word': '##onia',\n",
" 'start': 438,\n",
" 'end': 442},\n",
" {'entity': 'ok',\n",
" 'score': 0.74133,\n",
" 'index': 107,\n",
" 'word': ',',\n",
" 'start': 442,\n",
" 'end': 443},\n",
" {'entity': 'ok',\n",
" 'score': 0.976595,\n",
" 'index': 108,\n",
" 'word': 'only',\n",
" 'start': 444,\n",
" 'end': 448},\n",
" {'entity': 'ok',\n",
" 'score': 0.999059,\n",
" 'index': 109,\n",
" 'word': 'this',\n",
" 'start': 449,\n",
" 'end': 453},\n",
" {'entity': 'ok',\n",
" 'score': 0.9988238,\n",
" 'index': 110,\n",
" 'word': 'one',\n",
" 'start': 454,\n",
" 'end': 457},\n",
" {'entity': 'ok',\n",
" 'score': 0.8208375,\n",
" 'index': 111,\n",
" 'word': 'had',\n",
" 'start': 458,\n",
" 'end': 461},\n",
" {'entity': 'typo',\n",
" 'score': 0.70970184,\n",
" 'index': 112,\n",
" 'word': 'sh',\n",
" 'start': 462,\n",
" 'end': 464},\n",
" {'entity': 'ok',\n",
" 'score': 0.96421015,\n",
" 'index': 113,\n",
" 'word': '##adow',\n",
" 'start': 464,\n",
" 'end': 468},\n",
" {'entity': 'ok',\n",
" 'score': 0.9639605,\n",
" 'index': 114,\n",
" 'word': '##s',\n",
" 'start': 468,\n",
" 'end': 469},\n",
" {'entity': 'ok',\n",
" 'score': 0.90302604,\n",
" 'index': 115,\n",
" 'word': 'that',\n",
" 'start': 470,\n",
" 'end': 474},\n",
" {'entity': 'ok',\n",
" 'score': 0.6680046,\n",
" 'index': 116,\n",
" 'word': 'made',\n",
" 'start': 475,\n",
" 'end': 479},\n",
" {'entity': 'ok',\n",
" 'score': 0.9925667,\n",
" 'index': 117,\n",
" 'word': 'it',\n",
" 'start': 480,\n",
" 'end': 482},\n",
" {'entity': 'typo',\n",
" 'score': 0.5095613,\n",
" 'index': 118,\n",
" 'word': 'look',\n",
" 'start': 483,\n",
" 'end': 487},\n",
" {'entity': 'ok',\n",
" 'score': 0.9739361,\n",
" 'index': 119,\n",
" 'word': 'like',\n",
" 'start': 488,\n",
" 'end': 492},\n",
" {'entity': 'ok',\n",
" 'score': 0.9018896,\n",
" 'index': 120,\n",
" 'word': 'an',\n",
" 'start': 493,\n",
" 'end': 495},\n",
" {'entity': 'typo',\n",
" 'score': 0.5793097,\n",
" 'index': 121,\n",
" 'word': 'Egypt',\n",
" 'start': 496,\n",
" 'end': 501},\n",
" {'entity': 'ok',\n",
" 'score': 0.9924223,\n",
" 'index': 122,\n",
" 'word': '##ion',\n",
" 'start': 501,\n",
" 'end': 504},\n",
" {'entity': 'ok',\n",
" 'score': 0.9970651,\n",
" 'index': 123,\n",
" 'word': 'Ph',\n",
" 'start': 505,\n",
" 'end': 507},\n",
" {'entity': 'ok',\n",
" 'score': 0.99351513,\n",
" 'index': 124,\n",
" 'word': '##ara',\n",
" 'start': 507,\n",
" 'end': 510},\n",
" {'entity': 'ok',\n",
" 'score': 0.98944503,\n",
" 'index': 125,\n",
" 'word': '##oh',\n",
" 'start': 510,\n",
" 'end': 512},\n",
" {'entity': 'typo',\n",
" 'score': 0.44638777,\n",
" 'index': 126,\n",
" 'word': '.',\n",
" 'start': 512,\n",
" 'end': 513},\n",
" {'entity': 'typo',\n",
" 'score': 0.9027144,\n",
" 'index': 127,\n",
" 'word': 'Very',\n",
" 'start': 514,\n",
" 'end': 518},\n",
" {'entity': 'ok',\n",
" 'score': 0.95206773,\n",
" 'index': 128,\n",
" 'word': 'few',\n",
" 'start': 519,\n",
" 'end': 522},\n",
" {'entity': 'ok',\n",
" 'score': 0.88961726,\n",
" 'index': 129,\n",
" 'word': 'days',\n",
" 'start': 523,\n",
" 'end': 527},\n",
" {'entity': 'ok',\n",
" 'score': 0.9765072,\n",
" 'index': 130,\n",
" 'word': 'later',\n",
" 'start': 528,\n",
" 'end': 533},\n",
" {'entity': 'typo',\n",
" 'score': 0.884694,\n",
" 'index': 131,\n",
" 'word': ',',\n",
" 'start': 533,\n",
" 'end': 534},\n",
" {'entity': 'typo',\n",
" 'score': 0.9981382,\n",
" 'index': 132,\n",
" 'word': 'we',\n",
" 'start': 535,\n",
" 'end': 537},\n",
" {'entity': 'typo',\n",
" 'score': 0.8012792,\n",
" 'index': 133,\n",
" 'word': 'revealed',\n",
" 'start': 538,\n",
" 'end': 546},\n",
" {'entity': 'ok',\n",
" 'score': 0.91450936,\n",
" 'index': 134,\n",
" 'word': 'the',\n",
" 'start': 547,\n",
" 'end': 550},\n",
" {'entity': 'typo',\n",
" 'score': 0.9744159,\n",
" 'index': 135,\n",
" 'word': 'image',\n",
" 'start': 551,\n",
" 'end': 556},\n",
" {'entity': 'typo',\n",
" 'score': 0.9916164,\n",
" 'index': 136,\n",
" 'word': 'for',\n",
" 'start': 557,\n",
" 'end': 560},\n",
" {'entity': 'typo',\n",
" 'score': 0.5908708,\n",
" 'index': 137,\n",
" 'word': 'all',\n",
" 'start': 561,\n",
" 'end': 564},\n",
" {'entity': 'ok',\n",
" 'score': 0.99475324,\n",
" 'index': 138,\n",
" 'word': 'to',\n",
" 'start': 565,\n",
" 'end': 567},\n",
" {'entity': 'ok',\n",
" 'score': 0.8057313,\n",
" 'index': 139,\n",
" 'word': 'see',\n",
" 'start': 568,\n",
" 'end': 571},\n",
" {'entity': 'typo',\n",
" 'score': 0.61359376,\n",
" 'index': 140,\n",
" 'word': ',',\n",
" 'start': 571,\n",
" 'end': 572},\n",
" {'entity': 'ok',\n",
" 'score': 0.99861836,\n",
" 'index': 141,\n",
" 'word': 'and',\n",
" 'start': 573,\n",
" 'end': 576},\n",
" {'entity': 'ok',\n",
" 'score': 0.6784175,\n",
" 'index': 142,\n",
" 'word': 'we',\n",
" 'start': 577,\n",
" 'end': 579},\n",
" {'entity': 'ok',\n",
" 'score': 0.9465173,\n",
" 'index': 143,\n",
" 'word': 'made',\n",
" 'start': 580,\n",
" 'end': 584},\n",
" {'entity': 'ok',\n",
" 'score': 0.9947103,\n",
" 'index': 144,\n",
" 'word': 'sure',\n",
" 'start': 585,\n",
" 'end': 589},\n",
" {'entity': 'ok',\n",
" 'score': 0.99907136,\n",
" 'index': 145,\n",
" 'word': 'to',\n",
" 'start': 590,\n",
" 'end': 592},\n",
" {'entity': 'ok',\n",
" 'score': 0.98897797,\n",
" 'index': 146,\n",
" 'word': 'note',\n",
" 'start': 593,\n",
" 'end': 597},\n",
" {'entity': 'ok',\n",
" 'score': 0.9916495,\n",
" 'index': 147,\n",
" 'word': 'that',\n",
" 'start': 598,\n",
" 'end': 602},\n",
" {'entity': 'typo',\n",
" 'score': 0.8916498,\n",
" 'index': 148,\n",
" 'word': 'it',\n",
" 'start': 603,\n",
" 'end': 605},\n",
" {'entity': 'ok',\n",
" 'score': 0.59980196,\n",
" 'index': 149,\n",
" 'word': 'was',\n",
" 'start': 606,\n",
" 'end': 609},\n",
" {'entity': 'typo',\n",
" 'score': 0.6458891,\n",
" 'index': 150,\n",
" 'word': 'a',\n",
" 'start': 610,\n",
" 'end': 611},\n",
" {'entity': 'ok',\n",
" 'score': 0.86706686,\n",
" 'index': 151,\n",
" 'word': 'huge',\n",
" 'start': 612,\n",
" 'end': 616},\n",
" {'entity': 'ok',\n",
" 'score': 0.74633974,\n",
" 'index': 152,\n",
" 'word': 'rock',\n",
" 'start': 617,\n",
" 'end': 621},\n",
" {'entity': 'ok',\n",
" 'score': 0.9891533,\n",
" 'index': 153,\n",
" 'word': 'formation',\n",
" 'start': 622,\n",
" 'end': 631},\n",
" {'entity': 'ok',\n",
" 'score': 0.8717077,\n",
" 'index': 154,\n",
" 'word': 'that',\n",
" 'start': 632,\n",
" 'end': 636},\n",
" {'entity': 'typo',\n",
" 'score': 0.99484986,\n",
" 'index': 155,\n",
" 'word': 'just',\n",
" 'start': 637,\n",
" 'end': 641},\n",
" {'entity': 'typo',\n",
" 'score': 0.5499501,\n",
" 'index': 156,\n",
" 'word': 'res',\n",
" 'start': 642,\n",
" 'end': 645},\n",
" {'entity': 'ok',\n",
" 'score': 0.89776593,\n",
" 'index': 157,\n",
" 'word': '##emble',\n",
" 'start': 645,\n",
" 'end': 650},\n",
" {'entity': 'ok',\n",
" 'score': 0.9908867,\n",
" 'index': 158,\n",
" 'word': '##d',\n",
" 'start': 650,\n",
" 'end': 651},\n",
" {'entity': 'ok',\n",
" 'score': 0.747778,\n",
" 'index': 159,\n",
" 'word': 'a',\n",
" 'start': 652,\n",
" 'end': 653},\n",
" {'entity': 'ok',\n",
" 'score': 0.9308926,\n",
" 'index': 160,\n",
" 'word': 'human',\n",
" 'start': 654,\n",
" 'end': 659},\n",
" {'entity': 'ok',\n",
" 'score': 0.8658663,\n",
" 'index': 161,\n",
" 'word': 'head',\n",
" 'start': 660,\n",
" 'end': 664},\n",
" {'entity': 'ok',\n",
" 'score': 0.9994103,\n",
" 'index': 162,\n",
" 'word': 'and',\n",
" 'start': 665,\n",
" 'end': 668},\n",
" {'entity': 'ok',\n",
" 'score': 0.99929607,\n",
" 'index': 163,\n",
" 'word': 'face',\n",
" 'start': 669,\n",
" 'end': 673},\n",
" {'entity': 'ok',\n",
" 'score': 0.86389965,\n",
" 'index': 164,\n",
" 'word': ',',\n",
" 'start': 673,\n",
" 'end': 674},\n",
" {'entity': 'ok',\n",
" 'score': 0.99792165,\n",
" 'index': 165,\n",
" 'word': 'but',\n",
" 'start': 675,\n",
" 'end': 678},\n",
" {'entity': 'ok',\n",
" 'score': 0.9912469,\n",
" 'index': 166,\n",
" 'word': 'all',\n",
" 'start': 679,\n",
" 'end': 682},\n",
" {'entity': 'ok',\n",
" 'score': 0.9983467,\n",
" 'index': 167,\n",
" 'word': 'of',\n",
" 'start': 683,\n",
" 'end': 685},\n",
" {'entity': 'ok',\n",
" 'score': 0.9972486,\n",
" 'index': 168,\n",
" 'word': 'it',\n",
" 'start': 686,\n",
" 'end': 688},\n",
" {'entity': 'ok',\n",
" 'score': 0.99832314,\n",
" 'index': 169,\n",
" 'word': 'was',\n",
" 'start': 689,\n",
" 'end': 692},\n",
" {'entity': 'ok',\n",
" 'score': 0.9965006,\n",
" 'index': 170,\n",
" 'word': 'formed',\n",
" 'start': 693,\n",
" 'end': 699},\n",
" {'entity': 'ok',\n",
" 'score': 0.9985544,\n",
" 'index': 171,\n",
" 'word': 'by',\n",
" 'start': 700,\n",
" 'end': 702},\n",
" {'entity': 'ok',\n",
" 'score': 0.98051447,\n",
" 'index': 172,\n",
" 'word': 'sh',\n",
" 'start': 703,\n",
" 'end': 705},\n",
" {'entity': 'ok',\n",
" 'score': 0.9919695,\n",
" 'index': 173,\n",
" 'word': '##adow',\n",
" 'start': 705,\n",
" 'end': 709},\n",
" {'entity': 'ok',\n",
" 'score': 0.986326,\n",
" 'index': 174,\n",
" 'word': '##s',\n",
" 'start': 709,\n",
" 'end': 710},\n",
" {'entity': 'typo',\n",
" 'score': 0.79459554,\n",
" 'index': 175,\n",
" 'word': '.',\n",
" 'start': 710,\n",
" 'end': 711},\n",
" {'entity': 'typo',\n",
" 'score': 0.99913234,\n",
" 'index': 176,\n",
" 'word': 'We',\n",
" 'start': 712,\n",
" 'end': 714},\n",
" {'entity': 'ok',\n",
" 'score': 0.9165677,\n",
" 'index': 177,\n",
" 'word': 'only',\n",
" 'start': 715,\n",
" 'end': 719},\n",
" {'entity': 'ok',\n",
" 'score': 0.9994357,\n",
" 'index': 178,\n",
" 'word': 'announced',\n",
" 'start': 720,\n",
" 'end': 729},\n",
" {'entity': 'ok',\n",
" 'score': 0.9980812,\n",
" 'index': 179,\n",
" 'word': 'it',\n",
" 'start': 730,\n",
" 'end': 732},\n",
" {'entity': 'ok',\n",
" 'score': 0.996609,\n",
" 'index': 180,\n",
" 'word': 'because',\n",
" 'start': 733,\n",
" 'end': 740},\n",
" {'entity': 'ok',\n",
" 'score': 0.7698178,\n",
" 'index': 181,\n",
" 'word': 'we',\n",
" 'start': 741,\n",
" 'end': 743},\n",
" {'entity': 'ok',\n",
" 'score': 0.9095254,\n",
" 'index': 182,\n",
" 'word': 'thought',\n",
" 'start': 744,\n",
" 'end': 751},\n",
" {'entity': 'ok',\n",
" 'score': 0.9074025,\n",
" 'index': 183,\n",
" 'word': 'it',\n",
" 'start': 752,\n",
" 'end': 754},\n",
" {'entity': 'ok',\n",
" 'score': 0.94705737,\n",
" 'index': 184,\n",
" 'word': 'would',\n",
" 'start': 755,\n",
" 'end': 760},\n",
" {'entity': 'ok',\n",
" 'score': 0.8691749,\n",
" 'index': 185,\n",
" 'word': 'be',\n",
" 'start': 761,\n",
" 'end': 763},\n",
" {'entity': 'ok',\n",
" 'score': 0.7643121,\n",
" 'index': 186,\n",
" 'word': 'a',\n",
" 'start': 764,\n",
" 'end': 765},\n",
" {'entity': 'ok',\n",
" 'score': 0.92725027,\n",
" 'index': 187,\n",
" 'word': 'good',\n",
" 'start': 766,\n",
" 'end': 770},\n",
" {'entity': 'ok',\n",
" 'score': 0.98407435,\n",
" 'index': 188,\n",
" 'word': 'way',\n",
" 'start': 771,\n",
" 'end': 774},\n",
" {'entity': 'ok',\n",
" 'score': 0.9118401,\n",
" 'index': 189,\n",
" 'word': 'to',\n",
" 'start': 775,\n",
" 'end': 777},\n",
" {'entity': 'typo',\n",
" 'score': 0.80820084,\n",
" 'index': 190,\n",
" 'word': 'engage',\n",
" 'start': 778,\n",
" 'end': 784},\n",
" {'entity': 'ok',\n",
" 'score': 0.9221322,\n",
" 'index': 191,\n",
" 'word': 'the',\n",
" 'start': 785,\n",
" 'end': 788},\n",
" {'entity': 'ok',\n",
" 'score': 0.8702925,\n",
" 'index': 192,\n",
" 'word': 'public',\n",
" 'start': 789,\n",
" 'end': 795},\n",
" {'entity': 'ok',\n",
" 'score': 0.96217585,\n",
" 'index': 193,\n",
" 'word': 'with',\n",
" 'start': 796,\n",
" 'end': 800},\n",
" {'entity': 'ok',\n",
" 'score': 0.5905169,\n",
" 'index': 194,\n",
" 'word': 'NASA',\n",
" 'start': 801,\n",
" 'end': 805},\n",
" {'entity': 'ok',\n",
" 'score': 0.52509785,\n",
" 'index': 195,\n",
" 'word': \"'\",\n",
" 'start': 805,\n",
" 'end': 806},\n",
" {'entity': 'typo',\n",
" 'score': 0.62146723,\n",
" 'index': 196,\n",
" 'word': 's',\n",
" 'start': 806,\n",
" 'end': 807},\n",
" {'entity': 'typo',\n",
" 'score': 0.79807454,\n",
" 'index': 197,\n",
" 'word': 'findings',\n",
" 'start': 808,\n",
" 'end': 816},\n",
" {'entity': 'typo',\n",
" 'score': 0.99402726,\n",
" 'index': 198,\n",
" 'word': ',',\n",
" 'start': 816,\n",
" 'end': 817},\n",
" {'entity': 'ok',\n",
" 'score': 0.49941942,\n",
" 'index': 199,\n",
" 'word': 'and',\n",
" 'start': 818,\n",
" 'end': 821},\n",
" {'entity': 'typo',\n",
" 'score': 0.9993143,\n",
" 'index': 200,\n",
" 'word': 'at',\n",
" 'start': 822,\n",
" 'end': 824},\n",
" {'entity': 'ok',\n",
" 'score': 0.9600053,\n",
" 'index': 201,\n",
" 'word': '##rra',\n",
" 'start': 824,\n",
" 'end': 827},\n",
" {'entity': 'ok',\n",
" 'score': 0.98252696,\n",
" 'index': 202,\n",
" 'word': '##ct',\n",
" 'start': 827,\n",
" 'end': 829},\n",
" {'entity': 'ok',\n",
" 'score': 0.9993383,\n",
" 'index': 203,\n",
" 'word': 'attention',\n",
" 'start': 830,\n",
" 'end': 839},\n",
" {'entity': 'ok',\n",
" 'score': 0.8149123,\n",
" 'index': 204,\n",
" 'word': 'to',\n",
" 'start': 840,\n",
" 'end': 842},\n",
" {'entity': 'ok',\n",
" 'score': 0.928141,\n",
" 'index': 205,\n",
" 'word': 'Mars',\n",
" 'start': 843,\n",
" 'end': 847},\n",
" {'entity': 'typo',\n",
" 'score': 0.98144484,\n",
" 'index': 206,\n",
" 'word': '-',\n",
" 'start': 847,\n",
" 'end': 848},\n",
" {'entity': 'typo',\n",
" 'score': 0.9430255,\n",
" 'index': 207,\n",
" 'word': '-',\n",
" 'start': 848,\n",
" 'end': 849},\n",
" {'entity': 'ok',\n",
" 'score': 0.9964923,\n",
" 'index': 208,\n",
" 'word': 'and',\n",
" 'start': 850,\n",
" 'end': 853},\n",
" {'entity': 'ok',\n",
" 'score': 0.9620715,\n",
" 'index': 209,\n",
" 'word': 'it',\n",
" 'start': 854,\n",
" 'end': 856},\n",
" {'entity': 'ok',\n",
" 'score': 0.9969836,\n",
" 'index': 210,\n",
" 'word': 'did',\n",
" 'start': 857,\n",
" 'end': 860},\n",
" {'entity': 'ok',\n",
" 'score': 0.42969742,\n",
" 'index': 211,\n",
" 'word': '.',\n",
" 'start': 860,\n",
" 'end': 861},\n",
" {'entity': 'ok',\n",
" 'score': 0.7296629,\n",
" 'index': 212,\n",
" 'word': 'The',\n",
" 'start': 863,\n",
" 'end': 866},\n",
" {'entity': 'ok',\n",
" 'score': 0.6601624,\n",
" 'index': 213,\n",
" 'word': 'face',\n",
" 'start': 867,\n",
" 'end': 871},\n",
" {'entity': 'ok',\n",
" 'score': 0.9987452,\n",
" 'index': 214,\n",
" 'word': 'on',\n",
" 'start': 872,\n",
" 'end': 874},\n",
" {'entity': 'ok',\n",
" 'score': 0.995214,\n",
" 'index': 215,\n",
" 'word': 'Mars',\n",
" 'start': 875,\n",
" 'end': 879},\n",
" {'entity': 'typo',\n",
" 'score': 0.76099324,\n",
" 'index': 216,\n",
" 'word': 'soon',\n",
" 'start': 880,\n",
" 'end': 884},\n",
" {'entity': 'ok',\n",
" 'score': 0.93993735,\n",
" 'index': 217,\n",
" 'word': 'became',\n",
" 'start': 885,\n",
" 'end': 891},\n",
" {'entity': 'ok',\n",
" 'score': 0.76330084,\n",
" 'index': 218,\n",
" 'word': 'a',\n",
" 'start': 892,\n",
" 'end': 893},\n",
" {'entity': 'typo',\n",
" 'score': 0.7646893,\n",
" 'index': 219,\n",
" 'word': 'pop',\n",
" 'start': 894,\n",
" 'end': 897},\n",
" {'entity': 'ok',\n",
" 'score': 0.99559265,\n",
" 'index': 220,\n",
" 'word': 'i',\n",
" 'start': 898,\n",
" 'end': 899},\n",
" {'entity': 'ok',\n",
" 'score': 0.99413526,\n",
" 'index': 221,\n",
" 'word': '##con',\n",
" 'start': 899,\n",
" 'end': 902},\n",
" {'entity': 'ok',\n",
" 'score': 0.7378378,\n",
" 'index': 222,\n",
" 'word': ';',\n",
" 'start': 902,\n",
" 'end': 903},\n",
" {'entity': 'typo',\n",
" 'score': 0.90965664,\n",
" 'index': 223,\n",
" 'word': 'shot',\n",
" 'start': 904,\n",
" 'end': 908},\n",
" {'entity': 'ok',\n",
" 'score': 0.998401,\n",
" 'index': 224,\n",
" 'word': 'in',\n",
" 'start': 909,\n",
" 'end': 911},\n",
" {'entity': 'ok',\n",
" 'score': 0.99020493,\n",
" 'index': 225,\n",
" 'word': 'movies',\n",
" 'start': 912,\n",
" 'end': 918},\n",
" {'entity': 'ok',\n",
" 'score': 0.8732322,\n",
" 'index': 226,\n",
" 'word': ',',\n",
" 'start': 918,\n",
" 'end': 919},\n",
" {'entity': 'typo',\n",
" 'score': 0.9965013,\n",
" 'index': 227,\n",
" 'word': 'appeared',\n",
" 'start': 920,\n",
" 'end': 928},\n",
" {'entity': 'ok',\n",
" 'score': 0.99827015,\n",
" 'index': 228,\n",
" 'word': 'in',\n",
" 'start': 929,\n",
" 'end': 931},\n",
" {'entity': 'ok',\n",
" 'score': 0.94189245,\n",
" 'index': 229,\n",
" 'word': 'books',\n",
" 'start': 932,\n",
" 'end': 937},\n",
" {'entity': 'ok',\n",
" 'score': 0.99873203,\n",
" 'index': 230,\n",
" 'word': ',',\n",
" 'start': 937,\n",
" 'end': 938},\n",
" {'entity': 'ok',\n",
" 'score': 0.788286,\n",
" 'index': 231,\n",
" 'word': 'magazines',\n",
" 'start': 939,\n",
" 'end': 948},\n",
" {'entity': 'ok',\n",
" 'score': 0.9985929,\n",
" 'index': 232,\n",
" 'word': ',',\n",
" 'start': 948,\n",
" 'end': 949},\n",
" {'entity': 'typo',\n",
" 'score': 0.83558345,\n",
" 'index': 233,\n",
" 'word': 'radio',\n",
" 'start': 950,\n",
" 'end': 955},\n",
" {'entity': 'ok',\n",
" 'score': 0.9941037,\n",
" 'index': 234,\n",
" 'word': 'talk',\n",
" 'start': 956,\n",
" 'end': 960},\n",
" {'entity': 'ok',\n",
" 'score': 0.9955338,\n",
" 'index': 235,\n",
" 'word': 'shows',\n",
" 'start': 961,\n",
" 'end': 966},\n",
" {'entity': 'typo',\n",
" 'score': 0.8871708,\n",
" 'index': 236,\n",
" 'word': ',',\n",
" 'start': 966,\n",
" 'end': 967},\n",
" {'entity': 'typo',\n",
" 'score': 0.5271952,\n",
" 'index': 237,\n",
" 'word': 'and',\n",
" 'start': 968,\n",
" 'end': 971},\n",
" {'entity': 'typo',\n",
" 'score': 0.99819934,\n",
" 'index': 238,\n",
" 'word': 'hau',\n",
" 'start': 972,\n",
" 'end': 975},\n",
" {'entity': 'ok',\n",
" 'score': 0.8323784,\n",
" 'index': 239,\n",
" 'word': '##nted',\n",
" 'start': 975,\n",
" 'end': 979},\n",
" {'entity': 'typo',\n",
" 'score': 0.97368705,\n",
" 'index': 240,\n",
" 'word': 'gr',\n",
" 'start': 980,\n",
" 'end': 982},\n",
" {'entity': 'ok',\n",
" 'score': 0.54475385,\n",
" 'index': 241,\n",
" 'word': '##oce',\n",
" 'start': 982,\n",
" 'end': 985},\n",
" {'entity': 'ok',\n",
" 'score': 0.9955764,\n",
" 'index': 242,\n",
" 'word': '##ry',\n",
" 'start': 985,\n",
" 'end': 987},\n",
" {'entity': 'ok',\n",
" 'score': 0.9983236,\n",
" 'index': 243,\n",
" 'word': 'store',\n",
" 'start': 988,\n",
" 'end': 993},\n",
" {'entity': 'ok',\n",
" 'score': 0.7288064,\n",
" 'index': 244,\n",
" 'word': 'check',\n",
" 'start': 994,\n",
" 'end': 999},\n",
" {'entity': 'ok',\n",
" 'score': 0.9966523,\n",
" 'index': 245,\n",
" 'word': '##out',\n",
" 'start': 999,\n",
" 'end': 1002},\n",
" {'entity': 'ok',\n",
" 'score': 0.99845624,\n",
" 'index': 246,\n",
" 'word': 'lines',\n",
" 'start': 1003,\n",
" 'end': 1008},\n",
" {'entity': 'typo',\n",
" 'score': 0.6110894,\n",
" 'index': 247,\n",
" 'word': 'for',\n",
" 'start': 1009,\n",
" 'end': 1012},\n",
" {'entity': 'ok',\n",
" 'score': 0.7403333,\n",
" 'index': 248,\n",
" 'word': '25',\n",
" 'start': 1013,\n",
" 'end': 1015},\n",
" {'entity': 'ok',\n",
" 'score': 0.90443635,\n",
" 'index': 249,\n",
" 'word': 'years',\n",
" 'start': 1016,\n",
" 'end': 1021},\n",
" {'entity': 'typo',\n",
" 'score': 0.7376988,\n",
" 'index': 250,\n",
" 'word': '.',\n",
" 'start': 1021,\n",
" 'end': 1022},\n",
" {'entity': 'typo',\n",
" 'score': 0.7852515,\n",
" 'index': 251,\n",
" 'word': 'Some',\n",
" 'start': 1023,\n",
" 'end': 1027},\n",
" {'entity': 'ok',\n",
" 'score': 0.76370704,\n",
" 'index': 252,\n",
" 'word': 'people',\n",
" 'start': 1028,\n",
" 'end': 1034},\n",
" {'entity': 'typo',\n",
" 'score': 0.62583363,\n",
" 'index': 253,\n",
" 'word': 'thought',\n",
" 'start': 1035,\n",
" 'end': 1042},\n",
" {'entity': 'ok',\n",
" 'score': 0.5030125,\n",
" 'index': 254,\n",
" 'word': 'the',\n",
" 'start': 1043,\n",
" 'end': 1046},\n",
" {'entity': 'typo',\n",
" 'score': 0.720193,\n",
" 'index': 255,\n",
" 'word': 'natural',\n",
" 'start': 1047,\n",
" 'end': 1054},\n",
" {'entity': 'ok',\n",
" 'score': 0.9988502,\n",
" 'index': 256,\n",
" 'word': 'land',\n",
" 'start': 1055,\n",
" 'end': 1059},\n",
" {'entity': 'ok',\n",
" 'score': 0.97257465,\n",
" 'index': 257,\n",
" 'word': '##form',\n",
" 'start': 1059,\n",
" 'end': 1063},\n",
" {'entity': 'ok',\n",
" 'score': 0.8562199,\n",
" 'index': 258,\n",
" 'word': 'was',\n",
" 'start': 1064,\n",
" 'end': 1067},\n",
" {'entity': 'typo',\n",
" 'score': 0.8256486,\n",
" 'index': 259,\n",
" 'word': 'evidence',\n",
" 'start': 1068,\n",
" 'end': 1076},\n",
" {'entity': 'ok',\n",
" 'score': 0.99360114,\n",
" 'index': 260,\n",
" 'word': 'of',\n",
" 'start': 1077,\n",
" 'end': 1079},\n",
" {'entity': 'ok',\n",
" 'score': 0.8346857,\n",
" 'index': 261,\n",
" 'word': 'life',\n",
" 'start': 1080,\n",
" 'end': 1084},\n",
" {'entity': 'ok',\n",
" 'score': 0.982291,\n",
" 'index': 262,\n",
" 'word': 'on',\n",
" 'start': 1085,\n",
" 'end': 1087},\n",
" {'entity': 'ok',\n",
" 'score': 0.97399634,\n",
" 'index': 263,\n",
" 'word': 'Mars',\n",
" 'start': 1088,\n",
" 'end': 1092},\n",
" {'entity': 'typo',\n",
" 'score': 0.90118784,\n",
" 'index': 264,\n",
" 'word': ',',\n",
" 'start': 1092,\n",
" 'end': 1093},\n",
" {'entity': 'ok',\n",
" 'score': 0.59052175,\n",
" 'index': 265,\n",
" 'word': 'and',\n",
" 'start': 1094,\n",
" 'end': 1097},\n",
" {'entity': 'typo',\n",
" 'score': 0.9531032,\n",
" 'index': 266,\n",
" 'word': 'that',\n",
" 'start': 1098,\n",
" 'end': 1102},\n",
" {'entity': 'typo',\n",
" 'score': 0.5538794,\n",
" 'index': 267,\n",
" 'word': 'us',\n",
" 'start': 1103,\n",
" 'end': 1105},\n",
" {'entity': 'ok',\n",
" 'score': 0.996549,\n",
" 'index': 268,\n",
" 'word': 'scientists',\n",
" 'start': 1106,\n",
" 'end': 1116},\n",
" {'entity': 'typo',\n",
" 'score': 0.7577664,\n",
" 'index': 269,\n",
" 'word': 'wanted',\n",
" 'start': 1117,\n",
" 'end': 1123},\n",
" {'entity': 'ok',\n",
" 'score': 0.94258755,\n",
" 'index': 270,\n",
" 'word': 'to',\n",
" 'start': 1124,\n",
" 'end': 1126},\n",
" {'entity': 'typo',\n",
" 'score': 0.901848,\n",
" 'index': 271,\n",
" 'word': 'hide',\n",
" 'start': 1127,\n",
" 'end': 1131},\n",
" {'entity': 'ok',\n",
" 'score': 0.87200266,\n",
" 'index': 272,\n",
" 'word': 'it',\n",
" 'start': 1132,\n",
" 'end': 1134},\n",
" {'entity': 'typo',\n",
" 'score': 0.93054014,\n",
" 'index': 273,\n",
" 'word': ',',\n",
" 'start': 1134,\n",
" 'end': 1135},\n",
" {'entity': 'ok',\n",
" 'score': 0.7310256,\n",
" 'index': 274,\n",
" 'word': 'but',\n",
" 'start': 1136,\n",
" 'end': 1139},\n",
" {'entity': 'ok',\n",
" 'score': 0.98565906,\n",
" 'index': 275,\n",
" 'word': 'really',\n",
" 'start': 1140,\n",
" 'end': 1146},\n",
" {'entity': 'ok',\n",
" 'score': 0.94660145,\n",
" 'index': 276,\n",
" 'word': ',',\n",
" 'start': 1146,\n",
" 'end': 1147},\n",
" {'entity': 'typo',\n",
" 'score': 0.9548484,\n",
" 'index': 277,\n",
" 'word': 'the',\n",
" 'start': 1148,\n",
" 'end': 1151},\n",
" {'entity': 'ok',\n",
" 'score': 0.85286355,\n",
" 'index': 278,\n",
" 'word': 'defender',\n",
" 'start': 1152,\n",
" 'end': 1160},\n",
" {'entity': 'ok',\n",
" 'score': 0.919405,\n",
" 'index': 279,\n",
" 'word': '##s',\n",
" 'start': 1160,\n",
" 'end': 1161},\n",
" {'entity': 'ok',\n",
" 'score': 0.97006005,\n",
" 'index': 280,\n",
" 'word': 'of',\n",
" 'start': 1162,\n",
" 'end': 1164},\n",
" {'entity': 'ok',\n",
" 'score': 0.929363,\n",
" 'index': 281,\n",
" 'word': 'the',\n",
" 'start': 1165,\n",
" 'end': 1168},\n",
" {'entity': 'ok',\n",
" 'score': 0.7226242,\n",
" 'index': 282,\n",
" 'word': 'NASA',\n",
" 'start': 1169,\n",
" 'end': 1173},\n",
" {'entity': 'typo',\n",
" 'score': 0.8097878,\n",
" 'index': 283,\n",
" 'word': 'budget',\n",
" 'start': 1174,\n",
" 'end': 1180},\n",
" {'entity': 'ok',\n",
" 'score': 0.95355994,\n",
" 'index': 284,\n",
" 'word': 'wish',\n",
" 'start': 1181,\n",
" 'end': 1185},\n",
" {'entity': 'typo',\n",
" 'score': 0.9236313,\n",
" 'index': 285,\n",
" 'word': 'there',\n",
" 'start': 1186,\n",
" 'end': 1191},\n",
" {'entity': 'ok',\n",
" 'score': 0.7190513,\n",
" 'index': 286,\n",
" 'word': 'was',\n",
" 'start': 1192,\n",
" 'end': 1195},\n",
" {'entity': 'ok',\n",
" 'score': 0.66554,\n",
" 'index': 287,\n",
" 'word': 'ancient',\n",
" 'start': 1196,\n",
" 'end': 1203},\n",
" {'entity': 'ok',\n",
" 'score': 0.60441756,\n",
" 'index': 288,\n",
" 'word': 'civili',\n",
" 'start': 1204,\n",
" 'end': 1210},\n",
" {'entity': 'typo',\n",
" 'score': 0.6450192,\n",
" 'index': 289,\n",
" 'word': '##zation',\n",
" 'start': 1210,\n",
" 'end': 1216},\n",
" {'entity': 'ok',\n",
" 'score': 0.6214468,\n",
" 'index': 290,\n",
" 'word': 'on',\n",
" 'start': 1217,\n",
" 'end': 1219},\n",
" {'entity': 'ok',\n",
" 'score': 0.9483295,\n",
" 'index': 291,\n",
" 'word': 'Mars',\n",
" 'start': 1220,\n",
" 'end': 1224},\n",
" {'entity': 'typo',\n",
" 'score': 0.7586299,\n",
" 'index': 292,\n",
" 'word': '.',\n",
" 'start': 1224,\n",
" 'end': 1225},\n",
" {'entity': 'typo',\n",
" 'score': 0.8658028,\n",
" 'index': 293,\n",
" 'word': 'We',\n",
" 'start': 1226,\n",
" 'end': 1228},\n",
" {'entity': 'typo',\n",
" 'score': 0.65636414,\n",
" 'index': 294,\n",
" 'word': 'decided',\n",
" 'start': 1229,\n",
" 'end': 1236},\n",
" {'entity': 'ok',\n",
" 'score': 0.997926,\n",
" 'index': 295,\n",
" 'word': 'to',\n",
" 'start': 1237,\n",
" 'end': 1239},\n",
" {'entity': 'typo',\n",
" 'score': 0.5479873,\n",
" 'index': 296,\n",
" 'word': 'take',\n",
" 'start': 1240,\n",
" 'end': 1244},\n",
" {'entity': 'ok',\n",
" 'score': 0.99164516,\n",
" 'index': 297,\n",
" 'word': 'another',\n",
" 'start': 1245,\n",
" 'end': 1252},\n",
" {'entity': 'ok',\n",
" 'score': 0.9741786,\n",
" 'index': 298,\n",
" 'word': 'shot',\n",
" 'start': 1253,\n",
" 'end': 1257},\n",
" {'entity': 'typo',\n",
" 'score': 0.98279405,\n",
" 'index': 299,\n",
" 'word': 'just',\n",
" 'start': 1258,\n",
" 'end': 1262},\n",
" {'entity': 'ok',\n",
" 'score': 0.86857635,\n",
" 'index': 300,\n",
" 'word': 'to',\n",
" 'start': 1263,\n",
" 'end': 1265},\n",
" {'entity': 'ok',\n",
" 'score': 0.5942953,\n",
" 'index': 301,\n",
" 'word': 'make',\n",
" 'start': 1266,\n",
" 'end': 1270},\n",
" {'entity': 'ok',\n",
" 'score': 0.9985331,\n",
" 'index': 302,\n",
" 'word': 'sure',\n",
" 'start': 1271,\n",
" 'end': 1275},\n",
" {'entity': 'typo',\n",
" 'score': 0.7710986,\n",
" 'index': 303,\n",
" 'word': 'we',\n",
" 'start': 1276,\n",
" 'end': 1278},\n",
" {'entity': 'ok',\n",
" 'score': 0.6990816,\n",
" 'index': 304,\n",
" 'word': 'were',\n",
" 'start': 1279,\n",
" 'end': 1283},\n",
" {'entity': 'ok',\n",
" 'score': 0.8634561,\n",
" 'index': 305,\n",
" 'word': '##n',\n",
" 'start': 1283,\n",
" 'end': 1284},\n",
" {'entity': 'ok',\n",
" 'score': 0.99818283,\n",
" 'index': 306,\n",
" 'word': \"'\",\n",
" 'start': 1284,\n",
" 'end': 1285},\n",
" {'entity': 'ok',\n",
" 'score': 0.99803716,\n",
" 'index': 307,\n",
" 'word': 't',\n",
" 'start': 1285,\n",
" 'end': 1286},\n",
" {'entity': 'ok',\n",
" 'score': 0.9882159,\n",
" 'index': 308,\n",
" 'word': 'wrong',\n",
" 'start': 1287,\n",
" 'end': 1292},\n",
" {'entity': 'typo',\n",
" 'score': 0.88255525,\n",
" 'index': 309,\n",
" 'word': ',',\n",
" 'start': 1292,\n",
" 'end': 1293},\n",
" {'entity': 'typo',\n",
" 'score': 0.9208057,\n",
" 'index': 310,\n",
" 'word': 'on',\n",
" 'start': 1294,\n",
" 'end': 1296},\n",
" {'entity': 'typo',\n",
" 'score': 0.91730064,\n",
" 'index': 311,\n",
" 'word': 'April',\n",
" 'start': 1297,\n",
" 'end': 1302},\n",
" {'entity': 'ok',\n",
" 'score': 0.5459038,\n",
" 'index': 312,\n",
" 'word': '5',\n",
" 'start': 1303,\n",
" 'end': 1304},\n",
" {'entity': 'ok',\n",
" 'score': 0.5843325,\n",
" 'index': 313,\n",
" 'word': ',',\n",
" 'start': 1304,\n",
" 'end': 1305},\n",
" {'entity': 'typo',\n",
" 'score': 0.8402537,\n",
" 'index': 314,\n",
" 'word': '1998',\n",
" 'start': 1306,\n",
" 'end': 1310},\n",
" {'entity': 'typo',\n",
" 'score': 0.9408526,\n",
" 'index': 315,\n",
" 'word': '.',\n",
" 'start': 1310,\n",
" 'end': 1311},\n",
" {'entity': 'typo',\n",
" 'score': 0.9976394,\n",
" 'index': 316,\n",
" 'word': 'Michael',\n",
" 'start': 1312,\n",
" 'end': 1319},\n",
" {'entity': 'ok',\n",
" 'score': 0.96120375,\n",
" 'index': 317,\n",
" 'word': 'Mali',\n",
" 'start': 1320,\n",
" 'end': 1324},\n",
" {'entity': 'ok',\n",
" 'score': 0.923244,\n",
" 'index': 318,\n",
" 'word': '##n',\n",
" 'start': 1324,\n",
" 'end': 1325},\n",
" {'entity': 'ok',\n",
" 'score': 0.97935665,\n",
" 'index': 319,\n",
" 'word': 'and',\n",
" 'start': 1326,\n",
" 'end': 1329},\n",
" {'entity': 'typo',\n",
" 'score': 0.7523317,\n",
" 'index': 320,\n",
" 'word': 'his',\n",
" 'start': 1330,\n",
" 'end': 1333},\n",
" {'entity': 'typo',\n",
" 'score': 0.91937953,\n",
" 'index': 321,\n",
" 'word': 'Mars',\n",
" 'start': 1334,\n",
" 'end': 1338},\n",
" {'entity': 'ok',\n",
" 'score': 0.8556037,\n",
" 'index': 322,\n",
" 'word': 'Or',\n",
" 'start': 1339,\n",
" 'end': 1341},\n",
" {'entity': 'ok',\n",
" 'score': 0.84279835,\n",
" 'index': 323,\n",
" 'word': '##biter',\n",
" 'start': 1341,\n",
" 'end': 1346},\n",
" {'entity': 'typo',\n",
" 'score': 0.9986338,\n",
" 'index': 324,\n",
" 'word': 'camera',\n",
" 'start': 1347,\n",
" 'end': 1353},\n",
" {'entity': 'ok',\n",
" 'score': 0.90062076,\n",
" 'index': 325,\n",
" 'word': 'team',\n",
" 'start': 1354,\n",
" 'end': 1358},\n",
" {'entity': 'typo',\n",
" 'score': 0.91712064,\n",
" 'index': 326,\n",
" 'word': 'took',\n",
" 'start': 1359,\n",
" 'end': 1363},\n",
" {'entity': 'ok',\n",
" 'score': 0.785382,\n",
" 'index': 327,\n",
" 'word': 'a',\n",
" 'start': 1364,\n",
" 'end': 1365},\n",
" {'entity': 'typo',\n",
" 'score': 0.9040174,\n",
" 'index': 328,\n",
" 'word': 'picture',\n",
" 'start': 1366,\n",
" 'end': 1373},\n",
" {'entity': 'ok',\n",
" 'score': 0.9086188,\n",
" 'index': 329,\n",
" 'word': 'that',\n",
" 'start': 1374,\n",
" 'end': 1378},\n",
" {'entity': 'typo',\n",
" 'score': 0.65382206,\n",
" 'index': 330,\n",
" 'word': 'was',\n",
" 'start': 1379,\n",
" 'end': 1382},\n",
" {'entity': 'ok',\n",
" 'score': 0.95172024,\n",
" 'index': 331,\n",
" 'word': 'ten',\n",
" 'start': 1383,\n",
" 'end': 1386},\n",
" {'entity': 'ok',\n",
" 'score': 0.9873766,\n",
" 'index': 332,\n",
" 'word': 'times',\n",
" 'start': 1387,\n",
" 'end': 1392},\n",
" {'entity': 'ok',\n",
" 'score': 0.67130685,\n",
" 'index': 333,\n",
" 'word': 'sharp',\n",
" 'start': 1393,\n",
" 'end': 1398},\n",
" {'entity': 'ok',\n",
" 'score': 0.9719016,\n",
" 'index': 334,\n",
" 'word': '##er',\n",
" 'start': 1398,\n",
" 'end': 1400},\n",
" {'entity': 'ok',\n",
" 'score': 0.99000204,\n",
" 'index': 335,\n",
" 'word': 'than',\n",
" 'start': 1401,\n",
" 'end': 1405},\n",
" {'entity': 'ok',\n",
" 'score': 0.93401396,\n",
" 'index': 336,\n",
" 'word': 'the',\n",
" 'start': 1406,\n",
" 'end': 1409},\n",
" {'entity': 'ok',\n",
" 'score': 0.77596074,\n",
" 'index': 337,\n",
" 'word': 'original',\n",
" 'start': 1410,\n",
" 'end': 1418},\n",
" {'entity': 'typo',\n",
" 'score': 0.7519366,\n",
" 'index': 338,\n",
" 'word': 'Viking',\n",
" 'start': 1419,\n",
" 'end': 1425},\n",
" {'entity': 'ok',\n",
" 'score': 0.66892004,\n",
" 'index': 339,\n",
" 'word': 'photos',\n",
" 'start': 1426,\n",
" 'end': 1432},\n",
" {'entity': 'typo',\n",
" 'score': 0.98212206,\n",
" 'index': 340,\n",
" 'word': ',',\n",
" 'start': 1432,\n",
" 'end': 1433},\n",
" {'entity': 'typo',\n",
" 'score': 0.99039525,\n",
" 'index': 341,\n",
" 'word': 'reveal',\n",
" 'start': 1434,\n",
" 'end': 1440},\n",
" {'entity': 'ok',\n",
" 'score': 0.53409153,\n",
" 'index': 342,\n",
" 'word': '##ing',\n",
" 'start': 1440,\n",
" 'end': 1443},\n",
" {'entity': 'typo',\n",
" 'score': 0.86396545,\n",
" 'index': 343,\n",
" 'word': 'a',\n",
" 'start': 1444,\n",
" 'end': 1445},\n",
" {'entity': 'typo',\n",
" 'score': 0.6633433,\n",
" 'index': 344,\n",
" 'word': 'natural',\n",
" 'start': 1446,\n",
" 'end': 1453},\n",
" {'entity': 'ok',\n",
" 'score': 0.9973527,\n",
" 'index': 345,\n",
" 'word': 'land',\n",
" 'start': 1454,\n",
" 'end': 1458},\n",
" {'entity': 'ok',\n",
" 'score': 0.8840458,\n",
" 'index': 346,\n",
" 'word': '##form',\n",
" 'start': 1458,\n",
" 'end': 1462},\n",
" {'entity': 'typo',\n",
" 'score': 0.9950237,\n",
" 'index': 347,\n",
" 'word': ',',\n",
" 'start': 1462,\n",
" 'end': 1463},\n",
" {'entity': 'typo',\n",
" 'score': 0.9553996,\n",
" 'index': 348,\n",
" 'word': 'which',\n",
" 'start': 1464,\n",
" 'end': 1469},\n",
" {'entity': 'typo',\n",
" 'score': 0.9745846,\n",
" 'index': 349,\n",
" 'word': 'meant',\n",
" 'start': 1470,\n",
" 'end': 1475},\n",
" {'entity': 'typo',\n",
" 'score': 0.7203086,\n",
" 'index': 350,\n",
" 'word': 'no',\n",
" 'start': 1476,\n",
" 'end': 1478},\n",
" {'entity': 'ok',\n",
" 'score': 0.91405845,\n",
" 'index': 351,\n",
" 'word': 'alien',\n",
" 'start': 1479,\n",
" 'end': 1484},\n",
" {'entity': 'ok',\n",
" 'score': 0.99879634,\n",
" 'index': 352,\n",
" 'word': 'monument',\n",
" 'start': 1485,\n",
" 'end': 1493},\n",
" {'entity': 'typo',\n",
" 'score': 0.56567025,\n",
" 'index': 353,\n",
" 'word': '.',\n",
" 'start': 1493,\n",
" 'end': 1494},\n",
" {'entity': 'typo',\n",
" 'score': 0.971668,\n",
" 'index': 354,\n",
" 'word': '\"',\n",
" 'start': 1495,\n",
" 'end': 1496},\n",
" {'entity': 'ok',\n",
" 'score': 0.9891775,\n",
" 'index': 355,\n",
" 'word': 'But',\n",
" 'start': 1496,\n",
" 'end': 1499},\n",
" {'entity': 'ok',\n",
" 'score': 0.9991333,\n",
" 'index': 356,\n",
" 'word': 'that',\n",
" 'start': 1500,\n",
" 'end': 1504},\n",
" {'entity': 'ok',\n",
" 'score': 0.9964754,\n",
" 'index': 357,\n",
" 'word': 'picture',\n",
" 'start': 1505,\n",
" 'end': 1512},\n",
" {'entity': 'ok',\n",
" 'score': 0.89596707,\n",
" 'index': 358,\n",
" 'word': 'wasn',\n",
" 'start': 1513,\n",
" 'end': 1517},\n",
" {'entity': 'ok',\n",
" 'score': 0.986717,\n",
" 'index': 359,\n",
" 'word': \"'\",\n",
" 'start': 1517,\n",
" 'end': 1518},\n",
" {'entity': 'ok',\n",
" 'score': 0.98859054,\n",
" 'index': 360,\n",
" 'word': 't',\n",
" 'start': 1518,\n",
" 'end': 1519},\n",
" {'entity': 'ok',\n",
" 'score': 0.9508471,\n",
" 'index': 361,\n",
" 'word': 'very',\n",
" 'start': 1520,\n",
" 'end': 1524},\n",
" {'entity': 'ok',\n",
" 'score': 0.9967194,\n",
" 'index': 362,\n",
" 'word': 'clear',\n",
" 'start': 1525,\n",
" 'end': 1530},\n",
" {'entity': 'ok',\n",
" 'score': 0.69285977,\n",
" 'index': 363,\n",
" 'word': 'at',\n",
" 'start': 1531,\n",
" 'end': 1533},\n",
" {'entity': 'ok',\n",
" 'score': 0.9915514,\n",
" 'index': 364,\n",
" 'word': 'all',\n",
" 'start': 1534,\n",
" 'end': 1537},\n",
" {'entity': 'ok',\n",
" 'score': 0.5899336,\n",
" 'index': 365,\n",
" 'word': ',',\n",
" 'start': 1537,\n",
" 'end': 1538},\n",
" {'entity': 'ok',\n",
" 'score': 0.5996167,\n",
" 'index': 366,\n",
" 'word': 'which',\n",
" 'start': 1539,\n",
" 'end': 1544},\n",
" {'entity': 'typo',\n",
" 'score': 0.8670836,\n",
" 'index': 367,\n",
" 'word': 'could',\n",
" 'start': 1545,\n",
" 'end': 1550},\n",
" {'entity': 'ok',\n",
" 'score': 0.9761153,\n",
" 'index': 368,\n",
" 'word': 'mean',\n",
" 'start': 1551,\n",
" 'end': 1555},\n",
" {'entity': 'typo',\n",
" 'score': 0.7312487,\n",
" 'index': 369,\n",
" 'word': 'alien',\n",
" 'start': 1556,\n",
" 'end': 1561},\n",
" {'entity': 'ok',\n",
" 'score': 0.8062085,\n",
" 'index': 370,\n",
" 'word': 'marking',\n",
" 'start': 1562,\n",
" 'end': 1569},\n",
" {'entity': 'typo',\n",
" 'score': 0.80552965,\n",
" 'index': 371,\n",
" 'word': '##s',\n",
" 'start': 1569,\n",
" 'end': 1570},\n",
" {'entity': 'typo',\n",
" 'score': 0.97620255,\n",
" 'index': 372,\n",
" 'word': 'were',\n",
" 'start': 1571,\n",
" 'end': 1575},\n",
" {'entity': 'typo',\n",
" 'score': 0.91833067,\n",
" 'index': 373,\n",
" 'word': 'hidden',\n",
" 'start': 1576,\n",
" 'end': 1582},\n",
" {'entity': 'typo',\n",
" 'score': 0.8934933,\n",
" 'index': 374,\n",
" 'word': 'by',\n",
" 'start': 1583,\n",
" 'end': 1585},\n",
" {'entity': 'typo',\n",
" 'score': 0.9554798,\n",
" 'index': 375,\n",
" 'word': 'ha',\n",
" 'start': 1586,\n",
" 'end': 1588},\n",
" {'entity': 'ok',\n",
" 'score': 0.6562142,\n",
" 'index': 376,\n",
" 'word': '##ze',\n",
" 'start': 1588,\n",
" 'end': 1590},\n",
" {'entity': 'ok',\n",
" 'score': 0.64348394,\n",
" 'index': 377,\n",
" 'word': '\"',\n",
" 'start': 1590,\n",
" 'end': 1591},\n",
" {'entity': 'ok',\n",
" 'score': 0.949757,\n",
" 'index': 378,\n",
" 'word': 'Well',\n",
" 'start': 1592,\n",
" 'end': 1596},\n",
" {'entity': 'ok',\n",
" 'score': 0.99965537,\n",
" 'index': 379,\n",
" 'word': 'no',\n",
" 'start': 1597,\n",
" 'end': 1599},\n",
" {'entity': 'ok',\n",
" 'score': 0.9982284,\n",
" 'index': 380,\n",
" 'word': ',',\n",
" 'start': 1599,\n",
" 'end': 1600},\n",
" {'entity': 'typo',\n",
" 'score': 0.7986995,\n",
" 'index': 381,\n",
" 'word': 'ye',\n",
" 'start': 1601,\n",
" 'end': 1603},\n",
" {'entity': 'ok',\n",
" 'score': 0.9877543,\n",
" 'index': 382,\n",
" 'word': '##s',\n",
" 'start': 1603,\n",
" 'end': 1604},\n",
" {'entity': 'ok',\n",
" 'score': 0.9989477,\n",
" 'index': 383,\n",
" 'word': 'that',\n",
" 'start': 1605,\n",
" 'end': 1609},\n",
" {'entity': 'typo',\n",
" 'score': 0.8178804,\n",
" 'index': 384,\n",
" 'word': 'rum',\n",
" 'start': 1610,\n",
" 'end': 1613},\n",
" {'entity': 'ok',\n",
" 'score': 0.9868613,\n",
" 'index': 385,\n",
" 'word': '##or',\n",
" 'start': 1613,\n",
" 'end': 1615},\n",
" {'entity': 'ok',\n",
" 'score': 0.90991825,\n",
" 'index': 386,\n",
" 'word': 'started',\n",
" 'start': 1616,\n",
" 'end': 1623},\n",
" {'entity': 'typo',\n",
" 'score': 0.5242705,\n",
" 'index': 387,\n",
" 'word': ',',\n",
" 'start': 1623,\n",
" 'end': 1624},\n",
" {'entity': 'ok',\n",
" 'score': 0.9543154,\n",
" 'index': 388,\n",
" 'word': 'but',\n",
" 'start': 1625,\n",
" 'end': 1628},\n",
" {'entity': 'ok',\n",
" 'score': 0.86853844,\n",
" 'index': 389,\n",
" 'word': 'to',\n",
" 'start': 1629,\n",
" 'end': 1631},\n",
" {'entity': 'ok',\n",
" 'score': 0.7432325,\n",
" 'index': 390,\n",
" 'word': 'prove',\n",
" 'start': 1632,\n",
" 'end': 1637},\n",
" {'entity': 'ok',\n",
" 'score': 0.9467291,\n",
" 'index': 391,\n",
" 'word': 'them',\n",
" 'start': 1638,\n",
" 'end': 1642},\n",
" {'entity': 'typo',\n",
" 'score': 0.7128569,\n",
" 'index': 392,\n",
" 'word': 'wrong',\n",
" 'start': 1643,\n",
" 'end': 1648},\n",
" {'entity': 'ok',\n",
" 'score': 0.62498444,\n",
" 'index': 393,\n",
" 'word': 'on',\n",
" 'start': 1649,\n",
" 'end': 1651},\n",
" {'entity': 'typo',\n",
" 'score': 0.6839772,\n",
" 'index': 394,\n",
" 'word': 'April',\n",
" 'start': 1652,\n",
" 'end': 1657},\n",
" {'entity': 'ok',\n",
" 'score': 0.68691665,\n",
" 'index': 395,\n",
" 'word': '8',\n",
" 'start': 1658,\n",
" 'end': 1659},\n",
" {'entity': 'ok',\n",
" 'score': 0.84025437,\n",
" 'index': 396,\n",
" 'word': ',',\n",
" 'start': 1659,\n",
" 'end': 1660},\n",
" {'entity': 'typo',\n",
" 'score': 0.9157925,\n",
" 'index': 397,\n",
" 'word': '2001',\n",
" 'start': 1661,\n",
" 'end': 1665},\n",
" {'entity': 'typo',\n",
" 'score': 0.7844509,\n",
" 'index': 398,\n",
" 'word': 'we',\n",
" 'start': 1666,\n",
" 'end': 1668},\n",
" {'entity': 'typo',\n",
" 'score': 0.5589368,\n",
" 'index': 399,\n",
" 'word': 'decided',\n",
" 'start': 1669,\n",
" 'end': 1676},\n",
" {'entity': 'ok',\n",
" 'score': 0.99429286,\n",
" 'index': 400,\n",
" 'word': 'to',\n",
" 'start': 1677,\n",
" 'end': 1679},\n",
" {'entity': 'typo',\n",
" 'score': 0.84421164,\n",
" 'index': 401,\n",
" 'word': 'take',\n",
" 'start': 1680,\n",
" 'end': 1684},\n",
" {'entity': 'ok',\n",
" 'score': 0.978843,\n",
" 'index': 402,\n",
" 'word': 'another',\n",
" 'start': 1685,\n",
" 'end': 1692},\n",
" {'entity': 'ok',\n",
" 'score': 0.59987646,\n",
" 'index': 403,\n",
" 'word': 'picture',\n",
" 'start': 1693,\n",
" 'end': 1700},\n",
" {'entity': 'typo',\n",
" 'score': 0.7050702,\n",
" 'index': 404,\n",
" 'word': ',',\n",
" 'start': 1700,\n",
" 'end': 1701},\n",
" {'entity': 'ok',\n",
" 'score': 0.95890915,\n",
" 'index': 405,\n",
" 'word': 'making',\n",
" 'start': 1702,\n",
" 'end': 1708},\n",
" {'entity': 'ok',\n",
" 'score': 0.99975187,\n",
" 'index': 406,\n",
" 'word': 'sure',\n",
" 'start': 1709,\n",
" 'end': 1713},\n",
" {'entity': 'ok',\n",
" 'score': 0.98025346,\n",
" 'index': 407,\n",
" 'word': 'it',\n",
" 'start': 1714,\n",
" 'end': 1716},\n",
" {'entity': 'ok',\n",
" 'score': 0.83194023,\n",
" 'index': 408,\n",
" 'word': 'was',\n",
" 'start': 1717,\n",
" 'end': 1720},\n",
" {'entity': 'typo',\n",
" 'score': 0.9467474,\n",
" 'index': 409,\n",
" 'word': 'a',\n",
" 'start': 1721,\n",
" 'end': 1722},\n",
" {'entity': 'typo',\n",
" 'score': 0.91021883,\n",
" 'index': 410,\n",
" 'word': 'cloud',\n",
" 'start': 1723,\n",
" 'end': 1728},\n",
" {'entity': 'ok',\n",
" 'score': 0.9859971,\n",
" 'index': 411,\n",
" 'word': '##less',\n",
" 'start': 1728,\n",
" 'end': 1732},\n",
" {'entity': 'typo',\n",
" 'score': 0.98307884,\n",
" 'index': 412,\n",
" 'word': 'summer',\n",
" 'start': 1733,\n",
" 'end': 1739},\n",
" {'entity': 'ok',\n",
" 'score': 0.994193,\n",
" 'index': 413,\n",
" 'word': 'day',\n",
" 'start': 1740,\n",
" 'end': 1743},\n",
" {'entity': 'typo',\n",
" 'score': 0.4313977,\n",
" 'index': 414,\n",
" 'word': '.',\n",
" 'start': 1743,\n",
" 'end': 1744},\n",
" {'entity': 'typo',\n",
" 'score': 0.96405447,\n",
" 'index': 415,\n",
" 'word': 'Mali',\n",
" 'start': 1745,\n",
" 'end': 1749},\n",
" {'entity': 'ok',\n",
" 'score': 0.6155722,\n",
" 'index': 416,\n",
" 'word': '##n',\n",
" 'start': 1749,\n",
" 'end': 1750},\n",
" {'entity': 'ok',\n",
" 'score': 0.9290005,\n",
" 'index': 417,\n",
" 'word': \"'\",\n",
" 'start': 1750,\n",
" 'end': 1751},\n",
" {'entity': 'ok',\n",
" 'score': 0.90629065,\n",
" 'index': 418,\n",
" 'word': 's',\n",
" 'start': 1751,\n",
" 'end': 1752},\n",
" {'entity': 'ok',\n",
" 'score': 0.868706,\n",
" 'index': 419,\n",
" 'word': 'team',\n",
" 'start': 1753,\n",
" 'end': 1757},\n",
" {'entity': 'ok',\n",
" 'score': 0.8017526,\n",
" 'index': 420,\n",
" 'word': 'captured',\n",
" 'start': 1758,\n",
" 'end': 1766},\n",
" {'entity': 'ok',\n",
" 'score': 0.7134167,\n",
" 'index': 421,\n",
" 'word': 'an',\n",
" 'start': 1767,\n",
" 'end': 1769},\n",
" {'entity': 'typo',\n",
" 'score': 0.5621726,\n",
" 'index': 422,\n",
" 'word': 'ama',\n",
" 'start': 1770,\n",
" 'end': 1773},\n",
" {'entity': 'ok',\n",
" 'score': 0.7826029,\n",
" 'index': 423,\n",
" 'word': '##zing',\n",
" 'start': 1773,\n",
" 'end': 1777},\n",
" {'entity': 'ok',\n",
" 'score': 0.99448264,\n",
" 'index': 424,\n",
" 'word': 'photo',\n",
" 'start': 1778,\n",
" 'end': 1783},\n",
" {'entity': 'ok',\n",
" 'score': 0.73524755,\n",
" 'index': 425,\n",
" 'word': 'using',\n",
" 'start': 1784,\n",
" 'end': 1789},\n",
" {'entity': 'ok',\n",
" 'score': 0.6591615,\n",
" 'index': 426,\n",
" 'word': 'the',\n",
" 'start': 1790,\n",
" 'end': 1793},\n",
" {'entity': 'typo',\n",
" 'score': 0.86533505,\n",
" 'index': 427,\n",
" 'word': 'camera',\n",
" 'start': 1794,\n",
" 'end': 1800},\n",
" {'entity': 'ok',\n",
" 'score': 0.90541,\n",
" 'index': 428,\n",
" 'word': \"'\",\n",
" 'start': 1800,\n",
" 'end': 1801},\n",
" {'entity': 'ok',\n",
" 'score': 0.9281528,\n",
" 'index': 429,\n",
" 'word': 's',\n",
" 'start': 1801,\n",
" 'end': 1802},\n",
" {'entity': 'typo',\n",
" 'score': 0.8930355,\n",
" 'index': 430,\n",
" 'word': 'absolute',\n",
" 'start': 1803,\n",
" 'end': 1811},\n",
" {'entity': 'ok',\n",
" 'score': 0.9972451,\n",
" 'index': 431,\n",
" 'word': 'maximum',\n",
" 'start': 1812,\n",
" 'end': 1819},\n",
" {'entity': 'ok',\n",
" 'score': 0.999348,\n",
" 'index': 432,\n",
" 'word': 'revolution',\n",
" 'start': 1820,\n",
" 'end': 1830},\n",
" {'entity': 'typo',\n",
" 'score': 0.6555487,\n",
" 'index': 433,\n",
" 'word': '.',\n",
" 'start': 1830,\n",
" 'end': 1831},\n",
" {'entity': 'ok',\n",
" 'score': 0.5426304,\n",
" 'index': 434,\n",
" 'word': 'With',\n",
" 'start': 1832,\n",
" 'end': 1836},\n",
" {'entity': 'ok',\n",
" 'score': 0.9795935,\n",
" 'index': 435,\n",
" 'word': 'this',\n",
" 'start': 1837,\n",
" 'end': 1841},\n",
" {'entity': 'ok',\n",
" 'score': 0.9408998,\n",
" 'index': 436,\n",
" 'word': 'camera',\n",
" 'start': 1842,\n",
" 'end': 1848},\n",
" {'entity': 'typo',\n",
" 'score': 0.9998209,\n",
" 'index': 437,\n",
" 'word': 'you',\n",
" 'start': 1849,\n",
" 'end': 1852},\n",
" {'entity': 'ok',\n",
" 'score': 0.9576567,\n",
" 'index': 438,\n",
" 'word': 'can',\n",
" 'start': 1853,\n",
" 'end': 1856},\n",
" {'entity': 'typo',\n",
" 'score': 0.573261,\n",
" 'index': 439,\n",
" 'word': 'disc',\n",
" 'start': 1857,\n",
" 'end': 1861},\n",
" {'entity': 'ok',\n",
" 'score': 0.9691068,\n",
" 'index': 440,\n",
" 'word': '##ern',\n",
" 'start': 1861,\n",
" 'end': 1864},\n",
" {'entity': 'ok',\n",
" 'score': 0.8217143,\n",
" 'index': 441,\n",
" 'word': 'things',\n",
" 'start': 1865,\n",
" 'end': 1871},\n",
" {'entity': 'typo',\n",
" 'score': 0.82698673,\n",
" 'index': 442,\n",
" 'word': 'in',\n",
" 'start': 1872,\n",
" 'end': 1874},\n",
" {'entity': 'typo',\n",
" 'score': 0.78578687,\n",
" 'index': 443,\n",
" 'word': 'a',\n",
" 'start': 1875,\n",
" 'end': 1876},\n",
" {'entity': 'typo',\n",
" 'score': 0.74662477,\n",
" 'index': 444,\n",
" 'word': 'digital',\n",
" 'start': 1877,\n",
" 'end': 1884},\n",
" {'entity': 'ok',\n",
" 'score': 0.5583283,\n",
" 'index': 445,\n",
" 'word': 'image',\n",
" 'start': 1885,\n",
" 'end': 1890},\n",
" {'entity': 'typo',\n",
" 'score': 0.99453664,\n",
" 'index': 446,\n",
" 'word': ',',\n",
" 'start': 1890,\n",
" 'end': 1891},\n",
" {'entity': 'typo',\n",
" 'score': 0.99513924,\n",
" 'index': 447,\n",
" 'word': '3',\n",
" 'start': 1892,\n",
" 'end': 1893},\n",
" {'entity': 'ok',\n",
" 'score': 0.97854996,\n",
" 'index': 448,\n",
" 'word': 'times',\n",
" 'start': 1894,\n",
" 'end': 1899},\n",
" {'entity': 'ok',\n",
" 'score': 0.9518669,\n",
" 'index': 449,\n",
" 'word': 'bigger',\n",
" 'start': 1900,\n",
" 'end': 1906},\n",
" {'entity': 'ok',\n",
" 'score': 0.9959552,\n",
" 'index': 450,\n",
" 'word': 'than',\n",
" 'start': 1907,\n",
" 'end': 1911},\n",
" {'entity': 'ok',\n",
" 'score': 0.70804185,\n",
" 'index': 451,\n",
" 'word': 'the',\n",
" 'start': 1912,\n",
" 'end': 1915},\n",
" {'entity': 'ok',\n",
" 'score': 0.7743477,\n",
" 'index': 452,\n",
" 'word': 'pi',\n",
" 'start': 1916,\n",
" 'end': 1918},\n",
" {'entity': 'ok',\n",
" 'score': 0.8590469,\n",
" 'index': 453,\n",
" 'word': '##xel',\n",
" 'start': 1918,\n",
" 'end': 1921},\n",
" {'entity': 'typo',\n",
" 'score': 0.88074666,\n",
" 'index': 454,\n",
" 'word': 'size',\n",
" 'start': 1922,\n",
" 'end': 1926},\n",
" {'entity': 'typo',\n",
" 'score': 0.977126,\n",
" 'index': 455,\n",
" 'word': 'which',\n",
" 'start': 1927,\n",
" 'end': 1932},\n",
" {'entity': 'typo',\n",
" 'score': 0.71839124,\n",
" 'index': 456,\n",
" 'word': 'means',\n",
" 'start': 1933,\n",
" 'end': 1938},\n",
" {'entity': 'ok',\n",
" 'score': 0.7110502,\n",
" 'index': 457,\n",
" 'word': 'if',\n",
" 'start': 1939,\n",
" 'end': 1941},\n",
" {'entity': 'ok',\n",
" 'score': 0.72081035,\n",
" 'index': 458,\n",
" 'word': 'there',\n",
" 'start': 1942,\n",
" 'end': 1947},\n",
" {'entity': 'ok',\n",
" 'score': 0.90700245,\n",
" 'index': 459,\n",
" 'word': 'were',\n",
" 'start': 1948,\n",
" 'end': 1952},\n",
" {'entity': 'ok',\n",
" 'score': 0.9401142,\n",
" 'index': 460,\n",
" 'word': 'any',\n",
" 'start': 1953,\n",
" 'end': 1956},\n",
" {'entity': 'ok',\n",
" 'score': 0.8010327,\n",
" 'index': 461,\n",
" 'word': 'signs',\n",
" 'start': 1957,\n",
" 'end': 1962},\n",
" {'entity': 'ok',\n",
" 'score': 0.9774201,\n",
" 'index': 462,\n",
" 'word': 'of',\n",
" 'start': 1963,\n",
" 'end': 1965},\n",
" {'entity': 'ok',\n",
" 'score': 0.72982603,\n",
" 'index': 463,\n",
" 'word': 'life',\n",
" 'start': 1966,\n",
" 'end': 1970},\n",
" {'entity': 'typo',\n",
" 'score': 0.98334414,\n",
" 'index': 464,\n",
" 'word': ',',\n",
" 'start': 1970,\n",
" 'end': 1971},\n",
" {'entity': 'typo',\n",
" 'score': 0.9922644,\n",
" 'index': 465,\n",
" 'word': 'you',\n",
" 'start': 1972,\n",
" 'end': 1975},\n",
" {'entity': 'ok',\n",
" 'score': 0.9828555,\n",
" 'index': 466,\n",
" 'word': 'could',\n",
" 'start': 1976,\n",
" 'end': 1981},\n",
" {'entity': 'ok',\n",
" 'score': 0.9977181,\n",
" 'index': 467,\n",
" 'word': 'easily',\n",
" 'start': 1982,\n",
" 'end': 1988},\n",
" {'entity': 'ok',\n",
" 'score': 0.99048984,\n",
" 'index': 468,\n",
" 'word': 'see',\n",
" 'start': 1989,\n",
" 'end': 1992},\n",
" {'entity': 'ok',\n",
" 'score': 0.99270344,\n",
" 'index': 469,\n",
" 'word': 'what',\n",
" 'start': 1993,\n",
" 'end': 1997},\n",
" {'entity': 'ok',\n",
" 'score': 0.7724791,\n",
" 'index': 470,\n",
" 'word': 'they',\n",
" 'start': 1998,\n",
" 'end': 2002},\n",
" {'entity': 'ok',\n",
" 'score': 0.9829512,\n",
" 'index': 471,\n",
" 'word': 'were',\n",
" 'start': 2003,\n",
" 'end': 2007},\n",
" {'entity': 'typo',\n",
" 'score': 0.9201129,\n",
" 'index': 472,\n",
" 'word': '.',\n",
" 'start': 2007,\n",
" 'end': 2008},\n",
" {'entity': 'typo',\n",
" 'score': 0.51953334,\n",
" 'index': 473,\n",
" 'word': 'What',\n",
" 'start': 2009,\n",
" 'end': 2013},\n",
" {'entity': 'ok',\n",
" 'score': 0.9992943,\n",
" 'index': 474,\n",
" 'word': 'the',\n",
" 'start': 2014,\n",
" 'end': 2017},\n",
" {'entity': 'ok',\n",
" 'score': 0.98957855,\n",
" 'index': 475,\n",
" 'word': 'picture',\n",
" 'start': 2018,\n",
" 'end': 2025},\n",
" {'entity': 'typo',\n",
" 'score': 0.58303297,\n",
" 'index': 476,\n",
" 'word': 'showed',\n",
" 'start': 2026,\n",
" 'end': 2032},\n",
" {'entity': 'typo',\n",
" 'score': 0.8085752,\n",
" 'index': 477,\n",
" 'word': 'was',\n",
" 'start': 2033,\n",
" 'end': 2036},\n",
" {'entity': 'typo',\n",
" 'score': 0.9929074,\n",
" 'index': 478,\n",
" 'word': 'the',\n",
" 'start': 2037,\n",
" 'end': 2040},\n",
" {'entity': 'typo',\n",
" 'score': 0.9818252,\n",
" 'index': 479,\n",
" 'word': 'but',\n",
" 'start': 2041,\n",
" 'end': 2044},\n",
" {'entity': 'ok',\n",
" 'score': 0.5864657,\n",
" 'index': 480,\n",
" 'word': '##te',\n",
" 'start': 2044,\n",
" 'end': 2046},\n",
" {'entity': 'ok',\n",
" 'score': 0.9995679,\n",
" 'index': 481,\n",
" 'word': 'or',\n",
" 'start': 2047,\n",
" 'end': 2049},\n",
" {'entity': 'ok',\n",
" 'score': 0.8257528,\n",
" 'index': 482,\n",
" 'word': 'mesa',\n",
" 'start': 2050,\n",
" 'end': 2054},\n",
" {'entity': 'typo',\n",
" 'score': 0.78061324,\n",
" 'index': 483,\n",
" 'word': ',',\n",
" 'start': 2054,\n",
" 'end': 2055},\n",
" {'entity': 'typo',\n",
" 'score': 0.529569,\n",
" 'index': 484,\n",
" 'word': 'which',\n",
" 'start': 2056,\n",
" 'end': 2061},\n",
" {'entity': 'typo',\n",
" 'score': 0.98165315,\n",
" 'index': 485,\n",
" 'word': 'are',\n",
" 'start': 2062,\n",
" 'end': 2065},\n",
" {'entity': 'ok',\n",
" 'score': 0.7917957,\n",
" 'index': 486,\n",
" 'word': 'land',\n",
" 'start': 2066,\n",
" 'end': 2070},\n",
" {'entity': 'ok',\n",
" 'score': 0.8484906,\n",
" 'index': 487,\n",
" 'word': '##form',\n",
" 'start': 2070,\n",
" 'end': 2074},\n",
" {'entity': 'typo',\n",
" 'score': 0.5294084,\n",
" 'index': 488,\n",
" 'word': '##s',\n",
" 'start': 2074,\n",
" 'end': 2075},\n",
" {'entity': 'typo',\n",
" 'score': 0.66130173,\n",
" 'index': 489,\n",
" 'word': 'common',\n",
" 'start': 2076,\n",
" 'end': 2082},\n",
" {'entity': 'typo',\n",
" 'score': 0.79395646,\n",
" 'index': 490,\n",
" 'word': 'around',\n",
" 'start': 2083,\n",
" 'end': 2089},\n",
" {'entity': 'typo',\n",
" 'score': 0.80247337,\n",
" 'index': 491,\n",
" 'word': 'the',\n",
" 'start': 2090,\n",
" 'end': 2093},\n",
" {'entity': 'typo',\n",
" 'score': 0.90861714,\n",
" 'index': 492,\n",
" 'word': 'American',\n",
" 'start': 2094,\n",
" 'end': 2102},\n",
" {'entity': 'ok',\n",
" 'score': 0.916328,\n",
" 'index': 493,\n",
" 'word': 'West',\n",
" 'start': 2103,\n",
" 'end': 2107}]"
]
},
"execution_count": 62,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from transformers import pipeline\n",
"\n",
"typo_checker = pipeline(\n",
" \"ner\",\n",
" model=\"mrm8488/distilbert-base-multi-cased-finetuned-typo-detection\",\n",
" tokenizer=\"mrm8488/distilbert-base-multi-cased-finetuned-typo-detection\"\n",
")\n",
"\n",
"result = typo_checker(text)\n",
"result[1:-1]\n"
]
},
{
"cell_type": "code",
"execution_count": 63,
"id": "7b8311a5-68cc-45cb-b2a8-94495aa18113",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"entity\n",
"ok 328\n",
"typo 164\n",
"dtype: int64\n"
]
},
{
"data": {
"text/plain": [
"entity word \n",
"ok 7\n",
" ##adow 3\n",
" ##ara 1\n",
" ##biter 1\n",
" ##con 1\n",
" ..\n",
"typo when 1\n",
" which 4\n",
" wrong 1\n",
" ye 1\n",
" you 3\n",
"Length: 298, dtype: int64"
]
},
"execution_count": 63,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"with open(\"5 metricas mrm8488.json\", encoding='utf-8') as f:\n",
" data = json.load(f)\n",
"aux=obtener_dataframe(data)\n",
"aux\n",
"print(aux.groupby(['entity']).size())\n",
"aux.groupby(['entity', 'word']).size()"
]
},
{
"cell_type": "markdown",
"id": "25e685d8-ff8a-43ec-8ab0-958a89056ee8",
"metadata": {},
"source": [
"## 6 sagorsarker/codeswitch-spaeng-ner-lince "
]
},
{
"cell_type": "code",
"execution_count": 128,
"id": "5e5dcd1c-f49d-486d-9d72-e0ce4498a565",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Some weights of the model checkpoint at sagorsarker/codeswitch-spaeng-ner-lince were not used when initializing BertForTokenClassification: ['bert.pooler.dense.bias', 'bert.pooler.dense.weight']\n",
"- This IS expected if you are initializing BertForTokenClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
"- This IS NOT expected if you are initializing BertForTokenClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n"
]
},
{
"data": {
"text/plain": [
"[{'entity': 'B-ORG',\n",
" 'score': 0.90762955,\n",
" 'index': 8,\n",
" 'word': 'NASA',\n",
" 'start': 16,\n",
" 'end': 20},\n",
" {'entity': 'B-LOC',\n",
" 'score': 0.45779723,\n",
" 'index': 25,\n",
" 'word': 'Mars',\n",
" 'start': 96,\n",
" 'end': 100},\n",
" {'entity': 'B-LOC',\n",
" 'score': 0.67780584,\n",
" 'index': 37,\n",
" 'word': 'Mars',\n",
" 'start': 152,\n",
" 'end': 156},\n",
" {'entity': 'B-PROD',\n",
" 'score': 0.823176,\n",
" 'index': 60,\n",
" 'word': 'Viking',\n",
" 'start': 240,\n",
" 'end': 246},\n",
" {'entity': 'I-PROD',\n",
" 'score': 0.6758249,\n",
" 'index': 61,\n",
" 'word': '1',\n",
" 'start': 247,\n",
" 'end': 248},\n",
" {'entity': 'B-OTHER',\n",
" 'score': 0.5148923,\n",
" 'index': 98,\n",
" 'word': 'Mart',\n",
" 'start': 407,\n",
" 'end': 411},\n",
" {'entity': 'B-OTHER',\n",
" 'score': 0.42019445,\n",
" 'index': 99,\n",
" 'word': '##ian',\n",
" 'start': 411,\n",
" 'end': 414},\n",
" {'entity': 'B-LOC',\n",
" 'score': 0.76582533,\n",
" 'index': 104,\n",
" 'word': 'C',\n",
" 'start': 435,\n",
" 'end': 436},\n",
" {'entity': 'B-LOC',\n",
" 'score': 0.74930793,\n",
" 'index': 105,\n",
" 'word': '##yd',\n",
" 'start': 436,\n",
" 'end': 438},\n",
" {'entity': 'I-LOC',\n",
" 'score': 0.34308487,\n",
" 'index': 106,\n",
" 'word': '##onia',\n",
" 'start': 438,\n",
" 'end': 442},\n",
" {'entity': 'B-OTHER',\n",
" 'score': 0.60952294,\n",
" 'index': 121,\n",
" 'word': 'Egypt',\n",
" 'start': 496,\n",
" 'end': 501},\n",
" {'entity': 'I-OTHER',\n",
" 'score': 0.6259159,\n",
" 'index': 122,\n",
" 'word': '##ion',\n",
" 'start': 501,\n",
" 'end': 504},\n",
" {'entity': 'B-ORG',\n",
" 'score': 0.9731985,\n",
" 'index': 194,\n",
" 'word': 'NASA',\n",
" 'start': 801,\n",
" 'end': 805},\n",
" {'entity': 'B-LOC',\n",
" 'score': 0.62910813,\n",
" 'index': 205,\n",
" 'word': 'Mars',\n",
" 'start': 843,\n",
" 'end': 847},\n",
" {'entity': 'B-LOC',\n",
" 'score': 0.6685378,\n",
" 'index': 215,\n",
" 'word': 'Mars',\n",
" 'start': 875,\n",
" 'end': 879},\n",
" {'entity': 'B-LOC',\n",
" 'score': 0.71298903,\n",
" 'index': 263,\n",
" 'word': 'Mars',\n",
" 'start': 1088,\n",
" 'end': 1092},\n",
" {'entity': 'B-ORG',\n",
" 'score': 0.9437394,\n",
" 'index': 282,\n",
" 'word': 'NASA',\n",
" 'start': 1169,\n",
" 'end': 1173},\n",
" {'entity': 'B-LOC',\n",
" 'score': 0.60219324,\n",
" 'index': 291,\n",
" 'word': 'Mars',\n",
" 'start': 1220,\n",
" 'end': 1224},\n",
" {'entity': 'B-TIME',\n",
" 'score': 0.9826582,\n",
" 'index': 311,\n",
" 'word': 'April',\n",
" 'start': 1297,\n",
" 'end': 1302},\n",
" {'entity': 'I-TIME',\n",
" 'score': 0.9323657,\n",
" 'index': 312,\n",
" 'word': '5',\n",
" 'start': 1303,\n",
" 'end': 1304},\n",
" {'entity': 'I-TIME',\n",
" 'score': 0.8260366,\n",
" 'index': 313,\n",
" 'word': ',',\n",
" 'start': 1304,\n",
" 'end': 1305},\n",
" {'entity': 'I-TIME',\n",
" 'score': 0.887798,\n",
" 'index': 314,\n",
" 'word': '1998',\n",
" 'start': 1306,\n",
" 'end': 1310},\n",
" {'entity': 'B-PER',\n",
" 'score': 0.9961196,\n",
" 'index': 316,\n",
" 'word': 'Michael',\n",
" 'start': 1312,\n",
" 'end': 1319},\n",
" {'entity': 'I-PER',\n",
" 'score': 0.99166673,\n",
" 'index': 317,\n",
" 'word': 'Mali',\n",
" 'start': 1320,\n",
" 'end': 1324},\n",
" {'entity': 'I-PER',\n",
" 'score': 0.9911315,\n",
" 'index': 318,\n",
" 'word': '##n',\n",
" 'start': 1324,\n",
" 'end': 1325},\n",
" {'entity': 'B-PROD',\n",
" 'score': 0.6550399,\n",
" 'index': 321,\n",
" 'word': 'Mars',\n",
" 'start': 1334,\n",
" 'end': 1338},\n",
" {'entity': 'I-PROD',\n",
" 'score': 0.80179524,\n",
" 'index': 322,\n",
" 'word': 'Or',\n",
" 'start': 1339,\n",
" 'end': 1341},\n",
" {'entity': 'I-PROD',\n",
" 'score': 0.8049131,\n",
" 'index': 323,\n",
" 'word': '##biter',\n",
" 'start': 1341,\n",
" 'end': 1346},\n",
" {'entity': 'B-OTHER',\n",
" 'score': 0.35082105,\n",
" 'index': 338,\n",
" 'word': 'Viking',\n",
" 'start': 1419,\n",
" 'end': 1425},\n",
" {'entity': 'B-TIME',\n",
" 'score': 0.97395533,\n",
" 'index': 394,\n",
" 'word': 'April',\n",
" 'start': 1652,\n",
" 'end': 1657},\n",
" {'entity': 'I-TIME',\n",
" 'score': 0.66215277,\n",
" 'index': 395,\n",
" 'word': '8',\n",
" 'start': 1658,\n",
" 'end': 1659},\n",
" {'entity': 'B-GROUP',\n",
" 'score': 0.5102115,\n",
" 'index': 415,\n",
" 'word': 'Mali',\n",
" 'start': 1745,\n",
" 'end': 1749},\n",
" {'entity': 'B-GROUP',\n",
" 'score': 0.34341586,\n",
" 'index': 416,\n",
" 'word': '##n',\n",
" 'start': 1749,\n",
" 'end': 1750},\n",
" {'entity': 'B-LOC',\n",
" 'score': 0.8453891,\n",
" 'index': 492,\n",
" 'word': 'American',\n",
" 'start': 2094,\n",
" 'end': 2102},\n",
" {'entity': 'I-LOC',\n",
" 'score': 0.663866,\n",
" 'index': 493,\n",
" 'word': 'West',\n",
" 'start': 2103,\n",
" 'end': 2107}]"
]
},
"execution_count": 128,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from transformers import AutoTokenizer, AutoModelForTokenClassification, pipeline\n",
"\n",
"tokenizer = AutoTokenizer.from_pretrained(\"sagorsarker/codeswitch-spaeng-ner-lince\")\n",
"\n",
"model = AutoModelForTokenClassification.from_pretrained(\"sagorsarker/codeswitch-spaeng-ner-lince\")\n",
"\n",
"ner_model = pipeline('ner', model=model, tokenizer=tokenizer)\n",
"\n",
"ner_model(text)"
]
},
{
"cell_type": "code",
"execution_count": 66,
"id": "58c5a81d-c1c6-48db-b26d-edd1e01d1231",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"entity\n",
"B-GROUP 2\n",
"B-LOC 9\n",
"B-ORG 3\n",
"B-OTHER 4\n",
"B-PER 1\n",
"B-PROD 2\n",
"B-TIME 2\n",
"I-LOC 2\n",
"I-OTHER 1\n",
"I-PER 2\n",
"I-PROD 3\n",
"I-TIME 4\n",
"dtype: int64\n"
]
},
{
"data": {
"text/plain": [
"entity word \n",
"B-GROUP ##n 1\n",
" Mali 1\n",
"B-LOC ##yd 1\n",
" American 1\n",
" C 1\n",
" Mars 6\n",
"B-ORG NASA 3\n",
"B-OTHER ##ian 1\n",
" Egypt 1\n",
" Mart 1\n",
" Viking 1\n",
"B-PER Michael 1\n",
"B-PROD Mars 1\n",
" Viking 1\n",
"B-TIME April 2\n",
"I-LOC ##onia 1\n",
" West 1\n",
"I-OTHER ##ion 1\n",
"I-PER ##n 1\n",
" Mali 1\n",
"I-PROD ##biter 1\n",
" 1 1\n",
" Or 1\n",
"I-TIME , 1\n",
" 1998 1\n",
" 5 1\n",
" 8 1\n",
"dtype: int64"
]
},
"execution_count": 66,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"with open(\"6 sagorsarkercodeswitch-spaeng-ner-lince.json\", encoding='utf-8') as f:\n",
" data = json.load(f)\n",
"aux=obtener_dataframe(data)\n",
"aux\n",
"print(aux.groupby(['entity']).size())\n",
"aux.groupby(['entity', 'word']).size()"
]
},
{
"cell_type": "markdown",
"id": "9c02ef71-62d9-40f1-98a8-138419cff895",
"metadata": {},
"source": [
"## 7 gunghio/xlm-roberta-base-finetuned-panx-ner"
]
},
{
"cell_type": "code",
"execution_count": 68,
"id": "a8df4657-5842-43c0-97b8-15346f5a4578",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"[{'entity_group': 'ORG', 'score': 0.9839335, 'word': 'NASA', 'start': 16, 'end': 20}, {'entity_group': 'ORG', 'score': 0.9257912, 'word': 'FaceOnMars,', 'start': 88, 'end': 101}, {'entity_group': 'LOC', 'score': 0.8524113, 'word': 'Mars,', 'start': 152, 'end': 157}, {'entity_group': 'ORG', 'score': 0.9470773, 'word': 'Viking1spacecraft', 'start': 240, 'end': 259}, {'entity_group': 'ORG', 'score': 0.6181985, 'word': 'Martian', 'start': 407, 'end': 414}, {'entity_group': 'LOC', 'score': 0.49310815, 'word': 'mesa,', 'start': 415, 'end': 420}, {'entity_group': 'LOC', 'score': 0.8700732, 'word': 'Cydonia,', 'start': 435, 'end': 443}, {'entity_group': 'ORG', 'score': 0.57100993, 'word': 'EgyptionPharaoh.', 'start': 496, 'end': 513}, {'entity_group': 'LOC', 'score': 0.46428245, 'word': 'formation', 'start': 622, 'end': 631}, {'entity_group': 'ORG', 'score': 0.90882486, 'word': \"NASA's\", 'start': 801, 'end': 807}, {'entity_group': 'LOC', 'score': 0.5685518, 'word': 'Mars--', 'start': 843, 'end': 849}, {'entity_group': 'LOC', 'score': 0.89254224, 'word': 'Mars', 'start': 875, 'end': 879}, {'entity_group': 'LOC', 'score': 0.8768112, 'word': 'Mars,', 'start': 1088, 'end': 1093}, {'entity_group': 'ORG', 'score': 0.88662714, 'word': 'NASAbudget', 'start': 1169, 'end': 1180}, {'entity_group': 'ORG', 'score': 0.5386654, 'word': 'ancientcivilization', 'start': 1196, 'end': 1216}, {'entity_group': 'LOC', 'score': 0.6887058, 'word': 'Mars.', 'start': 1220, 'end': 1225}, {'entity_group': 'PER', 'score': 0.86852753, 'word': 'MichaelMalin', 'start': 1312, 'end': 1325}, {'entity_group': 'ORG', 'score': 0.9244041, 'word': 'MarsOrbitercamerateam', 'start': 1334, 'end': 1358}, {'entity_group': 'ORG', 'score': 0.65044224, 'word': 'Vikingphotos,', 'start': 1419, 'end': 1433}, {'entity_group': 'ORG', 'score': 0.6091294, 'word': 'digitalimage,', 'start': 1877, 'end': 1891}, {'entity_group': 'LOC', 'score': 0.93108547, 'word': 'AmericanWest.', 'start': 2094, 'end': 2108}]\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"C:\\Users\\NW\\anaconda3\\Lib\\site-packages\\transformers\\pipelines\\token_classification.py:392: UserWarning: Tokenizer does not support real words, using fallback heuristic\n",
" warnings.warn(\n"
]
}
],
"source": [
"from transformers import AutoTokenizer, AutoModelForTokenClassification\n",
"from transformers import pipeline\n",
"\n",
"tokenizer = AutoTokenizer.from_pretrained(\"gunghio/xlm-roberta-base-finetuned-panx-ner\")\n",
"model = AutoModelForTokenClassification.from_pretrained(\"gunghio/xlm-roberta-base-finetuned-panx-ner\")\n",
"\n",
"nlp = pipeline(\"ner\", model=model, tokenizer=tokenizer, aggregation_strategy=\"first\")\n",
"example = text\n",
"\n",
"ner_results = nlp(example)\n",
"print(ner_results)"
]
},
{
"cell_type": "code",
"execution_count": 69,
"id": "15eb816d-ac29-448d-8520-fd3cd2d3d957",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"entity_group\n",
"LOC 9\n",
"ORG 11\n",
"PER 1\n",
"dtype: int64\n"
]
},
{
"data": {
"text/plain": [
"entity_group word \n",
"LOC AmericanWest. 1\n",
" Cydonia, 1\n",
" Mars 1\n",
" Mars, 2\n",
" Mars-- 1\n",
" Mars. 1\n",
" formation 1\n",
" mesa, 1\n",
"ORG EgyptionPharaoh. 1\n",
" FaceOnMars, 1\n",
" MarsOrbitercamerateam 1\n",
" Martian 1\n",
" NASA 1\n",
" NASA's 1\n",
" NASAbudget 1\n",
" Viking1spacecraft 1\n",
" Vikingphotos, 1\n",
" ancientcivilization 1\n",
" digitalimage, 1\n",
"PER MichaelMalin 1\n",
"dtype: int64"
]
},
"execution_count": 69,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"\n",
"with open(\"7 gunghioxlm-roberta-base-finetuned-panx-ner.json\", encoding='utf-8') as f:\n",
" data = json.load(f)\n",
"aux=obtener_dataframe(data)\n",
"aux\n",
"print(aux.groupby(['entity_group']).size())\n",
"aux.groupby(['entity_group', 'word']).size()"
]
},
{
"cell_type": "markdown",
"id": "7a419ca8-e000-4bf0-81fa-de8970e98a22",
"metadata": {},
"source": [
"## 8 51la5/roberta-large-NER"
]
},
{
"cell_type": "code",
"execution_count": 71,
"id": "92d64393-7a73-4803-9fa5-043de160cd45",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"C:\\Users\\NW\\anaconda3\\Lib\\site-packages\\huggingface_hub\\file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.\n",
" warnings.warn(\n",
"Some weights of the model checkpoint at 51la5/roberta-large-NER were not used when initializing XLMRobertaForTokenClassification: ['roberta.pooler.dense.bias', 'roberta.pooler.dense.weight']\n",
"- This IS expected if you are initializing XLMRobertaForTokenClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
"- This IS NOT expected if you are initializing XLMRobertaForTokenClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
"Asking to truncate to max_length but no maximum length is provided and the model has no predefined maximum length. Default to no truncation.\n"
]
},
{
"data": {
"text/plain": [
"[{'entity': 'I-ORG',\n",
" 'score': 0.9999913,\n",
" 'index': 8,\n",
" 'word': '▁NASA',\n",
" 'start': 16,\n",
" 'end': 20},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.9999641,\n",
" 'index': 23,\n",
" 'word': '▁Face',\n",
" 'start': 88,\n",
" 'end': 92},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.99989665,\n",
" 'index': 24,\n",
" 'word': '▁On',\n",
" 'start': 93,\n",
" 'end': 95},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.97350365,\n",
" 'index': 25,\n",
" 'word': '▁Mars',\n",
" 'start': 96,\n",
" 'end': 100},\n",
" {'entity': 'I-LOC',\n",
" 'score': 0.9999362,\n",
" 'index': 36,\n",
" 'word': '▁Mars',\n",
" 'start': 152,\n",
" 'end': 156},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.9992086,\n",
" 'index': 58,\n",
" 'word': '▁Viking',\n",
" 'start': 240,\n",
" 'end': 246},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.9989502,\n",
" 'index': 59,\n",
" 'word': '▁1',\n",
" 'start': 247,\n",
" 'end': 248},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.999977,\n",
" 'index': 97,\n",
" 'word': '▁Marti',\n",
" 'start': 407,\n",
" 'end': 412},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.99619055,\n",
" 'index': 98,\n",
" 'word': 'an',\n",
" 'start': 412,\n",
" 'end': 414},\n",
" {'entity': 'I-LOC',\n",
" 'score': 0.9999354,\n",
" 'index': 103,\n",
" 'word': '▁Cy',\n",
" 'start': 435,\n",
" 'end': 437},\n",
" {'entity': 'I-LOC',\n",
" 'score': 0.99994576,\n",
" 'index': 104,\n",
" 'word': 'do',\n",
" 'start': 437,\n",
" 'end': 439},\n",
" {'entity': 'I-LOC',\n",
" 'score': 0.99992585,\n",
" 'index': 105,\n",
" 'word': 'nia',\n",
" 'start': 439,\n",
" 'end': 442},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.9999789,\n",
" 'index': 119,\n",
" 'word': '▁Egypt',\n",
" 'start': 496,\n",
" 'end': 501},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.9614088,\n",
" 'index': 120,\n",
" 'word': 'ion',\n",
" 'start': 501,\n",
" 'end': 504},\n",
" {'entity': 'I-ORG',\n",
" 'score': 0.99997246,\n",
" 'index': 193,\n",
" 'word': '▁NASA',\n",
" 'start': 801,\n",
" 'end': 805},\n",
" {'entity': 'I-LOC',\n",
" 'score': 0.99979633,\n",
" 'index': 205,\n",
" 'word': '▁Mars',\n",
" 'start': 843,\n",
" 'end': 847},\n",
" {'entity': 'I-LOC',\n",
" 'score': 0.9998061,\n",
" 'index': 215,\n",
" 'word': '▁Mars',\n",
" 'start': 875,\n",
" 'end': 879},\n",
" {'entity': 'I-LOC',\n",
" 'score': 0.99984956,\n",
" 'index': 264,\n",
" 'word': '▁Mars',\n",
" 'start': 1088,\n",
" 'end': 1092},\n",
" {'entity': 'I-ORG',\n",
" 'score': 0.99996305,\n",
" 'index': 285,\n",
" 'word': '▁NASA',\n",
" 'start': 1169,\n",
" 'end': 1173},\n",
" {'entity': 'I-LOC',\n",
" 'score': 0.9998203,\n",
" 'index': 295,\n",
" 'word': '▁Mars',\n",
" 'start': 1220,\n",
" 'end': 1224},\n",
" {'entity': 'I-PER',\n",
" 'score': 0.9999932,\n",
" 'index': 319,\n",
" 'word': '▁Michael',\n",
" 'start': 1312,\n",
" 'end': 1319},\n",
" {'entity': 'I-PER',\n",
" 'score': 0.99999106,\n",
" 'index': 320,\n",
" 'word': '▁Malin',\n",
" 'start': 1320,\n",
" 'end': 1325},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.94105357,\n",
" 'index': 323,\n",
" 'word': '▁Mars',\n",
" 'start': 1334,\n",
" 'end': 1338},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.9839579,\n",
" 'index': 324,\n",
" 'word': '▁Or',\n",
" 'start': 1339,\n",
" 'end': 1341},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.9913346,\n",
" 'index': 325,\n",
" 'word': 'bit',\n",
" 'start': 1341,\n",
" 'end': 1344},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.9759228,\n",
" 'index': 326,\n",
" 'word': 'er',\n",
" 'start': 1344,\n",
" 'end': 1346},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.999749,\n",
" 'index': 341,\n",
" 'word': '▁Viking',\n",
" 'start': 1419,\n",
" 'end': 1425},\n",
" {'entity': 'I-PER',\n",
" 'score': 0.9999914,\n",
" 'index': 416,\n",
" 'word': '▁Malin',\n",
" 'start': 1745,\n",
" 'end': 1750},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.92417294,\n",
" 'index': 491,\n",
" 'word': '▁American',\n",
" 'start': 2094,\n",
" 'end': 2102},\n",
" {'entity': 'I-LOC',\n",
" 'score': 0.99954396,\n",
" 'index': 492,\n",
" 'word': '▁West',\n",
" 'start': 2103,\n",
" 'end': 2107}]"
]
},
"execution_count": 71,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from transformers import pipeline\n",
"tokenizer = AutoTokenizer.from_pretrained(\"51la5/roberta-large-NER\")\n",
"model = AutoModelForTokenClassification.from_pretrained(\"51la5/roberta-large-NER\")\n",
"classifier = pipeline(\"ner\", model=model, tokenizer=tokenizer)\n",
"classifier(text)"
]
},
{
"cell_type": "code",
"execution_count": 72,
"id": "9cf28bd0-f78c-48f1-98f0-e6b8dea07e21",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"entity\n",
"I-LOC 9\n",
"I-MISC 15\n",
"I-ORG 3\n",
"I-PER 3\n",
"dtype: int64\n"
]
},
{
"data": {
"text/plain": [
"entity word \n",
"I-LOC do 1\n",
" nia 1\n",
" ▁Cy 1\n",
" ▁Mars 5\n",
" ▁West 1\n",
"I-MISC an 1\n",
" bit 1\n",
" er 1\n",
" ion 1\n",
" ▁1 1\n",
" ▁American 1\n",
" ▁Egypt 1\n",
" ▁Face 1\n",
" ▁Mars 2\n",
" ▁Marti 1\n",
" ▁On 1\n",
" ▁Or 1\n",
" ▁Viking 2\n",
"I-ORG ▁NASA 3\n",
"I-PER ▁Malin 2\n",
" ▁Michael 1\n",
"dtype: int64"
]
},
"execution_count": 72,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"with open(\"8 51la5roberta-large-NER.json\", encoding='utf-8') as f:\n",
" data = json.load(f)\n",
"aux=obtener_dataframe(data)\n",
"aux\n",
"print(aux.groupby(['entity']).size())\n",
"aux.groupby(['entity', 'word']).size()"
]
},
{
"cell_type": "markdown",
"id": "c6a46082-0ffa-413c-ae69-4ffbf2081e71",
"metadata": {},
"source": [
"## 9 dmargutierrezdistilbert-base-multilingual-cased-mapa_coarse-ner"
]
},
{
"cell_type": "code",
"execution_count": 74,
"id": "e832ae0c-3ecd-4080-aeee-70b0fac8db0b",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[{'entity': 'B-ADDRESS',\n",
" 'score': 0.89953065,\n",
" 'index': 104,\n",
" 'word': 'C',\n",
" 'start': 435,\n",
" 'end': 436},\n",
" {'entity': 'I-ADDRESS',\n",
" 'score': 0.776557,\n",
" 'index': 105,\n",
" 'word': '##yd',\n",
" 'start': 436,\n",
" 'end': 438},\n",
" {'entity': 'I-ADDRESS',\n",
" 'score': 0.87639356,\n",
" 'index': 106,\n",
" 'word': '##onia',\n",
" 'start': 438,\n",
" 'end': 442},\n",
" {'entity': 'B-AMOUNT',\n",
" 'score': 0.94337094,\n",
" 'index': 248,\n",
" 'word': '25',\n",
" 'start': 1013,\n",
" 'end': 1015},\n",
" {'entity': 'I-AMOUNT',\n",
" 'score': 0.83502764,\n",
" 'index': 249,\n",
" 'word': 'years',\n",
" 'start': 1016,\n",
" 'end': 1021},\n",
" {'entity': 'B-DATE',\n",
" 'score': 0.993107,\n",
" 'index': 311,\n",
" 'word': 'April',\n",
" 'start': 1297,\n",
" 'end': 1302},\n",
" {'entity': 'I-DATE',\n",
" 'score': 0.9911287,\n",
" 'index': 312,\n",
" 'word': '5',\n",
" 'start': 1303,\n",
" 'end': 1304},\n",
" {'entity': 'I-DATE',\n",
" 'score': 0.9842742,\n",
" 'index': 313,\n",
" 'word': ',',\n",
" 'start': 1304,\n",
" 'end': 1305},\n",
" {'entity': 'I-DATE',\n",
" 'score': 0.9907127,\n",
" 'index': 314,\n",
" 'word': '1998',\n",
" 'start': 1306,\n",
" 'end': 1310},\n",
" {'entity': 'B-PERSON',\n",
" 'score': 0.9272426,\n",
" 'index': 316,\n",
" 'word': 'Michael',\n",
" 'start': 1312,\n",
" 'end': 1319},\n",
" {'entity': 'I-PERSON',\n",
" 'score': 0.9829417,\n",
" 'index': 317,\n",
" 'word': 'Mali',\n",
" 'start': 1320,\n",
" 'end': 1324},\n",
" {'entity': 'I-PERSON',\n",
" 'score': 0.96556324,\n",
" 'index': 318,\n",
" 'word': '##n',\n",
" 'start': 1324,\n",
" 'end': 1325},\n",
" {'entity': 'B-ORGANISATION',\n",
" 'score': 0.92684674,\n",
" 'index': 321,\n",
" 'word': 'Mars',\n",
" 'start': 1334,\n",
" 'end': 1338},\n",
" {'entity': 'I-ORGANISATION',\n",
" 'score': 0.9111312,\n",
" 'index': 322,\n",
" 'word': 'Or',\n",
" 'start': 1339,\n",
" 'end': 1341},\n",
" {'entity': 'I-ORGANISATION',\n",
" 'score': 0.8044608,\n",
" 'index': 323,\n",
" 'word': '##biter',\n",
" 'start': 1341,\n",
" 'end': 1346},\n",
" {'entity': 'B-DATE',\n",
" 'score': 0.9925527,\n",
" 'index': 394,\n",
" 'word': 'April',\n",
" 'start': 1652,\n",
" 'end': 1657},\n",
" {'entity': 'I-DATE',\n",
" 'score': 0.989077,\n",
" 'index': 395,\n",
" 'word': '8',\n",
" 'start': 1658,\n",
" 'end': 1659},\n",
" {'entity': 'I-DATE',\n",
" 'score': 0.9804143,\n",
" 'index': 396,\n",
" 'word': ',',\n",
" 'start': 1659,\n",
" 'end': 1660},\n",
" {'entity': 'I-DATE',\n",
" 'score': 0.989312,\n",
" 'index': 397,\n",
" 'word': '2001',\n",
" 'start': 1661,\n",
" 'end': 1665},\n",
" {'entity': 'B-PERSON',\n",
" 'score': 0.6016297,\n",
" 'index': 415,\n",
" 'word': 'Mali',\n",
" 'start': 1745,\n",
" 'end': 1749},\n",
" {'entity': 'I-PERSON',\n",
" 'score': 0.7576901,\n",
" 'index': 416,\n",
" 'word': '##n',\n",
" 'start': 1749,\n",
" 'end': 1750},\n",
" {'entity': 'B-ADDRESS',\n",
" 'score': 0.48142406,\n",
" 'index': 492,\n",
" 'word': 'American',\n",
" 'start': 2094,\n",
" 'end': 2102},\n",
" {'entity': 'I-ADDRESS',\n",
" 'score': 0.456201,\n",
" 'index': 493,\n",
" 'word': 'West',\n",
" 'start': 2103,\n",
" 'end': 2107}]"
]
},
"execution_count": 74,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from transformers import pipeline\n",
"tokenizer = AutoTokenizer.from_pretrained(\"dmargutierrez/distilbert-base-multilingual-cased-mapa_coarse-ner\")\n",
"model = AutoModelForTokenClassification.from_pretrained(\"dmargutierrez/distilbert-base-multilingual-cased-mapa_coarse-ner\")\n",
"classifier = pipeline(\"ner\", model=model, tokenizer=tokenizer)\n",
"classifier(text)"
]
},
{
"cell_type": "code",
"execution_count": 75,
"id": "3464c046-a851-4cb6-a7fe-5a71e03b2c70",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"entity\n",
"B-ADDRESS 2\n",
"B-AMOUNT 1\n",
"B-DATE 2\n",
"B-ORGANISATION 1\n",
"B-PERSON 2\n",
"I-ADDRESS 3\n",
"I-AMOUNT 1\n",
"I-DATE 6\n",
"I-ORGANISATION 2\n",
"I-PERSON 3\n",
"dtype: int64\n"
]
},
{
"data": {
"text/plain": [
"entity word \n",
"B-ADDRESS American 1\n",
" C 1\n",
"B-AMOUNT 25 1\n",
"B-DATE April 2\n",
"B-ORGANISATION Mars 1\n",
"B-PERSON Mali 1\n",
" Michael 1\n",
"I-ADDRESS ##onia 1\n",
" ##yd 1\n",
" West 1\n",
"I-AMOUNT years 1\n",
"I-DATE , 2\n",
" 1998 1\n",
" 2001 1\n",
" 5 1\n",
" 8 1\n",
"I-ORGANISATION ##biter 1\n",
" Or 1\n",
"I-PERSON ##n 2\n",
" Mali 1\n",
"dtype: int64"
]
},
"execution_count": 75,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"with open(\"9 dmargutierrezdistilbert-base-multilingual-cased-mapa_coarse-ner.json\", encoding='utf-8') as f:\n",
" data = json.load(f)\n",
"aux=obtener_dataframe(data)\n",
"aux\n",
"print(aux.groupby(['entity']).size())\n",
"aux.groupby(['entity', 'word']).size()"
]
},
{
"cell_type": "markdown",
"id": "d464a1a6-7472-4961-907c-a58292fb42b2",
"metadata": {},
"source": [
"## 10 mbrutonspa_enpt_mBERT"
]
},
{
"cell_type": "code",
"execution_count": 77,
"id": "038960d7-d348-4025-b336-acd0b7a6827b",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[{'entity': 'r0:arg1|tem',\n",
" 'score': 0.8560656,\n",
" 'index': 4,\n",
" 'word': 'you',\n",
" 'start': 7,\n",
" 'end': 10},\n",
" {'entity': 'r0:root',\n",
" 'score': 0.99593973,\n",
" 'index': 6,\n",
" 'word': 're',\n",
" 'start': 11,\n",
" 'end': 13},\n",
" {'entity': 'r1:arg1|tem',\n",
" 'score': 0.7426193,\n",
" 'index': 11,\n",
" 'word': 'you',\n",
" 'start': 32,\n",
" 'end': 35},\n",
" {'entity': 'r1:root',\n",
" 'score': 0.94838774,\n",
" 'index': 13,\n",
" 'word': 'be',\n",
" 'start': 43,\n",
" 'end': 45},\n",
" {'entity': 'r1:arg2|atr',\n",
" 'score': 0.64391094,\n",
" 'index': 14,\n",
" 'word': 'able',\n",
" 'start': 46,\n",
" 'end': 50},\n",
" {'entity': 'r2:root',\n",
" 'score': 0.80442923,\n",
" 'index': 16,\n",
" 'word': 'tell',\n",
" 'start': 54,\n",
" 'end': 58},\n",
" {'entity': 'r1:arg1|pat',\n",
" 'score': 0.33438408,\n",
" 'index': 20,\n",
" 'word': 'story',\n",
" 'start': 72,\n",
" 'end': 77},\n",
" {'entity': 'r4:arg1|tem',\n",
" 'score': 0.34810358,\n",
" 'index': 27,\n",
" 'word': 'which',\n",
" 'start': 102,\n",
" 'end': 107},\n",
" {'entity': 'r5:root',\n",
" 'score': 0.5292147,\n",
" 'index': 30,\n",
" 'word': 'is',\n",
" 'start': 118,\n",
" 'end': 120},\n",
" {'entity': 'r4:arg2|atr',\n",
" 'score': 0.46503437,\n",
" 'index': 31,\n",
" 'word': 'evidence',\n",
" 'start': 121,\n",
" 'end': 129},\n",
" {'entity': 'r6:arg2|atr',\n",
" 'score': 0.07937594,\n",
" 'index': 33,\n",
" 'word': 'there',\n",
" 'start': 135,\n",
" 'end': 140},\n",
" {'entity': 'r6:root',\n",
" 'score': 0.55574423,\n",
" 'index': 34,\n",
" 'word': 'is',\n",
" 'start': 141,\n",
" 'end': 143},\n",
" {'entity': 'r5:arg1|tem',\n",
" 'score': 0.38110724,\n",
" 'index': 35,\n",
" 'word': 'life',\n",
" 'start': 144,\n",
" 'end': 148},\n",
" {'entity': 'r7:arg1|pat',\n",
" 'score': 0.16291595,\n",
" 'index': 42,\n",
" 'word': 'face',\n",
" 'start': 171,\n",
" 'end': 175},\n",
" {'entity': 'r7:root',\n",
" 'score': 0.6100769,\n",
" 'index': 44,\n",
" 'word': 'created',\n",
" 'start': 180,\n",
" 'end': 187},\n",
" {'entity': 'r7:arg0|agt',\n",
" 'score': 0.15812016,\n",
" 'index': 45,\n",
" 'word': 'by',\n",
" 'start': 188,\n",
" 'end': 190},\n",
" {'entity': 'r6:arg0|agt',\n",
" 'score': 0.18646379,\n",
" 'index': 62,\n",
" 'word': 'spacecraft',\n",
" 'start': 249,\n",
" 'end': 259},\n",
" {'entity': 'r7:root',\n",
" 'score': 0.6611128,\n",
" 'index': 64,\n",
" 'word': 'ci',\n",
" 'start': 264,\n",
" 'end': 266},\n",
" {'entity': 'r7:root',\n",
" 'score': 0.5775215,\n",
" 'index': 65,\n",
" 'word': '##rc',\n",
" 'start': 266,\n",
" 'end': 268},\n",
" {'entity': 'r7:root',\n",
" 'score': 0.35779056,\n",
" 'index': 66,\n",
" 'word': '##ling',\n",
" 'start': 268,\n",
" 'end': 272},\n",
" {'entity': 'r7:arg1|pat',\n",
" 'score': 0.39385104,\n",
" 'index': 68,\n",
" 'word': 'planet',\n",
" 'start': 277,\n",
" 'end': 283},\n",
" {'entity': 'r7:root',\n",
" 'score': 0.42999077,\n",
" 'index': 70,\n",
" 'word': 'sna',\n",
" 'start': 285,\n",
" 'end': 288},\n",
" {'entity': 'r8:root',\n",
" 'score': 0.26388708,\n",
" 'index': 71,\n",
" 'word': '##pping',\n",
" 'start': 288,\n",
" 'end': 293},\n",
" {'entity': 'r7:arg1|pat',\n",
" 'score': 0.2658881,\n",
" 'index': 72,\n",
" 'word': 'photos',\n",
" 'start': 294,\n",
" 'end': 300},\n",
" {'entity': 'r8:arg0|agt',\n",
" 'score': 0.19982801,\n",
" 'index': 75,\n",
" 'word': 'it',\n",
" 'start': 307,\n",
" 'end': 309},\n",
" {'entity': 'r8:root',\n",
" 'score': 0.3022762,\n",
" 'index': 76,\n",
" 'word': 'spotted',\n",
" 'start': 310,\n",
" 'end': 317},\n",
" {'entity': 'r8:arg1|pat',\n",
" 'score': 0.08930407,\n",
" 'index': 81,\n",
" 'word': 'like',\n",
" 'start': 330,\n",
" 'end': 334},\n",
" {'entity': 'r8:arg1|pat',\n",
" 'score': 0.09168834,\n",
" 'index': 82,\n",
" 'word': '##ness',\n",
" 'start': 334,\n",
" 'end': 338},\n",
" {'entity': 'r8:arg0|agt',\n",
" 'score': 0.054486226,\n",
" 'index': 88,\n",
" 'word': 'Us',\n",
" 'start': 356,\n",
" 'end': 358},\n",
" {'entity': 'r8:arg0|agt',\n",
" 'score': 0.10709534,\n",
" 'index': 89,\n",
" 'word': 'scientists',\n",
" 'start': 359,\n",
" 'end': 369},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.18592234,\n",
" 'index': 90,\n",
" 'word': 'figure',\n",
" 'start': 370,\n",
" 'end': 376},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.15209025,\n",
" 'index': 91,\n",
" 'word': '##d',\n",
" 'start': 376,\n",
" 'end': 377},\n",
" {'entity': 'r8:arg1|tem',\n",
" 'score': 0.045862645,\n",
" 'index': 94,\n",
" 'word': 'it',\n",
" 'start': 387,\n",
" 'end': 389},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.115167715,\n",
" 'index': 95,\n",
" 'word': 'was',\n",
" 'start': 390,\n",
" 'end': 393},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.04374532,\n",
" 'index': 102,\n",
" 'word': 'common',\n",
" 'start': 421,\n",
" 'end': 427},\n",
" {'entity': 'r8:arg1|tem',\n",
" 'score': 0.040294692,\n",
" 'index': 110,\n",
" 'word': 'one',\n",
" 'start': 454,\n",
" 'end': 457},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.1052284,\n",
" 'index': 111,\n",
" 'word': 'had',\n",
" 'start': 458,\n",
" 'end': 461},\n",
" {'entity': 'r8:arg2|atr',\n",
" 'score': 0.04526653,\n",
" 'index': 112,\n",
" 'word': 'sh',\n",
" 'start': 462,\n",
" 'end': 464},\n",
" {'entity': 'r8:arg2|atr',\n",
" 'score': 0.050974954,\n",
" 'index': 113,\n",
" 'word': '##adow',\n",
" 'start': 464,\n",
" 'end': 468},\n",
" {'entity': 'r8:arg0|agt',\n",
" 'score': 0.04716369,\n",
" 'index': 115,\n",
" 'word': 'that',\n",
" 'start': 470,\n",
" 'end': 474},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.12286918,\n",
" 'index': 116,\n",
" 'word': 'made',\n",
" 'start': 475,\n",
" 'end': 479},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.034002524,\n",
" 'index': 117,\n",
" 'word': 'it',\n",
" 'start': 480,\n",
" 'end': 482},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.085215025,\n",
" 'index': 118,\n",
" 'word': 'look',\n",
" 'start': 483,\n",
" 'end': 487},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.04240514,\n",
" 'index': 119,\n",
" 'word': 'like',\n",
" 'start': 488,\n",
" 'end': 492},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.04686627,\n",
" 'index': 129,\n",
" 'word': 'days',\n",
" 'start': 523,\n",
" 'end': 527},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.040811997,\n",
" 'index': 130,\n",
" 'word': 'later',\n",
" 'start': 528,\n",
" 'end': 533},\n",
" {'entity': 'r10:root',\n",
" 'score': 0.040206842,\n",
" 'index': 132,\n",
" 'word': 'we',\n",
" 'start': 535,\n",
" 'end': 537},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.080649935,\n",
" 'index': 133,\n",
" 'word': 'revealed',\n",
" 'start': 538,\n",
" 'end': 546},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.042861167,\n",
" 'index': 135,\n",
" 'word': 'image',\n",
" 'start': 551,\n",
" 'end': 556},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.042270634,\n",
" 'index': 136,\n",
" 'word': 'for',\n",
" 'start': 557,\n",
" 'end': 560},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.08127655,\n",
" 'index': 139,\n",
" 'word': 'see',\n",
" 'start': 568,\n",
" 'end': 571},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.044283886,\n",
" 'index': 142,\n",
" 'word': 'we',\n",
" 'start': 577,\n",
" 'end': 579},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.08544878,\n",
" 'index': 143,\n",
" 'word': 'made',\n",
" 'start': 580,\n",
" 'end': 584},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.040639073,\n",
" 'index': 144,\n",
" 'word': 'sure',\n",
" 'start': 585,\n",
" 'end': 589},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.0474719,\n",
" 'index': 148,\n",
" 'word': 'it',\n",
" 'start': 603,\n",
" 'end': 605},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.078273796,\n",
" 'index': 149,\n",
" 'word': 'was',\n",
" 'start': 606,\n",
" 'end': 609},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.060446404,\n",
" 'index': 153,\n",
" 'word': 'formation',\n",
" 'start': 622,\n",
" 'end': 631},\n",
" {'entity': 'r9:arg1|pat',\n",
" 'score': 0.030689918,\n",
" 'index': 154,\n",
" 'word': 'that',\n",
" 'start': 632,\n",
" 'end': 636},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.08887379,\n",
" 'index': 156,\n",
" 'word': 'res',\n",
" 'start': 642,\n",
" 'end': 645},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.07661998,\n",
" 'index': 157,\n",
" 'word': '##emble',\n",
" 'start': 645,\n",
" 'end': 650},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.0793471,\n",
" 'index': 158,\n",
" 'word': '##d',\n",
" 'start': 650,\n",
" 'end': 651},\n",
" {'entity': 'r8:arg2|atr',\n",
" 'score': 0.034773763,\n",
" 'index': 161,\n",
" 'word': 'head',\n",
" 'start': 660,\n",
" 'end': 664},\n",
" {'entity': 'r9:arg1|pat',\n",
" 'score': 0.034561444,\n",
" 'index': 166,\n",
" 'word': 'all',\n",
" 'start': 679,\n",
" 'end': 682},\n",
" {'entity': 'r9:arg1|pat',\n",
" 'score': 0.038585175,\n",
" 'index': 168,\n",
" 'word': 'it',\n",
" 'start': 686,\n",
" 'end': 688},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.087780945,\n",
" 'index': 169,\n",
" 'word': 'was',\n",
" 'start': 689,\n",
" 'end': 692},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.0999068,\n",
" 'index': 170,\n",
" 'word': 'formed',\n",
" 'start': 693,\n",
" 'end': 699},\n",
" {'entity': 'r8:arg2|atr',\n",
" 'score': 0.027370188,\n",
" 'index': 171,\n",
" 'word': 'by',\n",
" 'start': 700,\n",
" 'end': 702},\n",
" {'entity': 'r8:arg0|agt',\n",
" 'score': 0.078722976,\n",
" 'index': 176,\n",
" 'word': 'We',\n",
" 'start': 712,\n",
" 'end': 714},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.12920254,\n",
" 'index': 178,\n",
" 'word': 'announced',\n",
" 'start': 720,\n",
" 'end': 729},\n",
" {'entity': 'r8:arg1|pat',\n",
" 'score': 0.07986989,\n",
" 'index': 179,\n",
" 'word': 'it',\n",
" 'start': 730,\n",
" 'end': 732},\n",
" {'entity': 'r8:arg0|agt',\n",
" 'score': 0.048153095,\n",
" 'index': 181,\n",
" 'word': 'we',\n",
" 'start': 741,\n",
" 'end': 743},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.10950655,\n",
" 'index': 182,\n",
" 'word': 'thought',\n",
" 'start': 744,\n",
" 'end': 751},\n",
" {'entity': 'r8:arg1|tem',\n",
" 'score': 0.041486118,\n",
" 'index': 183,\n",
" 'word': 'it',\n",
" 'start': 752,\n",
" 'end': 754},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.05073052,\n",
" 'index': 184,\n",
" 'word': 'would',\n",
" 'start': 755,\n",
" 'end': 760},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.10624526,\n",
" 'index': 185,\n",
" 'word': 'be',\n",
" 'start': 761,\n",
" 'end': 763},\n",
" {'entity': 'r8:arg2|atr',\n",
" 'score': 0.054641366,\n",
" 'index': 188,\n",
" 'word': 'way',\n",
" 'start': 771,\n",
" 'end': 774},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.10240572,\n",
" 'index': 190,\n",
" 'word': 'engage',\n",
" 'start': 778,\n",
" 'end': 784},\n",
" {'entity': 'r9:arg1|pat',\n",
" 'score': 0.044021852,\n",
" 'index': 192,\n",
" 'word': 'public',\n",
" 'start': 789,\n",
" 'end': 795},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.090464294,\n",
" 'index': 200,\n",
" 'word': 'at',\n",
" 'start': 822,\n",
" 'end': 824},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.09293412,\n",
" 'index': 201,\n",
" 'word': '##rra',\n",
" 'start': 824,\n",
" 'end': 827},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.09475694,\n",
" 'index': 202,\n",
" 'word': '##ct',\n",
" 'start': 827,\n",
" 'end': 829},\n",
" {'entity': 'r9:arg1|pat',\n",
" 'score': 0.056167223,\n",
" 'index': 203,\n",
" 'word': 'attention',\n",
" 'start': 830,\n",
" 'end': 839},\n",
" {'entity': 'r8:arg2|atr',\n",
" 'score': 0.03035166,\n",
" 'index': 204,\n",
" 'word': 'to',\n",
" 'start': 840,\n",
" 'end': 842},\n",
" {'entity': 'r9:arg1|pat',\n",
" 'score': 0.032432236,\n",
" 'index': 209,\n",
" 'word': 'it',\n",
" 'start': 854,\n",
" 'end': 856},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.083309464,\n",
" 'index': 210,\n",
" 'word': 'did',\n",
" 'start': 857,\n",
" 'end': 860},\n",
" {'entity': 'r8:arg1|tem',\n",
" 'score': 0.038414363,\n",
" 'index': 213,\n",
" 'word': 'face',\n",
" 'start': 867,\n",
" 'end': 871},\n",
" {'entity': 'r8:arg2|atr',\n",
" 'score': 0.02903995,\n",
" 'index': 216,\n",
" 'word': 'soon',\n",
" 'start': 880,\n",
" 'end': 884},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.07692599,\n",
" 'index': 217,\n",
" 'word': 'became',\n",
" 'start': 885,\n",
" 'end': 891},\n",
" {'entity': 'r8:arg2|atr',\n",
" 'score': 0.051249504,\n",
" 'index': 219,\n",
" 'word': 'pop',\n",
" 'start': 894,\n",
" 'end': 897},\n",
" {'entity': 'r8:arg2|atr',\n",
" 'score': 0.04843377,\n",
" 'index': 220,\n",
" 'word': 'i',\n",
" 'start': 898,\n",
" 'end': 899},\n",
" {'entity': 'r8:arg2|atr',\n",
" 'score': 0.04545699,\n",
" 'index': 221,\n",
" 'word': '##con',\n",
" 'start': 899,\n",
" 'end': 902},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.08494211,\n",
" 'index': 223,\n",
" 'word': 'shot',\n",
" 'start': 904,\n",
" 'end': 908},\n",
" {'entity': 'r8:arg2|atr',\n",
" 'score': 0.03157478,\n",
" 'index': 224,\n",
" 'word': 'in',\n",
" 'start': 909,\n",
" 'end': 911},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.1036089,\n",
" 'index': 227,\n",
" 'word': 'appeared',\n",
" 'start': 920,\n",
" 'end': 928},\n",
" {'entity': 'r8:arg2|atr',\n",
" 'score': 0.03158019,\n",
" 'index': 228,\n",
" 'word': 'in',\n",
" 'start': 929,\n",
" 'end': 931},\n",
" {'entity': 'r8:argM|tmp',\n",
" 'score': 0.091016516,\n",
" 'index': 247,\n",
" 'word': 'for',\n",
" 'start': 1009,\n",
" 'end': 1012},\n",
" {'entity': 'r8:arg1|tem',\n",
" 'score': 0.06676676,\n",
" 'index': 252,\n",
" 'word': 'people',\n",
" 'start': 1028,\n",
" 'end': 1034},\n",
" {'entity': 'r8:root',\n",
" 'score': 0.20488475,\n",
" 'index': 253,\n",
" 'word': 'thought',\n",
" 'start': 1035,\n",
" 'end': 1042},\n",
" {'entity': 'r7:arg1|tem',\n",
" 'score': 0.36603826,\n",
" 'index': 256,\n",
" 'word': 'land',\n",
" 'start': 1055,\n",
" 'end': 1059},\n",
" {'entity': 'r7:arg1|tem',\n",
" 'score': 0.31067976,\n",
" 'index': 257,\n",
" 'word': '##form',\n",
" 'start': 1059,\n",
" 'end': 1063},\n",
" {'entity': 'r7:root',\n",
" 'score': 0.48778686,\n",
" 'index': 258,\n",
" 'word': 'was',\n",
" 'start': 1064,\n",
" 'end': 1067},\n",
" {'entity': 'r7:arg2|atr',\n",
" 'score': 0.13593948,\n",
" 'index': 259,\n",
" 'word': 'evidence',\n",
" 'start': 1068,\n",
" 'end': 1076},\n",
" {'entity': 'r6:arg0|agt',\n",
" 'score': 0.21349978,\n",
" 'index': 268,\n",
" 'word': 'scientists',\n",
" 'start': 1106,\n",
" 'end': 1116},\n",
" {'entity': 'r7:root',\n",
" 'score': 0.48215854,\n",
" 'index': 269,\n",
" 'word': 'wanted',\n",
" 'start': 1117,\n",
" 'end': 1123},\n",
" {'entity': 'r7:root',\n",
" 'score': 0.66229856,\n",
" 'index': 271,\n",
" 'word': 'hide',\n",
" 'start': 1127,\n",
" 'end': 1131},\n",
" {'entity': 'r7:arg1|pat',\n",
" 'score': 0.34266004,\n",
" 'index': 272,\n",
" 'word': 'it',\n",
" 'start': 1132,\n",
" 'end': 1134},\n",
" {'entity': 'r6:arg0|agt',\n",
" 'score': 0.27256963,\n",
" 'index': 278,\n",
" 'word': 'defender',\n",
" 'start': 1152,\n",
" 'end': 1160},\n",
" {'entity': 'r5:root',\n",
" 'score': 0.4016244,\n",
" 'index': 286,\n",
" 'word': 'was',\n",
" 'start': 1192,\n",
" 'end': 1195},\n",
" {'entity': 'r6:arg0|agt',\n",
" 'score': 0.3085603,\n",
" 'index': 293,\n",
" 'word': 'We',\n",
" 'start': 1226,\n",
" 'end': 1228},\n",
" {'entity': 'r7:root',\n",
" 'score': 0.68402517,\n",
" 'index': 294,\n",
" 'word': 'decided',\n",
" 'start': 1229,\n",
" 'end': 1236},\n",
" {'entity': 'r7:root',\n",
" 'score': 0.6312899,\n",
" 'index': 296,\n",
" 'word': 'take',\n",
" 'start': 1240,\n",
" 'end': 1244},\n",
" {'entity': 'r7:arg1|pat',\n",
" 'score': 0.2447523,\n",
" 'index': 298,\n",
" 'word': 'shot',\n",
" 'start': 1253,\n",
" 'end': 1257},\n",
" {'entity': 'r5:argM|fin',\n",
" 'score': 0.21691667,\n",
" 'index': 300,\n",
" 'word': 'to',\n",
" 'start': 1263,\n",
" 'end': 1265},\n",
" {'entity': 'r8:arg1|tem',\n",
" 'score': 0.0507591,\n",
" 'index': 303,\n",
" 'word': 'we',\n",
" 'start': 1276,\n",
" 'end': 1278},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.110183075,\n",
" 'index': 304,\n",
" 'word': 'were',\n",
" 'start': 1279,\n",
" 'end': 1283},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.0827311,\n",
" 'index': 305,\n",
" 'word': '##n',\n",
" 'start': 1283,\n",
" 'end': 1284},\n",
" {'entity': 'r8:arg2|atr',\n",
" 'score': 0.07966802,\n",
" 'index': 308,\n",
" 'word': 'wrong',\n",
" 'start': 1287,\n",
" 'end': 1292},\n",
" {'entity': 'r8:argM|tmp',\n",
" 'score': 0.047043335,\n",
" 'index': 310,\n",
" 'word': 'on',\n",
" 'start': 1294,\n",
" 'end': 1296},\n",
" {'entity': 'r8:arg0|agt',\n",
" 'score': 0.037761096,\n",
" 'index': 316,\n",
" 'word': 'Michael',\n",
" 'start': 1312,\n",
" 'end': 1319},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.11788476,\n",
" 'index': 326,\n",
" 'word': 'took',\n",
" 'start': 1359,\n",
" 'end': 1363},\n",
" {'entity': 'r8:arg1|pat',\n",
" 'score': 0.061609197,\n",
" 'index': 328,\n",
" 'word': 'picture',\n",
" 'start': 1366,\n",
" 'end': 1373},\n",
" {'entity': 'r8:arg1|tem',\n",
" 'score': 0.04639362,\n",
" 'index': 329,\n",
" 'word': 'that',\n",
" 'start': 1374,\n",
" 'end': 1378},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.10497463,\n",
" 'index': 330,\n",
" 'word': 'was',\n",
" 'start': 1379,\n",
" 'end': 1382},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.037833076,\n",
" 'index': 332,\n",
" 'word': 'times',\n",
" 'start': 1387,\n",
" 'end': 1392},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.04543908,\n",
" 'index': 333,\n",
" 'word': 'sharp',\n",
" 'start': 1393,\n",
" 'end': 1398},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.043668192,\n",
" 'index': 334,\n",
" 'word': '##er',\n",
" 'start': 1398,\n",
" 'end': 1400},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.06038779,\n",
" 'index': 339,\n",
" 'word': 'photos',\n",
" 'start': 1426,\n",
" 'end': 1432},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.08361747,\n",
" 'index': 341,\n",
" 'word': 'reveal',\n",
" 'start': 1434,\n",
" 'end': 1440},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.0819967,\n",
" 'index': 342,\n",
" 'word': '##ing',\n",
" 'start': 1440,\n",
" 'end': 1443},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.050429754,\n",
" 'index': 345,\n",
" 'word': 'land',\n",
" 'start': 1454,\n",
" 'end': 1458},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.055836514,\n",
" 'index': 346,\n",
" 'word': '##form',\n",
" 'start': 1458,\n",
" 'end': 1462},\n",
" {'entity': 'r10:root',\n",
" 'score': 0.030725654,\n",
" 'index': 348,\n",
" 'word': 'which',\n",
" 'start': 1464,\n",
" 'end': 1469},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.07811248,\n",
" 'index': 349,\n",
" 'word': 'meant',\n",
" 'start': 1470,\n",
" 'end': 1475},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.056091927,\n",
" 'index': 352,\n",
" 'word': 'monument',\n",
" 'start': 1485,\n",
" 'end': 1493},\n",
" {'entity': 'r10:root',\n",
" 'score': 0.037297357,\n",
" 'index': 357,\n",
" 'word': 'picture',\n",
" 'start': 1505,\n",
" 'end': 1512},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.08379,\n",
" 'index': 358,\n",
" 'word': 'wasn',\n",
" 'start': 1513,\n",
" 'end': 1517},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.046028677,\n",
" 'index': 362,\n",
" 'word': 'clear',\n",
" 'start': 1525,\n",
" 'end': 1530},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.030750155,\n",
" 'index': 363,\n",
" 'word': 'at',\n",
" 'start': 1531,\n",
" 'end': 1533},\n",
" {'entity': 'r8:arg1|tem',\n",
" 'score': 0.029666642,\n",
" 'index': 366,\n",
" 'word': 'which',\n",
" 'start': 1539,\n",
" 'end': 1544},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.07766737,\n",
" 'index': 368,\n",
" 'word': 'mean',\n",
" 'start': 1551,\n",
" 'end': 1555},\n",
" {'entity': 'r9:arg1|pat',\n",
" 'score': 0.037586644,\n",
" 'index': 370,\n",
" 'word': 'marking',\n",
" 'start': 1562,\n",
" 'end': 1569},\n",
" {'entity': 'r9:arg1|pat',\n",
" 'score': 0.039157256,\n",
" 'index': 371,\n",
" 'word': '##s',\n",
" 'start': 1569,\n",
" 'end': 1570},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.06878179,\n",
" 'index': 372,\n",
" 'word': 'were',\n",
" 'start': 1571,\n",
" 'end': 1575},\n",
" {'entity': 'r10:root',\n",
" 'score': 0.06736001,\n",
" 'index': 373,\n",
" 'word': 'hidden',\n",
" 'start': 1576,\n",
" 'end': 1582},\n",
" {'entity': 'r8:arg2|atr',\n",
" 'score': 0.030598668,\n",
" 'index': 374,\n",
" 'word': 'by',\n",
" 'start': 1583,\n",
" 'end': 1585},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.04365823,\n",
" 'index': 381,\n",
" 'word': 'ye',\n",
" 'start': 1601,\n",
" 'end': 1603},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.04754055,\n",
" 'index': 382,\n",
" 'word': '##s',\n",
" 'start': 1603,\n",
" 'end': 1604},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.04762847,\n",
" 'index': 384,\n",
" 'word': 'rum',\n",
" 'start': 1610,\n",
" 'end': 1613},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.038992863,\n",
" 'index': 385,\n",
" 'word': '##or',\n",
" 'start': 1613,\n",
" 'end': 1615},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.06939269,\n",
" 'index': 386,\n",
" 'word': 'started',\n",
" 'start': 1616,\n",
" 'end': 1623},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.08153154,\n",
" 'index': 390,\n",
" 'word': 'prove',\n",
" 'start': 1632,\n",
" 'end': 1637},\n",
" {'entity': 'r9:arg1|pat',\n",
" 'score': 0.042257246,\n",
" 'index': 391,\n",
" 'word': 'them',\n",
" 'start': 1638,\n",
" 'end': 1642},\n",
" {'entity': 'r8:arg2|atr',\n",
" 'score': 0.04214874,\n",
" 'index': 392,\n",
" 'word': 'wrong',\n",
" 'start': 1643,\n",
" 'end': 1648},\n",
" {'entity': 'r8:argM|tmp',\n",
" 'score': 0.036950577,\n",
" 'index': 393,\n",
" 'word': 'on',\n",
" 'start': 1649,\n",
" 'end': 1651},\n",
" {'entity': 'r8:arg0|agt',\n",
" 'score': 0.033291806,\n",
" 'index': 398,\n",
" 'word': 'we',\n",
" 'start': 1666,\n",
" 'end': 1668},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.0738949,\n",
" 'index': 399,\n",
" 'word': 'decided',\n",
" 'start': 1669,\n",
" 'end': 1676},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.08140796,\n",
" 'index': 401,\n",
" 'word': 'take',\n",
" 'start': 1680,\n",
" 'end': 1684},\n",
" {'entity': 'r9:arg1|pat',\n",
" 'score': 0.04987211,\n",
" 'index': 403,\n",
" 'word': 'picture',\n",
" 'start': 1693,\n",
" 'end': 1700},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.06430703,\n",
" 'index': 405,\n",
" 'word': 'making',\n",
" 'start': 1702,\n",
" 'end': 1708},\n",
" {'entity': 'r8:arg2|atr',\n",
" 'score': 0.038376,\n",
" 'index': 406,\n",
" 'word': 'sure',\n",
" 'start': 1709,\n",
" 'end': 1713},\n",
" {'entity': 'r9:arg1|pat',\n",
" 'score': 0.03541435,\n",
" 'index': 407,\n",
" 'word': 'it',\n",
" 'start': 1714,\n",
" 'end': 1716},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.074262,\n",
" 'index': 408,\n",
" 'word': 'was',\n",
" 'start': 1717,\n",
" 'end': 1720},\n",
" {'entity': 'r8:arg2|atr',\n",
" 'score': 0.044221986,\n",
" 'index': 410,\n",
" 'word': 'cloud',\n",
" 'start': 1723,\n",
" 'end': 1728},\n",
" {'entity': 'r8:arg2|atr',\n",
" 'score': 0.039963495,\n",
" 'index': 413,\n",
" 'word': 'day',\n",
" 'start': 1740,\n",
" 'end': 1743},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.03479836,\n",
" 'index': 415,\n",
" 'word': 'Mali',\n",
" 'start': 1745,\n",
" 'end': 1749},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.032437824,\n",
" 'index': 419,\n",
" 'word': 'team',\n",
" 'start': 1753,\n",
" 'end': 1757},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.08305774,\n",
" 'index': 420,\n",
" 'word': 'captured',\n",
" 'start': 1758,\n",
" 'end': 1766},\n",
" {'entity': 'r9:arg1|pat',\n",
" 'score': 0.04045077,\n",
" 'index': 424,\n",
" 'word': 'photo',\n",
" 'start': 1778,\n",
" 'end': 1783},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.066850476,\n",
" 'index': 425,\n",
" 'word': 'using',\n",
" 'start': 1784,\n",
" 'end': 1789},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.03912363,\n",
" 'index': 427,\n",
" 'word': 'camera',\n",
" 'start': 1794,\n",
" 'end': 1800},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.0366046,\n",
" 'index': 434,\n",
" 'word': 'With',\n",
" 'start': 1832,\n",
" 'end': 1836},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.034650415,\n",
" 'index': 437,\n",
" 'word': 'you',\n",
" 'start': 1849,\n",
" 'end': 1852},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.07881917,\n",
" 'index': 439,\n",
" 'word': 'disc',\n",
" 'start': 1857,\n",
" 'end': 1861},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.08386986,\n",
" 'index': 440,\n",
" 'word': '##ern',\n",
" 'start': 1861,\n",
" 'end': 1864},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.035086773,\n",
" 'index': 441,\n",
" 'word': 'things',\n",
" 'start': 1865,\n",
" 'end': 1871},\n",
" {'entity': 'r10:root',\n",
" 'score': 0.033257537,\n",
" 'index': 442,\n",
" 'word': 'in',\n",
" 'start': 1872,\n",
" 'end': 1874},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.035879184,\n",
" 'index': 455,\n",
" 'word': 'which',\n",
" 'start': 1927,\n",
" 'end': 1932},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.0820557,\n",
" 'index': 456,\n",
" 'word': 'means',\n",
" 'start': 1933,\n",
" 'end': 1938},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.045568023,\n",
" 'index': 458,\n",
" 'word': 'there',\n",
" 'start': 1942,\n",
" 'end': 1947},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.07430672,\n",
" 'index': 459,\n",
" 'word': 'were',\n",
" 'start': 1948,\n",
" 'end': 1952},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.048614994,\n",
" 'index': 461,\n",
" 'word': 'signs',\n",
" 'start': 1957,\n",
" 'end': 1962},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.040848035,\n",
" 'index': 465,\n",
" 'word': 'you',\n",
" 'start': 1972,\n",
" 'end': 1975},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.06783971,\n",
" 'index': 466,\n",
" 'word': 'could',\n",
" 'start': 1976,\n",
" 'end': 1981},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.043203566,\n",
" 'index': 467,\n",
" 'word': 'easily',\n",
" 'start': 1982,\n",
" 'end': 1988},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.07753727,\n",
" 'index': 468,\n",
" 'word': 'see',\n",
" 'start': 1989,\n",
" 'end': 1992},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.03982749,\n",
" 'index': 469,\n",
" 'word': 'what',\n",
" 'start': 1993,\n",
" 'end': 1997},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.050202448,\n",
" 'index': 470,\n",
" 'word': 'they',\n",
" 'start': 1998,\n",
" 'end': 2002},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.07350303,\n",
" 'index': 471,\n",
" 'word': 'were',\n",
" 'start': 2003,\n",
" 'end': 2007},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.041813992,\n",
" 'index': 473,\n",
" 'word': 'What',\n",
" 'start': 2009,\n",
" 'end': 2013},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.043242317,\n",
" 'index': 475,\n",
" 'word': 'picture',\n",
" 'start': 2018,\n",
" 'end': 2025},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.071942516,\n",
" 'index': 476,\n",
" 'word': 'showed',\n",
" 'start': 2026,\n",
" 'end': 2032},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.065713726,\n",
" 'index': 477,\n",
" 'word': 'was',\n",
" 'start': 2033,\n",
" 'end': 2036},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.047005344,\n",
" 'index': 479,\n",
" 'word': 'but',\n",
" 'start': 2041,\n",
" 'end': 2044},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.053594884,\n",
" 'index': 480,\n",
" 'word': '##te',\n",
" 'start': 2044,\n",
" 'end': 2046},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.038346287,\n",
" 'index': 484,\n",
" 'word': 'which',\n",
" 'start': 2056,\n",
" 'end': 2061},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.071865395,\n",
" 'index': 485,\n",
" 'word': 'are',\n",
" 'start': 2062,\n",
" 'end': 2065},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.052885562,\n",
" 'index': 486,\n",
" 'word': 'land',\n",
" 'start': 2066,\n",
" 'end': 2070},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.058789182,\n",
" 'index': 487,\n",
" 'word': '##form',\n",
" 'start': 2070,\n",
" 'end': 2074},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.05849257,\n",
" 'index': 488,\n",
" 'word': '##s',\n",
" 'start': 2074,\n",
" 'end': 2075},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.05117639,\n",
" 'index': 489,\n",
" 'word': 'common',\n",
" 'start': 2076,\n",
" 'end': 2082},\n",
" {'entity': 'r9:root',\n",
" 'score': 0.051266044,\n",
" 'index': 490,\n",
" 'word': 'around',\n",
" 'start': 2083,\n",
" 'end': 2089}]"
]
},
"execution_count": 77,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from transformers import pipeline\n",
"tokenizer = AutoTokenizer.from_pretrained(\"mbruton/spa_enpt_mBERT\")\n",
"model = AutoModelForTokenClassification.from_pretrained(\"mbruton/spa_enpt_mBERT\")\n",
"classifier = pipeline(\"ner\", model=model, tokenizer=tokenizer)\n",
"classifier(text)"
]
},
{
"cell_type": "code",
"execution_count": 78,
"id": "a5c41e30-7978-45e4-9eba-c3eaf8fcd77d",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"entity\n",
"r0:arg1|tem 1\n",
"r0:root 1\n",
"r10:root 5\n",
"r1:arg1|pat 1\n",
"r1:arg1|tem 1\n",
"r1:arg2|atr 1\n",
"r1:root 1\n",
"r2:root 1\n",
"r4:arg1|tem 1\n",
"r4:arg2|atr 1\n",
"r5:arg1|tem 1\n",
"r5:argM|fin 1\n",
"r5:root 2\n",
"r6:arg0|agt 4\n",
"r6:arg2|atr 1\n",
"r6:root 1\n",
"r7:arg0|agt 1\n",
"r7:arg1|pat 5\n",
"r7:arg1|tem 2\n",
"r7:arg2|atr 1\n",
"r7:root 10\n",
"r8:arg0|agt 8\n",
"r8:arg1|pat 4\n",
"r8:arg1|tem 8\n",
"r8:arg2|atr 18\n",
"r8:argM|tmp 3\n",
"r8:root 3\n",
"r9:arg1|pat 12\n",
"r9:root 102\n",
"dtype: int64\n"
]
},
{
"data": {
"text/plain": [
"entity word \n",
"r0:arg1|tem you 1\n",
"r0:root re 1\n",
"r10:root hidden 1\n",
" in 1\n",
" picture 1\n",
" ..\n",
"r9:root what 1\n",
" which 2\n",
" would 1\n",
" ye 1\n",
" you 2\n",
"Length: 174, dtype: int64"
]
},
"execution_count": 78,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"\n",
"with open(\"10 mbrutonspa_enpt_mBERT.json\", encoding='utf-8') as f:\n",
" data = json.load(f)\n",
"aux=obtener_dataframe(data)\n",
"aux\n",
"print(aux.groupby(['entity']).size())\n",
"aux.groupby(['entity', 'word']).size()"
]
},
{
"cell_type": "markdown",
"id": "6da949b6-cae9-409d-a05d-d93f9f634bfd",
"metadata": {},
"source": [
"## 11 benjamin/wtp-bert-mini"
]
},
{
"cell_type": "code",
"execution_count": 80,
"id": "6b9d327a-5f62-435e-923e-9a51d91c958a",
"metadata": {},
"outputs": [
{
"ename": "ValueError",
"evalue": "The checkpoint you are trying to load has model type `bert-char` but Transformers does not recognize this architecture. This could be because of an issue with the checkpoint, or because your version of Transformers is out of date.",
"output_type": "error",
"traceback": [
"\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[1;31mKeyError\u001b[0m Traceback (most recent call last)",
"File \u001b[1;32m~\\anaconda3\\Lib\\site-packages\\transformers\\models\\auto\\configuration_auto.py:951\u001b[0m, in \u001b[0;36mAutoConfig.from_pretrained\u001b[1;34m(cls, pretrained_model_name_or_path, **kwargs)\u001b[0m\n\u001b[0;32m 950\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m--> 951\u001b[0m config_class \u001b[38;5;241m=\u001b[39m CONFIG_MAPPING[config_dict[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmodel_type\u001b[39m\u001b[38;5;124m\"\u001b[39m]]\n\u001b[0;32m 952\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m:\n",
"File \u001b[1;32m~\\anaconda3\\Lib\\site-packages\\transformers\\models\\auto\\configuration_auto.py:653\u001b[0m, in \u001b[0;36m_LazyConfigMapping.__getitem__\u001b[1;34m(self, key)\u001b[0m\n\u001b[0;32m 652\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m key \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_mapping:\n\u001b[1;32m--> 653\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m(key)\n\u001b[0;32m 654\u001b[0m value \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_mapping[key]\n",
"\u001b[1;31mKeyError\u001b[0m: 'bert-char'",
"\nDuring handling of the above exception, another exception occurred:\n",
"\u001b[1;31mValueError\u001b[0m Traceback (most recent call last)",
"Cell \u001b[1;32mIn[80], line 2\u001b[0m\n\u001b[0;32m 1\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtransformers\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m pipeline\n\u001b[1;32m----> 2\u001b[0m tokenizer \u001b[38;5;241m=\u001b[39m AutoTokenizer\u001b[38;5;241m.\u001b[39mfrom_pretrained(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mbenjamin/wtp-bert-mini\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m 3\u001b[0m model \u001b[38;5;241m=\u001b[39m AutoModelForTokenClassification\u001b[38;5;241m.\u001b[39mfrom_pretrained(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mbenjamin/wtp-bert-mini\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m 4\u001b[0m classifier \u001b[38;5;241m=\u001b[39m pipeline(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mner\u001b[39m\u001b[38;5;124m\"\u001b[39m, model\u001b[38;5;241m=\u001b[39mmodel, tokenizer\u001b[38;5;241m=\u001b[39mtokenizer)\n",
"File \u001b[1;32m~\\anaconda3\\Lib\\site-packages\\transformers\\models\\auto\\tokenization_auto.py:837\u001b[0m, in \u001b[0;36mAutoTokenizer.from_pretrained\u001b[1;34m(cls, pretrained_model_name_or_path, *inputs, **kwargs)\u001b[0m\n\u001b[0;32m 835\u001b[0m config \u001b[38;5;241m=\u001b[39m AutoConfig\u001b[38;5;241m.\u001b[39mfor_model(\u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mconfig_dict)\n\u001b[0;32m 836\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m--> 837\u001b[0m config \u001b[38;5;241m=\u001b[39m AutoConfig\u001b[38;5;241m.\u001b[39mfrom_pretrained(\n\u001b[0;32m 838\u001b[0m pretrained_model_name_or_path, trust_remote_code\u001b[38;5;241m=\u001b[39mtrust_remote_code, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs\n\u001b[0;32m 839\u001b[0m )\n\u001b[0;32m 840\u001b[0m config_tokenizer_class \u001b[38;5;241m=\u001b[39m config\u001b[38;5;241m.\u001b[39mtokenizer_class\n\u001b[0;32m 841\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mhasattr\u001b[39m(config, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mauto_map\u001b[39m\u001b[38;5;124m\"\u001b[39m) \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mAutoTokenizer\u001b[39m\u001b[38;5;124m\"\u001b[39m \u001b[38;5;129;01min\u001b[39;00m config\u001b[38;5;241m.\u001b[39mauto_map:\n",
"File \u001b[1;32m~\\anaconda3\\Lib\\site-packages\\transformers\\models\\auto\\configuration_auto.py:953\u001b[0m, in \u001b[0;36mAutoConfig.from_pretrained\u001b[1;34m(cls, pretrained_model_name_or_path, **kwargs)\u001b[0m\n\u001b[0;32m 951\u001b[0m config_class \u001b[38;5;241m=\u001b[39m CONFIG_MAPPING[config_dict[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmodel_type\u001b[39m\u001b[38;5;124m\"\u001b[39m]]\n\u001b[0;32m 952\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m:\n\u001b[1;32m--> 953\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\n\u001b[0;32m 954\u001b[0m \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mThe checkpoint you are trying to load has model type `\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mconfig_dict[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mmodel_type\u001b[39m\u001b[38;5;124m'\u001b[39m]\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m` \u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m 955\u001b[0m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mbut Transformers does not recognize this architecture. This could be because of an \u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m 956\u001b[0m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124missue with the checkpoint, or because your version of Transformers is out of date.\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m 957\u001b[0m )\n\u001b[0;32m 958\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m config_class\u001b[38;5;241m.\u001b[39mfrom_dict(config_dict, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39munused_kwargs)\n\u001b[0;32m 959\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m 960\u001b[0m \u001b[38;5;66;03m# Fallback: use pattern matching on the string.\u001b[39;00m\n\u001b[0;32m 961\u001b[0m \u001b[38;5;66;03m# We go from longer names to shorter names to catch roberta before bert (for instance)\u001b[39;00m\n",
"\u001b[1;31mValueError\u001b[0m: The checkpoint you are trying to load has model type `bert-char` but Transformers does not recognize this architecture. This could be because of an issue with the checkpoint, or because your version of Transformers is out of date."
]
}
],
"source": [
"from transformers import pipeline\n",
"tokenizer = AutoTokenizer.from_pretrained(\"benjamin/wtp-bert-mini\")\n",
"model = AutoModelForTokenClassification.from_pretrained(\"benjamin/wtp-bert-mini\")\n",
"classifier = pipeline(\"ner\", model=model, tokenizer=tokenizer)\n",
"classifier(text)"
]
},
{
"cell_type": "markdown",
"id": "6b2ab65e-9856-45a9-aa54-c230abfe8d4b",
"metadata": {},
"source": [
"## 12 Babelscapewikineural-multilingual-ner"
]
},
{
"cell_type": "code",
"execution_count": 83,
"id": "d9c12157-5280-4753-b0e6-b1cd7de813bd",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[{'entity': 'B-ORG',\n",
" 'score': 0.9951147,\n",
" 'index': 8,\n",
" 'word': 'NASA',\n",
" 'start': 16,\n",
" 'end': 20},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.6191017,\n",
" 'index': 23,\n",
" 'word': 'Face',\n",
" 'start': 88,\n",
" 'end': 92},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.5708344,\n",
" 'index': 24,\n",
" 'word': 'On',\n",
" 'start': 93,\n",
" 'end': 95},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.5786274,\n",
" 'index': 25,\n",
" 'word': 'Mars',\n",
" 'start': 96,\n",
" 'end': 100},\n",
" {'entity': 'B-LOC',\n",
" 'score': 0.877606,\n",
" 'index': 37,\n",
" 'word': 'Mars',\n",
" 'start': 152,\n",
" 'end': 156},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.9051992,\n",
" 'index': 60,\n",
" 'word': 'Viking',\n",
" 'start': 240,\n",
" 'end': 246},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.9834109,\n",
" 'index': 61,\n",
" 'word': '1',\n",
" 'start': 247,\n",
" 'end': 248},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.48295248,\n",
" 'index': 98,\n",
" 'word': 'Mart',\n",
" 'start': 407,\n",
" 'end': 411},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.47647634,\n",
" 'index': 99,\n",
" 'word': '##ian',\n",
" 'start': 411,\n",
" 'end': 414},\n",
" {'entity': 'B-LOC',\n",
" 'score': 0.97810775,\n",
" 'index': 104,\n",
" 'word': 'C',\n",
" 'start': 435,\n",
" 'end': 436},\n",
" {'entity': 'I-LOC',\n",
" 'score': 0.9512793,\n",
" 'index': 105,\n",
" 'word': '##yd',\n",
" 'start': 436,\n",
" 'end': 438},\n",
" {'entity': 'I-LOC',\n",
" 'score': 0.9480485,\n",
" 'index': 106,\n",
" 'word': '##onia',\n",
" 'start': 438,\n",
" 'end': 442},\n",
" {'entity': 'B-LOC',\n",
" 'score': 0.4522933,\n",
" 'index': 121,\n",
" 'word': 'Egypt',\n",
" 'start': 496,\n",
" 'end': 501},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.43941417,\n",
" 'index': 122,\n",
" 'word': '##ion',\n",
" 'start': 501,\n",
" 'end': 504},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.33125964,\n",
" 'index': 123,\n",
" 'word': 'Ph',\n",
" 'start': 505,\n",
" 'end': 507},\n",
" {'entity': 'I-PER',\n",
" 'score': 0.2833639,\n",
" 'index': 124,\n",
" 'word': '##ara',\n",
" 'start': 507,\n",
" 'end': 510},\n",
" {'entity': 'B-ORG',\n",
" 'score': 0.9865861,\n",
" 'index': 194,\n",
" 'word': 'NASA',\n",
" 'start': 801,\n",
" 'end': 805},\n",
" {'entity': 'B-LOC',\n",
" 'score': 0.814626,\n",
" 'index': 205,\n",
" 'word': 'Mars',\n",
" 'start': 843,\n",
" 'end': 847},\n",
" {'entity': 'B-LOC',\n",
" 'score': 0.7951029,\n",
" 'index': 215,\n",
" 'word': 'Mars',\n",
" 'start': 875,\n",
" 'end': 879},\n",
" {'entity': 'B-LOC',\n",
" 'score': 0.87542975,\n",
" 'index': 263,\n",
" 'word': 'Mars',\n",
" 'start': 1088,\n",
" 'end': 1092},\n",
" {'entity': 'B-ORG',\n",
" 'score': 0.99368435,\n",
" 'index': 282,\n",
" 'word': 'NASA',\n",
" 'start': 1169,\n",
" 'end': 1173},\n",
" {'entity': 'B-LOC',\n",
" 'score': 0.7881979,\n",
" 'index': 291,\n",
" 'word': 'Mars',\n",
" 'start': 1220,\n",
" 'end': 1224},\n",
" {'entity': 'B-PER',\n",
" 'score': 0.9987973,\n",
" 'index': 316,\n",
" 'word': 'Michael',\n",
" 'start': 1312,\n",
" 'end': 1319},\n",
" {'entity': 'I-PER',\n",
" 'score': 0.99919313,\n",
" 'index': 317,\n",
" 'word': 'Mali',\n",
" 'start': 1320,\n",
" 'end': 1324},\n",
" {'entity': 'I-PER',\n",
" 'score': 0.99887544,\n",
" 'index': 318,\n",
" 'word': '##n',\n",
" 'start': 1324,\n",
" 'end': 1325},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.8786556,\n",
" 'index': 321,\n",
" 'word': 'Mars',\n",
" 'start': 1334,\n",
" 'end': 1338},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.9564052,\n",
" 'index': 322,\n",
" 'word': 'Or',\n",
" 'start': 1339,\n",
" 'end': 1341},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.96700704,\n",
" 'index': 323,\n",
" 'word': '##biter',\n",
" 'start': 1341,\n",
" 'end': 1346},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.89913684,\n",
" 'index': 338,\n",
" 'word': 'Viking',\n",
" 'start': 1419,\n",
" 'end': 1425},\n",
" {'entity': 'B-PER',\n",
" 'score': 0.97243416,\n",
" 'index': 415,\n",
" 'word': 'Mali',\n",
" 'start': 1745,\n",
" 'end': 1749},\n",
" {'entity': 'I-PER',\n",
" 'score': 0.963992,\n",
" 'index': 416,\n",
" 'word': '##n',\n",
" 'start': 1749,\n",
" 'end': 1750},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.7103455,\n",
" 'index': 492,\n",
" 'word': 'American',\n",
" 'start': 2094,\n",
" 'end': 2102},\n",
" {'entity': 'I-MISC',\n",
" 'score': 0.5068105,\n",
" 'index': 493,\n",
" 'word': 'West',\n",
" 'start': 2103,\n",
" 'end': 2107}]"
]
},
"execution_count": 83,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from transformers import pipeline\n",
"tokenizer = AutoTokenizer.from_pretrained(\"Babelscape/wikineural-multilingual-ner\")\n",
"model = AutoModelForTokenClassification.from_pretrained(\"Babelscape/wikineural-multilingual-ner\")\n",
"classifier = pipeline(\"ner\", model=model, tokenizer=tokenizer)\n",
"classifier(text)"
]
},
{
"cell_type": "code",
"execution_count": 84,
"id": "60a70391-65f5-4c73-9cd3-553d635386d9",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"entity\n",
"B-LOC 7\n",
"B-ORG 3\n",
"B-PER 2\n",
"I-LOC 2\n",
"I-MISC 15\n",
"I-PER 4\n",
"dtype: int64\n"
]
},
{
"data": {
"text/plain": [
"entity word \n",
"B-LOC C 1\n",
" Egypt 1\n",
" Mars 5\n",
"B-ORG NASA 3\n",
"B-PER Mali 1\n",
" Michael 1\n",
"I-LOC ##onia 1\n",
" ##yd 1\n",
"I-MISC ##biter 1\n",
" ##ian 1\n",
" ##ion 1\n",
" 1 1\n",
" American 1\n",
" Face 1\n",
" Mars 2\n",
" Mart 1\n",
" On 1\n",
" Or 1\n",
" Ph 1\n",
" Viking 2\n",
" West 1\n",
"I-PER ##ara 1\n",
" ##n 2\n",
" Mali 1\n",
"dtype: int64"
]
},
"execution_count": 84,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"\n",
"with open(\"12 Babelscapewikineural-multilingual-ner.json\", encoding='utf-8') as f:\n",
" data = json.load(f)\n",
"aux=obtener_dataframe(data)\n",
"aux\n",
"print(aux.groupby(['entity']).size())\n",
"aux.groupby(['entity', 'word']).size()"
]
},
{
"cell_type": "markdown",
"id": "b152a918-a4b1-4f0d-aa5a-48158eb9bee8",
"metadata": {},
"source": [
"## 13 julian-schelb/roberta-ner-multilingual"
]
},
{
"cell_type": "code",
"execution_count": 88,
"id": "982634a2-a7f4-4c9d-92d4-f3b34ba89931",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[{'entity': 'B-ORG',\n",
" 'score': 0.8837392,\n",
" 'index': 8,\n",
" 'word': '▁NASA',\n",
" 'start': 16,\n",
" 'end': 20},\n",
" {'entity': 'B-ORG',\n",
" 'score': 0.38925776,\n",
" 'index': 97,\n",
" 'word': '▁Marti',\n",
" 'start': 407,\n",
" 'end': 412},\n",
" {'entity': 'B-LOC',\n",
" 'score': 0.7155649,\n",
" 'index': 103,\n",
" 'word': '▁Cy',\n",
" 'start': 435,\n",
" 'end': 437},\n",
" {'entity': 'B-LOC',\n",
" 'score': 0.64458394,\n",
" 'index': 104,\n",
" 'word': 'do',\n",
" 'start': 437,\n",
" 'end': 439},\n",
" {'entity': 'B-LOC',\n",
" 'score': 0.623109,\n",
" 'index': 105,\n",
" 'word': 'nia',\n",
" 'start': 439,\n",
" 'end': 442},\n",
" {'entity': 'B-ORG',\n",
" 'score': 0.34994408,\n",
" 'index': 119,\n",
" 'word': '▁Egypt',\n",
" 'start': 496,\n",
" 'end': 501},\n",
" {'entity': 'B-ORG',\n",
" 'score': 0.32543704,\n",
" 'index': 120,\n",
" 'word': 'ion',\n",
" 'start': 501,\n",
" 'end': 504},\n",
" {'entity': 'I-PER',\n",
" 'score': 0.4674562,\n",
" 'index': 121,\n",
" 'word': '▁Phar',\n",
" 'start': 505,\n",
" 'end': 509},\n",
" {'entity': 'I-PER',\n",
" 'score': 0.53341544,\n",
" 'index': 122,\n",
" 'word': 'a',\n",
" 'start': 509,\n",
" 'end': 510},\n",
" {'entity': 'I-PER',\n",
" 'score': 0.50114465,\n",
" 'index': 123,\n",
" 'word': 'oh',\n",
" 'start': 510,\n",
" 'end': 512},\n",
" {'entity': 'B-ORG',\n",
" 'score': 0.5386203,\n",
" 'index': 193,\n",
" 'word': '▁NASA',\n",
" 'start': 801,\n",
" 'end': 805},\n",
" {'entity': 'B-ORG',\n",
" 'score': 0.6421071,\n",
" 'index': 285,\n",
" 'word': '▁NASA',\n",
" 'start': 1169,\n",
" 'end': 1173},\n",
" {'entity': 'B-PER',\n",
" 'score': 0.521761,\n",
" 'index': 319,\n",
" 'word': '▁Michael',\n",
" 'start': 1312,\n",
" 'end': 1319},\n",
" {'entity': 'I-PER',\n",
" 'score': 0.66133285,\n",
" 'index': 320,\n",
" 'word': '▁Malin',\n",
" 'start': 1320,\n",
" 'end': 1325},\n",
" {'entity': 'B-ORG',\n",
" 'score': 0.87530375,\n",
" 'index': 323,\n",
" 'word': '▁Mars',\n",
" 'start': 1334,\n",
" 'end': 1338},\n",
" {'entity': 'I-ORG',\n",
" 'score': 0.775388,\n",
" 'index': 324,\n",
" 'word': '▁Or',\n",
" 'start': 1339,\n",
" 'end': 1341},\n",
" {'entity': 'I-ORG',\n",
" 'score': 0.74558014,\n",
" 'index': 325,\n",
" 'word': 'bit',\n",
" 'start': 1341,\n",
" 'end': 1344},\n",
" {'entity': 'I-ORG',\n",
" 'score': 0.77897793,\n",
" 'index': 326,\n",
" 'word': 'er',\n",
" 'start': 1344,\n",
" 'end': 1346},\n",
" {'entity': 'B-LOC',\n",
" 'score': 0.781982,\n",
" 'index': 491,\n",
" 'word': '▁American',\n",
" 'start': 2094,\n",
" 'end': 2102},\n",
" {'entity': 'I-LOC',\n",
" 'score': 0.5271412,\n",
" 'index': 492,\n",
" 'word': '▁West',\n",
" 'start': 2103,\n",
" 'end': 2107}]"
]
},
"execution_count": 88,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from transformers import pipeline\n",
"tokenizer = AutoTokenizer.from_pretrained(\"julian-schelb/roberta-ner-multilingual\")\n",
"model = AutoModelForTokenClassification.from_pretrained(\"julian-schelb/roberta-ner-multilingual\")\n",
"classifier = pipeline(\"ner\", model=model, tokenizer=tokenizer)\n",
"classifier(text)"
]
},
{
"cell_type": "code",
"execution_count": 89,
"id": "1836b86a-e0c8-42c4-b199-b45b3e4c6235",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"entity\n",
"B-LOC 4\n",
"B-ORG 7\n",
"B-PER 1\n",
"I-LOC 1\n",
"I-ORG 3\n",
"I-PER 4\n",
"dtype: int64\n"
]
},
{
"data": {
"text/plain": [
"entity word \n",
"B-LOC do 1\n",
" nia 1\n",
" ▁American 1\n",
" ▁Cy 1\n",
"B-ORG ion 1\n",
" ▁Egypt 1\n",
" ▁Mars 1\n",
" ▁Marti 1\n",
" ▁NASA 3\n",
"B-PER ▁Michael 1\n",
"I-LOC ▁West 1\n",
"I-ORG bit 1\n",
" er 1\n",
" ▁Or 1\n",
"I-PER a 1\n",
" oh 1\n",
" ▁Malin 1\n",
" ▁Phar 1\n",
"dtype: int64"
]
},
"execution_count": 89,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"with open(\"13 julian-schelbroberta-ner-multilingual.json\", encoding='utf-8') as f:\n",
" data = json.load(f)\n",
"aux=obtener_dataframe(data)\n",
"aux\n",
"print(aux.groupby(['entity']).size())\n",
"aux.groupby(['entity', 'word']).size()"
]
},
{
"cell_type": "markdown",
"id": "16569163-0440-4747-8af0-167e67d576a5",
"metadata": {},
"source": [
"## 14 FacebookAI/xlm-roberta-large-finetuned-conll03-german"
]
},
{
"cell_type": "code",
"execution_count": 91,
"id": "3792dc25-ec55-4d6d-822c-e921c9129cd0",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Some weights of the model checkpoint at FacebookAI/xlm-roberta-large-finetuned-conll03-english were not used when initializing XLMRobertaForTokenClassification: ['roberta.pooler.dense.bias', 'roberta.pooler.dense.weight']\n",
"- This IS expected if you are initializing XLMRobertaForTokenClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
"- This IS NOT expected if you are initializing XLMRobertaForTokenClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[{'entity': 'I-ORG', 'score': 0.9999913, 'index': 8, 'word': '▁NASA', 'start': 16, 'end': 20}, {'entity': 'I-MISC', 'score': 0.9999641, 'index': 23, 'word': '▁Face', 'start': 88, 'end': 92}, {'entity': 'I-MISC', 'score': 0.99989665, 'index': 24, 'word': '▁On', 'start': 93, 'end': 95}, {'entity': 'I-MISC', 'score': 0.97350365, 'index': 25, 'word': '▁Mars', 'start': 96, 'end': 100}, {'entity': 'I-LOC', 'score': 0.9999362, 'index': 36, 'word': '▁Mars', 'start': 152, 'end': 156}, {'entity': 'I-MISC', 'score': 0.9992086, 'index': 58, 'word': '▁Viking', 'start': 240, 'end': 246}, {'entity': 'I-MISC', 'score': 0.9989502, 'index': 59, 'word': '▁1', 'start': 247, 'end': 248}, {'entity': 'I-MISC', 'score': 0.999977, 'index': 97, 'word': '▁Marti', 'start': 407, 'end': 412}, {'entity': 'I-MISC', 'score': 0.99619055, 'index': 98, 'word': 'an', 'start': 412, 'end': 414}, {'entity': 'I-LOC', 'score': 0.9999354, 'index': 103, 'word': '▁Cy', 'start': 435, 'end': 437}, {'entity': 'I-LOC', 'score': 0.99994576, 'index': 104, 'word': 'do', 'start': 437, 'end': 439}, {'entity': 'I-LOC', 'score': 0.99992585, 'index': 105, 'word': 'nia', 'start': 439, 'end': 442}, {'entity': 'I-MISC', 'score': 0.9999789, 'index': 119, 'word': '▁Egypt', 'start': 496, 'end': 501}, {'entity': 'I-MISC', 'score': 0.9614088, 'index': 120, 'word': 'ion', 'start': 501, 'end': 504}, {'entity': 'I-ORG', 'score': 0.99997246, 'index': 193, 'word': '▁NASA', 'start': 801, 'end': 805}, {'entity': 'I-LOC', 'score': 0.99979633, 'index': 205, 'word': '▁Mars', 'start': 843, 'end': 847}, {'entity': 'I-LOC', 'score': 0.9998061, 'index': 215, 'word': '▁Mars', 'start': 875, 'end': 879}, {'entity': 'I-LOC', 'score': 0.99984956, 'index': 264, 'word': '▁Mars', 'start': 1088, 'end': 1092}, {'entity': 'I-ORG', 'score': 0.99996305, 'index': 285, 'word': '▁NASA', 'start': 1169, 'end': 1173}, {'entity': 'I-LOC', 'score': 0.9998203, 'index': 295, 'word': '▁Mars', 'start': 1220, 'end': 1224}, {'entity': 'I-PER', 'score': 0.9999932, 'index': 319, 'word': '▁Michael', 'start': 1312, 'end': 1319}, {'entity': 'I-PER', 'score': 0.99999106, 'index': 320, 'word': '▁Malin', 'start': 1320, 'end': 1325}, {'entity': 'I-MISC', 'score': 0.94105357, 'index': 323, 'word': '▁Mars', 'start': 1334, 'end': 1338}, {'entity': 'I-MISC', 'score': 0.9839579, 'index': 324, 'word': '▁Or', 'start': 1339, 'end': 1341}, {'entity': 'I-MISC', 'score': 0.9913346, 'index': 325, 'word': 'bit', 'start': 1341, 'end': 1344}, {'entity': 'I-MISC', 'score': 0.9759228, 'index': 326, 'word': 'er', 'start': 1344, 'end': 1346}, {'entity': 'I-MISC', 'score': 0.999749, 'index': 341, 'word': '▁Viking', 'start': 1419, 'end': 1425}, {'entity': 'I-PER', 'score': 0.9999914, 'index': 416, 'word': '▁Malin', 'start': 1745, 'end': 1750}, {'entity': 'I-MISC', 'score': 0.92417294, 'index': 491, 'word': '▁American', 'start': 2094, 'end': 2102}, {'entity': 'I-LOC', 'score': 0.99954396, 'index': 492, 'word': '▁West', 'start': 2103, 'end': 2107}]\n"
]
}
],
"source": [
"import json\n",
"from transformers import pipeline\n",
"tokenizer = AutoTokenizer.from_pretrained(\"FacebookAI/xlm-roberta-large-finetuned-conll03-german\")\n",
"model = AutoModelForTokenClassification.from_pretrained(\"FacebookAI/xlm-roberta-large-finetuned-conll03-english\")\n",
"classifier = pipeline(\"ner\", model=model, tokenizer=tokenizer)\n",
"salida=classifier(text)\n",
"print(salida)\n"
]
},
{
"cell_type": "code",
"execution_count": 92,
"id": "356a5837-b0f4-4c63-95f0-42a562b8553e",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"entity\n",
"I-LOC 9\n",
"I-MISC 15\n",
"I-ORG 3\n",
"I-PER 3\n",
"dtype: int64\n"
]
},
{
"data": {
"text/plain": [
"entity word \n",
"I-LOC do 1\n",
" nia 1\n",
" ▁Cy 1\n",
" ▁Mars 5\n",
" ▁West 1\n",
"I-MISC an 1\n",
" bit 1\n",
" er 1\n",
" ion 1\n",
" ▁1 1\n",
" ▁American 1\n",
" ▁Egypt 1\n",
" ▁Face 1\n",
" ▁Mars 2\n",
" ▁Marti 1\n",
" ▁On 1\n",
" ▁Or 1\n",
" ▁Viking 2\n",
"I-ORG ▁NASA 3\n",
"I-PER ▁Malin 2\n",
" ▁Michael 1\n",
"dtype: int64"
]
},
"execution_count": 92,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"with open(\"14 FacebookAIxlm-roberta-large-finetuned-conll03-german.json\", encoding='utf-8') as f:\n",
" data = json.load(f)\n",
"aux=obtener_dataframe(data)\n",
"aux\n",
"print(aux.groupby(['entity']).size())\n",
"aux.groupby(['entity', 'word']).size()"
]
},
{
"cell_type": "markdown",
"id": "f34e36ec-8d26-49b4-89a5-15eaa203e168",
"metadata": {},
"source": [
"## 15 jplu/tf-xlm-r-ner-40-lang"
]
},
{
"cell_type": "code",
"execution_count": 103,
"id": "293a3320-8385-42a0-9aaf-fd24487ade80",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Requirement already satisfied: transformers[sentencepiece] in c:\\users\\nw\\anaconda3\\lib\\site-packages (4.41.2)\n",
"Requirement already satisfied: filelock in c:\\users\\nw\\anaconda3\\lib\\site-packages (from transformers[sentencepiece]) (3.13.1)\n",
"Requirement already satisfied: huggingface-hub<1.0,>=0.23.0 in c:\\users\\nw\\anaconda3\\lib\\site-packages (from transformers[sentencepiece]) (0.23.4)\n",
"Requirement already satisfied: numpy>=1.17 in c:\\users\\nw\\anaconda3\\lib\\site-packages (from transformers[sentencepiece]) (1.26.4)\n",
"Requirement already satisfied: packaging>=20.0 in c:\\users\\nw\\anaconda3\\lib\\site-packages (from transformers[sentencepiece]) (23.1)\n",
"Requirement already satisfied: pyyaml>=5.1 in c:\\users\\nw\\anaconda3\\lib\\site-packages (from transformers[sentencepiece]) (6.0.1)\n",
"Requirement already satisfied: regex!=2019.12.17 in c:\\users\\nw\\anaconda3\\lib\\site-packages (from transformers[sentencepiece]) (2023.10.3)\n",
"Requirement already satisfied: requests in c:\\users\\nw\\anaconda3\\lib\\site-packages (from transformers[sentencepiece]) (2.32.3)\n",
"Requirement already satisfied: tokenizers<0.20,>=0.19 in c:\\users\\nw\\anaconda3\\lib\\site-packages (from transformers[sentencepiece]) (0.19.1)\n",
"Requirement already satisfied: safetensors>=0.4.1 in c:\\users\\nw\\anaconda3\\lib\\site-packages (from transformers[sentencepiece]) (0.4.3)\n",
"Requirement already satisfied: tqdm>=4.27 in c:\\users\\nw\\anaconda3\\lib\\site-packages (from transformers[sentencepiece]) (4.66.4)\n",
"Requirement already satisfied: sentencepiece!=0.1.92,>=0.1.91 in c:\\users\\nw\\anaconda3\\lib\\site-packages (from transformers[sentencepiece]) (0.2.0)\n",
"Requirement already satisfied: protobuf in c:\\users\\nw\\anaconda3\\lib\\site-packages (from transformers[sentencepiece]) (3.20.3)\n",
"Requirement already satisfied: fsspec>=2023.5.0 in c:\\users\\nw\\anaconda3\\lib\\site-packages (from huggingface-hub<1.0,>=0.23.0->transformers[sentencepiece]) (2023.10.0)\n",
"Requirement already satisfied: typing-extensions>=3.7.4.3 in c:\\users\\nw\\anaconda3\\lib\\site-packages (from huggingface-hub<1.0,>=0.23.0->transformers[sentencepiece]) (4.9.0)\n",
"Requirement already satisfied: colorama in c:\\users\\nw\\anaconda3\\lib\\site-packages (from tqdm>=4.27->transformers[sentencepiece]) (0.4.6)\n",
"Requirement already satisfied: charset-normalizer<4,>=2 in c:\\users\\nw\\anaconda3\\lib\\site-packages (from requests->transformers[sentencepiece]) (2.0.4)\n",
"Requirement already satisfied: idna<4,>=2.5 in c:\\users\\nw\\anaconda3\\lib\\site-packages (from requests->transformers[sentencepiece]) (3.4)\n",
"Requirement already satisfied: urllib3<3,>=1.21.1 in c:\\users\\nw\\anaconda3\\lib\\site-packages (from requests->transformers[sentencepiece]) (2.0.7)\n",
"Requirement already satisfied: certifi>=2017.4.17 in c:\\users\\nw\\anaconda3\\lib\\site-packages (from requests->transformers[sentencepiece]) (2024.2.2)\n",
"Note: you may need to restart the kernel to use updated packages.\n"
]
}
],
"source": [
"pip install transformers[sentencepiece]"
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "a6a0c627-2ac4-48f0-8870-fd1a2479245c",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Some layers from the model checkpoint at jplu/tf-xlm-r-ner-40-lang were not used when initializing TFXLMRobertaForTokenClassification: ['dropout_38']\n",
"- This IS expected if you are initializing TFXLMRobertaForTokenClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
"- This IS NOT expected if you are initializing TFXLMRobertaForTokenClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
"All the layers of TFXLMRobertaForTokenClassification were initialized from the model checkpoint at jplu/tf-xlm-r-ner-40-lang.\n",
"If your task is similar to the task the model of the checkpoint was trained on, you can already use TFXLMRobertaForTokenClassification for predictions without further training.\n",
"Asking to truncate to max_length but no maximum length is provided and the model has no predefined maximum length. Default to no truncation.\n"
]
},
{
"data": {
"text/plain": [
"[{'entity': 'ORG',\n",
" 'score': 0.94845986,\n",
" 'index': 8,\n",
" 'word': '▁NASA',\n",
" 'start': 15,\n",
" 'end': 20},\n",
" {'entity': 'ORG',\n",
" 'score': 0.7998288,\n",
" 'index': 58,\n",
" 'word': '▁Viking',\n",
" 'start': 239,\n",
" 'end': 246},\n",
" {'entity': 'ORG',\n",
" 'score': 0.7579509,\n",
" 'index': 59,\n",
" 'word': '▁1',\n",
" 'start': 246,\n",
" 'end': 248},\n",
" {'entity': 'ORG',\n",
" 'score': 0.52262145,\n",
" 'index': 60,\n",
" 'word': '▁space',\n",
" 'start': 248,\n",
" 'end': 254},\n",
" {'entity': 'ORG',\n",
" 'score': 0.5880066,\n",
" 'index': 193,\n",
" 'word': '▁NASA',\n",
" 'start': 800,\n",
" 'end': 805},\n",
" {'entity': 'ORG',\n",
" 'score': 0.7987309,\n",
" 'index': 285,\n",
" 'word': '▁NASA',\n",
" 'start': 1168,\n",
" 'end': 1173},\n",
" {'entity': 'PER',\n",
" 'score': 0.9565463,\n",
" 'index': 319,\n",
" 'word': '▁Michael',\n",
" 'start': 1311,\n",
" 'end': 1319},\n",
" {'entity': 'PER',\n",
" 'score': 0.9528012,\n",
" 'index': 320,\n",
" 'word': '▁Malin',\n",
" 'start': 1319,\n",
" 'end': 1325},\n",
" {'entity': 'ORG',\n",
" 'score': 0.5696624,\n",
" 'index': 323,\n",
" 'word': '▁Mars',\n",
" 'start': 1333,\n",
" 'end': 1338},\n",
" {'entity': 'LOC',\n",
" 'score': 0.93097985,\n",
" 'index': 491,\n",
" 'word': '▁American',\n",
" 'start': 2093,\n",
" 'end': 2102},\n",
" {'entity': 'LOC',\n",
" 'score': 0.89757425,\n",
" 'index': 492,\n",
" 'word': '▁West',\n",
" 'start': 2102,\n",
" 'end': 2107}]"
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from transformers import pipeline\n",
"\n",
"nlp_ner = pipeline(\n",
" \"ner\",\n",
" model=\"jplu/tf-xlm-r-ner-40-lang\",\n",
" tokenizer=(\n",
" 'jplu/tf-xlm-r-ner-40-lang'),\n",
" framework=\"tf\"\n",
")\n",
"\n",
"nlp_ner(text)\n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": 7,
"id": "8ea0424d-e2da-4958-be66-2dc6e6556456",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"entity\n",
"LOC 2\n",
"ORG 7\n",
"PER 2\n",
"dtype: int64\n"
]
},
{
"data": {
"text/plain": [
"entity word \n",
"LOC ▁American 1\n",
" ▁West 1\n",
"ORG ▁1 1\n",
" ▁Mars 1\n",
" ▁NASA 3\n",
" ▁Viking 1\n",
" ▁space 1\n",
"PER ▁Malin 1\n",
" ▁Michael 1\n",
"dtype: int64"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"with open(\"15 jplutf-xlm-r-ner-40-lang.json\", encoding='utf-8') as f:\n",
" data = json.load(f)\n",
"aux=obtener_dataframe(data)\n",
"aux\n",
"print(aux.groupby(['entity']).size())\n",
"aux.groupby(['entity', 'word']).size()"
]
},
{
"cell_type": "markdown",
"id": "6fc9092a-d4c6-4365-9d73-205ea982bde6",
"metadata": {},
"source": [
"## 16 sagorsarker/codeswitch-spaeng-lid-lince"
]
},
{
"cell_type": "code",
"execution_count": 127,
"id": "c71bfa05-c1e7-4bae-bcc6-78962e062f1f",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Some weights of the model checkpoint at sagorsarker/codeswitch-spaeng-lid-lince were not used when initializing BertForTokenClassification: ['bert.pooler.dense.bias', 'bert.pooler.dense.weight']\n",
"- This IS expected if you are initializing BertForTokenClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
"- This IS NOT expected if you are initializing BertForTokenClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n"
]
},
{
"data": {
"text/plain": [
"[{'entity': 'en',\n",
" 'score': 0.9998629,\n",
" 'index': 1,\n",
" 'word': 'So',\n",
" 'start': 0,\n",
" 'end': 2},\n",
" {'entity': 'other',\n",
" 'score': 0.9999267,\n",
" 'index': 2,\n",
" 'word': ',',\n",
" 'start': 2,\n",
" 'end': 3},\n",
" {'entity': 'en',\n",
" 'score': 0.99985707,\n",
" 'index': 3,\n",
" 'word': 'if',\n",
" 'start': 4,\n",
" 'end': 6},\n",
" {'entity': 'en',\n",
" 'score': 0.99984396,\n",
" 'index': 4,\n",
" 'word': 'you',\n",
" 'start': 7,\n",
" 'end': 10},\n",
" {'entity': 'en',\n",
" 'score': 0.9998392,\n",
" 'index': 5,\n",
" 'word': \"'\",\n",
" 'start': 10,\n",
" 'end': 11},\n",
" {'entity': 'en',\n",
" 'score': 0.9998178,\n",
" 'index': 6,\n",
" 'word': 're',\n",
" 'start': 11,\n",
" 'end': 13},\n",
" {'entity': 'en',\n",
" 'score': 0.99961334,\n",
" 'index': 7,\n",
" 'word': 'a',\n",
" 'start': 14,\n",
" 'end': 15},\n",
" {'entity': 'ne',\n",
" 'score': 0.99565876,\n",
" 'index': 8,\n",
" 'word': 'NASA',\n",
" 'start': 16,\n",
" 'end': 20},\n",
" {'entity': 'en',\n",
" 'score': 0.9997341,\n",
" 'index': 9,\n",
" 'word': 'scientist',\n",
" 'start': 21,\n",
" 'end': 30},\n",
" {'entity': 'other',\n",
" 'score': 0.99992585,\n",
" 'index': 10,\n",
" 'word': ',',\n",
" 'start': 30,\n",
" 'end': 31},\n",
" {'entity': 'en',\n",
" 'score': 0.99986625,\n",
" 'index': 11,\n",
" 'word': 'you',\n",
" 'start': 32,\n",
" 'end': 35},\n",
" {'entity': 'en',\n",
" 'score': 0.99986553,\n",
" 'index': 12,\n",
" 'word': 'should',\n",
" 'start': 36,\n",
" 'end': 42},\n",
" {'entity': 'en',\n",
" 'score': 0.9998518,\n",
" 'index': 13,\n",
" 'word': 'be',\n",
" 'start': 43,\n",
" 'end': 45},\n",
" {'entity': 'en',\n",
" 'score': 0.9998604,\n",
" 'index': 14,\n",
" 'word': 'able',\n",
" 'start': 46,\n",
" 'end': 50},\n",
" {'entity': 'en',\n",
" 'score': 0.99985576,\n",
" 'index': 15,\n",
" 'word': 'to',\n",
" 'start': 51,\n",
" 'end': 53},\n",
" {'entity': 'en',\n",
" 'score': 0.999874,\n",
" 'index': 16,\n",
" 'word': 'tell',\n",
" 'start': 54,\n",
" 'end': 58},\n",
" {'entity': 'en',\n",
" 'score': 0.9998467,\n",
" 'index': 17,\n",
" 'word': 'me',\n",
" 'start': 59,\n",
" 'end': 61},\n",
" {'entity': 'en',\n",
" 'score': 0.99984443,\n",
" 'index': 18,\n",
" 'word': 'the',\n",
" 'start': 62,\n",
" 'end': 65},\n",
" {'entity': 'en',\n",
" 'score': 0.99985886,\n",
" 'index': 19,\n",
" 'word': 'whole',\n",
" 'start': 66,\n",
" 'end': 71},\n",
" {'entity': 'en',\n",
" 'score': 0.9998666,\n",
" 'index': 20,\n",
" 'word': 'story',\n",
" 'start': 72,\n",
" 'end': 77},\n",
" {'entity': 'en',\n",
" 'score': 0.99987817,\n",
" 'index': 21,\n",
" 'word': 'about',\n",
" 'start': 78,\n",
" 'end': 83},\n",
" {'entity': 'en',\n",
" 'score': 0.99987066,\n",
" 'index': 22,\n",
" 'word': 'the',\n",
" 'start': 84,\n",
" 'end': 87},\n",
" {'entity': 'en',\n",
" 'score': 0.9998729,\n",
" 'index': 23,\n",
" 'word': 'Face',\n",
" 'start': 88,\n",
" 'end': 92},\n",
" {'entity': 'en',\n",
" 'score': 0.9998679,\n",
" 'index': 24,\n",
" 'word': 'On',\n",
" 'start': 93,\n",
" 'end': 95},\n",
" {'entity': 'ne',\n",
" 'score': 0.9943815,\n",
" 'index': 25,\n",
" 'word': 'Mars',\n",
" 'start': 96,\n",
" 'end': 100},\n",
" {'entity': 'other',\n",
" 'score': 0.9999286,\n",
" 'index': 26,\n",
" 'word': ',',\n",
" 'start': 100,\n",
" 'end': 101},\n",
" {'entity': 'en',\n",
" 'score': 0.9998467,\n",
" 'index': 27,\n",
" 'word': 'which',\n",
" 'start': 102,\n",
" 'end': 107},\n",
" {'entity': 'en',\n",
" 'score': 0.9998258,\n",
" 'index': 28,\n",
" 'word': 'obvious',\n",
" 'start': 108,\n",
" 'end': 115},\n",
" {'entity': 'en',\n",
" 'score': 0.99984264,\n",
" 'index': 29,\n",
" 'word': '##ly',\n",
" 'start': 115,\n",
" 'end': 117},\n",
" {'entity': 'en',\n",
" 'score': 0.9998216,\n",
" 'index': 30,\n",
" 'word': 'is',\n",
" 'start': 118,\n",
" 'end': 120},\n",
" {'entity': 'en',\n",
" 'score': 0.9998173,\n",
" 'index': 31,\n",
" 'word': 'evidence',\n",
" 'start': 121,\n",
" 'end': 129},\n",
" {'entity': 'en',\n",
" 'score': 0.9998375,\n",
" 'index': 32,\n",
" 'word': 'that',\n",
" 'start': 130,\n",
" 'end': 134},\n",
" {'entity': 'en',\n",
" 'score': 0.9998186,\n",
" 'index': 33,\n",
" 'word': 'there',\n",
" 'start': 135,\n",
" 'end': 140},\n",
" {'entity': 'en',\n",
" 'score': 0.99982053,\n",
" 'index': 34,\n",
" 'word': 'is',\n",
" 'start': 141,\n",
" 'end': 143},\n",
" {'entity': 'en',\n",
" 'score': 0.99982506,\n",
" 'index': 35,\n",
" 'word': 'life',\n",
" 'start': 144,\n",
" 'end': 148},\n",
" {'entity': 'en',\n",
" 'score': 0.99981946,\n",
" 'index': 36,\n",
" 'word': 'on',\n",
" 'start': 149,\n",
" 'end': 151},\n",
" {'entity': 'ne',\n",
" 'score': 0.99421823,\n",
" 'index': 37,\n",
" 'word': 'Mars',\n",
" 'start': 152,\n",
" 'end': 156},\n",
" {'entity': 'other',\n",
" 'score': 0.9999292,\n",
" 'index': 38,\n",
" 'word': ',',\n",
" 'start': 156,\n",
" 'end': 157},\n",
" {'entity': 'en',\n",
" 'score': 0.99983776,\n",
" 'index': 39,\n",
" 'word': 'and',\n",
" 'start': 158,\n",
" 'end': 161},\n",
" {'entity': 'en',\n",
" 'score': 0.9998344,\n",
" 'index': 40,\n",
" 'word': 'that',\n",
" 'start': 162,\n",
" 'end': 166},\n",
" {'entity': 'en',\n",
" 'score': 0.9997547,\n",
" 'index': 41,\n",
" 'word': 'the',\n",
" 'start': 167,\n",
" 'end': 170},\n",
" {'entity': 'en',\n",
" 'score': 0.99978274,\n",
" 'index': 42,\n",
" 'word': 'face',\n",
" 'start': 171,\n",
" 'end': 175},\n",
" {'entity': 'en',\n",
" 'score': 0.9997836,\n",
" 'index': 43,\n",
" 'word': 'was',\n",
" 'start': 176,\n",
" 'end': 179},\n",
" {'entity': 'en',\n",
" 'score': 0.99971956,\n",
" 'index': 44,\n",
" 'word': 'created',\n",
" 'start': 180,\n",
" 'end': 187},\n",
" {'entity': 'en',\n",
" 'score': 0.99975187,\n",
" 'index': 45,\n",
" 'word': 'by',\n",
" 'start': 188,\n",
" 'end': 190},\n",
" {'entity': 'en',\n",
" 'score': 0.9997137,\n",
" 'index': 46,\n",
" 'word': 'alien',\n",
" 'start': 191,\n",
" 'end': 196},\n",
" {'entity': 'en',\n",
" 'score': 0.99977976,\n",
" 'index': 47,\n",
" 'word': '##s',\n",
" 'start': 196,\n",
" 'end': 197},\n",
" {'entity': 'other',\n",
" 'score': 0.9999249,\n",
" 'index': 48,\n",
" 'word': ',',\n",
" 'start': 197,\n",
" 'end': 198},\n",
" {'entity': 'en',\n",
" 'score': 0.99981195,\n",
" 'index': 49,\n",
" 'word': 'correct',\n",
" 'start': 199,\n",
" 'end': 206},\n",
" {'entity': 'other',\n",
" 'score': 0.9999312,\n",
" 'index': 50,\n",
" 'word': '?',\n",
" 'start': 206,\n",
" 'end': 207},\n",
" {'entity': 'other',\n",
" 'score': 0.9999089,\n",
" 'index': 51,\n",
" 'word': '\"',\n",
" 'start': 207,\n",
" 'end': 208},\n",
" {'entity': 'en',\n",
" 'score': 0.9864689,\n",
" 'index': 52,\n",
" 'word': 'No',\n",
" 'start': 209,\n",
" 'end': 211},\n",
" {'entity': 'other',\n",
" 'score': 0.99993014,\n",
" 'index': 53,\n",
" 'word': ',',\n",
" 'start': 211,\n",
" 'end': 212},\n",
" {'entity': 'en',\n",
" 'score': 0.9997601,\n",
" 'index': 54,\n",
" 'word': 'twenty',\n",
" 'start': 213,\n",
" 'end': 219},\n",
" {'entity': 'en',\n",
" 'score': 0.9996724,\n",
" 'index': 55,\n",
" 'word': 'five',\n",
" 'start': 220,\n",
" 'end': 224},\n",
" {'entity': 'en',\n",
" 'score': 0.99973506,\n",
" 'index': 56,\n",
" 'word': 'years',\n",
" 'start': 225,\n",
" 'end': 230},\n",
" {'entity': 'en',\n",
" 'score': 0.9997675,\n",
" 'index': 57,\n",
" 'word': 'ago',\n",
" 'start': 231,\n",
" 'end': 234},\n",
" {'entity': 'other',\n",
" 'score': 0.9999273,\n",
" 'index': 58,\n",
" 'word': ',',\n",
" 'start': 234,\n",
" 'end': 235},\n",
" {'entity': 'en',\n",
" 'score': 0.9992092,\n",
" 'index': 59,\n",
" 'word': 'our',\n",
" 'start': 236,\n",
" 'end': 239},\n",
" {'entity': 'ne',\n",
" 'score': 0.62706536,\n",
" 'index': 60,\n",
" 'word': 'Viking',\n",
" 'start': 240,\n",
" 'end': 246},\n",
" {'entity': 'other',\n",
" 'score': 0.9996927,\n",
" 'index': 61,\n",
" 'word': '1',\n",
" 'start': 247,\n",
" 'end': 248},\n",
" {'entity': 'en',\n",
" 'score': 0.9995969,\n",
" 'index': 62,\n",
" 'word': 'spacecraft',\n",
" 'start': 249,\n",
" 'end': 259},\n",
" {'entity': 'en',\n",
" 'score': 0.9997328,\n",
" 'index': 63,\n",
" 'word': 'was',\n",
" 'start': 260,\n",
" 'end': 263},\n",
" {'entity': 'en',\n",
" 'score': 0.9997286,\n",
" 'index': 64,\n",
" 'word': 'ci',\n",
" 'start': 264,\n",
" 'end': 266},\n",
" {'entity': 'en',\n",
" 'score': 0.9997693,\n",
" 'index': 65,\n",
" 'word': '##rc',\n",
" 'start': 266,\n",
" 'end': 268},\n",
" {'entity': 'en',\n",
" 'score': 0.99981683,\n",
" 'index': 66,\n",
" 'word': '##ling',\n",
" 'start': 268,\n",
" 'end': 272},\n",
" {'entity': 'en',\n",
" 'score': 0.99974436,\n",
" 'index': 67,\n",
" 'word': 'the',\n",
" 'start': 273,\n",
" 'end': 276},\n",
" {'entity': 'en',\n",
" 'score': 0.999772,\n",
" 'index': 68,\n",
" 'word': 'planet',\n",
" 'start': 277,\n",
" 'end': 283},\n",
" {'entity': 'other',\n",
" 'score': 0.9999294,\n",
" 'index': 69,\n",
" 'word': ',',\n",
" 'start': 283,\n",
" 'end': 284},\n",
" {'entity': 'en',\n",
" 'score': 0.9995789,\n",
" 'index': 70,\n",
" 'word': 'sna',\n",
" 'start': 285,\n",
" 'end': 288},\n",
" {'entity': 'en',\n",
" 'score': 0.9996741,\n",
" 'index': 71,\n",
" 'word': '##pping',\n",
" 'start': 288,\n",
" 'end': 293},\n",
" {'entity': 'en',\n",
" 'score': 0.99957246,\n",
" 'index': 72,\n",
" 'word': 'photos',\n",
" 'start': 294,\n",
" 'end': 300},\n",
" {'entity': 'other',\n",
" 'score': 0.99993145,\n",
" 'index': 73,\n",
" 'word': ',',\n",
" 'start': 300,\n",
" 'end': 301},\n",
" {'entity': 'en',\n",
" 'score': 0.9997521,\n",
" 'index': 74,\n",
" 'word': 'when',\n",
" 'start': 302,\n",
" 'end': 306},\n",
" {'entity': 'en',\n",
" 'score': 0.99976677,\n",
" 'index': 75,\n",
" 'word': 'it',\n",
" 'start': 307,\n",
" 'end': 309},\n",
" {'entity': 'en',\n",
" 'score': 0.99971706,\n",
" 'index': 76,\n",
" 'word': 'spotted',\n",
" 'start': 310,\n",
" 'end': 317},\n",
" {'entity': 'en',\n",
" 'score': 0.99961495,\n",
" 'index': 77,\n",
" 'word': 'the',\n",
" 'start': 318,\n",
" 'end': 321},\n",
" {'entity': 'en',\n",
" 'score': 0.9996594,\n",
" 'index': 78,\n",
" 'word': 'sh',\n",
" 'start': 322,\n",
" 'end': 324},\n",
" {'entity': 'en',\n",
" 'score': 0.99976844,\n",
" 'index': 79,\n",
" 'word': '##adow',\n",
" 'start': 324,\n",
" 'end': 328},\n",
" {'entity': 'en',\n",
" 'score': 0.9997464,\n",
" 'index': 80,\n",
" 'word': '##y',\n",
" 'start': 328,\n",
" 'end': 329},\n",
" {'entity': 'en',\n",
" 'score': 0.9997322,\n",
" 'index': 81,\n",
" 'word': 'like',\n",
" 'start': 330,\n",
" 'end': 334},\n",
" {'entity': 'en',\n",
" 'score': 0.99969065,\n",
" 'index': 82,\n",
" 'word': '##ness',\n",
" 'start': 334,\n",
" 'end': 338},\n",
" {'entity': 'en',\n",
" 'score': 0.9996531,\n",
" 'index': 83,\n",
" 'word': 'of',\n",
" 'start': 339,\n",
" 'end': 341},\n",
" {'entity': 'en',\n",
" 'score': 0.9992725,\n",
" 'index': 84,\n",
" 'word': 'a',\n",
" 'start': 342,\n",
" 'end': 343},\n",
" {'entity': 'en',\n",
" 'score': 0.99959975,\n",
" 'index': 85,\n",
" 'word': 'human',\n",
" 'start': 344,\n",
" 'end': 349},\n",
" {'entity': 'en',\n",
" 'score': 0.99973136,\n",
" 'index': 86,\n",
" 'word': 'face',\n",
" 'start': 350,\n",
" 'end': 354},\n",
" {'entity': 'other',\n",
" 'score': 0.9999311,\n",
" 'index': 87,\n",
" 'word': '.',\n",
" 'start': 354,\n",
" 'end': 355},\n",
" {'entity': 'en',\n",
" 'score': 0.9996131,\n",
" 'index': 88,\n",
" 'word': 'Us',\n",
" 'start': 356,\n",
" 'end': 358},\n",
" {'entity': 'en',\n",
" 'score': 0.9996656,\n",
" 'index': 89,\n",
" 'word': 'scientists',\n",
" 'start': 359,\n",
" 'end': 369},\n",
" {'entity': 'en',\n",
" 'score': 0.99980336,\n",
" 'index': 90,\n",
" 'word': 'figure',\n",
" 'start': 370,\n",
" 'end': 376},\n",
" {'entity': 'en',\n",
" 'score': 0.999795,\n",
" 'index': 91,\n",
" 'word': '##d',\n",
" 'start': 376,\n",
" 'end': 377},\n",
" {'entity': 'en',\n",
" 'score': 0.99979943,\n",
" 'index': 92,\n",
" 'word': 'out',\n",
" 'start': 378,\n",
" 'end': 381},\n",
" {'entity': 'en',\n",
" 'score': 0.9997974,\n",
" 'index': 93,\n",
" 'word': 'that',\n",
" 'start': 382,\n",
" 'end': 386},\n",
" {'entity': 'en',\n",
" 'score': 0.99973387,\n",
" 'index': 94,\n",
" 'word': 'it',\n",
" 'start': 387,\n",
" 'end': 389},\n",
" {'entity': 'en',\n",
" 'score': 0.9997162,\n",
" 'index': 95,\n",
" 'word': 'was',\n",
" 'start': 390,\n",
" 'end': 393},\n",
" {'entity': 'en',\n",
" 'score': 0.9996761,\n",
" 'index': 96,\n",
" 'word': 'just',\n",
" 'start': 394,\n",
" 'end': 398},\n",
" {'entity': 'en',\n",
" 'score': 0.9995012,\n",
" 'index': 97,\n",
" 'word': 'another',\n",
" 'start': 399,\n",
" 'end': 406},\n",
" {'entity': 'en',\n",
" 'score': 0.92382264,\n",
" 'index': 98,\n",
" 'word': 'Mart',\n",
" 'start': 407,\n",
" 'end': 411},\n",
" {'entity': 'en',\n",
" 'score': 0.9973562,\n",
" 'index': 99,\n",
" 'word': '##ian',\n",
" 'start': 411,\n",
" 'end': 414},\n",
" {'entity': 'en',\n",
" 'score': 0.7907492,\n",
" 'index': 100,\n",
" 'word': 'mesa',\n",
" 'start': 415,\n",
" 'end': 419},\n",
" {'entity': 'other',\n",
" 'score': 0.9999194,\n",
" 'index': 101,\n",
" 'word': ',',\n",
" 'start': 419,\n",
" 'end': 420},\n",
" {'entity': 'en',\n",
" 'score': 0.9995608,\n",
" 'index': 102,\n",
" 'word': 'common',\n",
" 'start': 421,\n",
" 'end': 427},\n",
" {'entity': 'en',\n",
" 'score': 0.9996966,\n",
" 'index': 103,\n",
" 'word': 'around',\n",
" 'start': 428,\n",
" 'end': 434},\n",
" {'entity': 'ne',\n",
" 'score': 0.9975068,\n",
" 'index': 104,\n",
" 'word': 'C',\n",
" 'start': 435,\n",
" 'end': 436},\n",
" {'entity': 'ne',\n",
" 'score': 0.9957877,\n",
" 'index': 105,\n",
" 'word': '##yd',\n",
" 'start': 436,\n",
" 'end': 438},\n",
" {'entity': 'ne',\n",
" 'score': 0.9961337,\n",
" 'index': 106,\n",
" 'word': '##onia',\n",
" 'start': 438,\n",
" 'end': 442},\n",
" {'entity': 'other',\n",
" 'score': 0.9999093,\n",
" 'index': 107,\n",
" 'word': ',',\n",
" 'start': 442,\n",
" 'end': 443},\n",
" {'entity': 'en',\n",
" 'score': 0.99976605,\n",
" 'index': 108,\n",
" 'word': 'only',\n",
" 'start': 444,\n",
" 'end': 448},\n",
" {'entity': 'en',\n",
" 'score': 0.9997769,\n",
" 'index': 109,\n",
" 'word': 'this',\n",
" 'start': 449,\n",
" 'end': 453},\n",
" {'entity': 'en',\n",
" 'score': 0.99976176,\n",
" 'index': 110,\n",
" 'word': 'one',\n",
" 'start': 454,\n",
" 'end': 457},\n",
" {'entity': 'en',\n",
" 'score': 0.9997123,\n",
" 'index': 111,\n",
" 'word': 'had',\n",
" 'start': 458,\n",
" 'end': 461},\n",
" {'entity': 'en',\n",
" 'score': 0.9997372,\n",
" 'index': 112,\n",
" 'word': 'sh',\n",
" 'start': 462,\n",
" 'end': 464},\n",
" {'entity': 'en',\n",
" 'score': 0.9998115,\n",
" 'index': 113,\n",
" 'word': '##adow',\n",
" 'start': 464,\n",
" 'end': 468},\n",
" {'entity': 'en',\n",
" 'score': 0.99980253,\n",
" 'index': 114,\n",
" 'word': '##s',\n",
" 'start': 468,\n",
" 'end': 469},\n",
" {'entity': 'en',\n",
" 'score': 0.99979705,\n",
" 'index': 115,\n",
" 'word': 'that',\n",
" 'start': 470,\n",
" 'end': 474},\n",
" {'entity': 'en',\n",
" 'score': 0.99979264,\n",
" 'index': 116,\n",
" 'word': 'made',\n",
" 'start': 475,\n",
" 'end': 479},\n",
" {'entity': 'en',\n",
" 'score': 0.9998178,\n",
" 'index': 117,\n",
" 'word': 'it',\n",
" 'start': 480,\n",
" 'end': 482},\n",
" {'entity': 'en',\n",
" 'score': 0.99979025,\n",
" 'index': 118,\n",
" 'word': 'look',\n",
" 'start': 483,\n",
" 'end': 487},\n",
" {'entity': 'en',\n",
" 'score': 0.99977773,\n",
" 'index': 119,\n",
" 'word': 'like',\n",
" 'start': 488,\n",
" 'end': 492},\n",
" {'entity': 'en',\n",
" 'score': 0.99962866,\n",
" 'index': 120,\n",
" 'word': 'an',\n",
" 'start': 493,\n",
" 'end': 495},\n",
" {'entity': 'ne',\n",
" 'score': 0.68299395,\n",
" 'index': 121,\n",
" 'word': 'Egypt',\n",
" 'start': 496,\n",
" 'end': 501},\n",
" {'entity': 'en',\n",
" 'score': 0.9760886,\n",
" 'index': 122,\n",
" 'word': '##ion',\n",
" 'start': 501,\n",
" 'end': 504},\n",
" {'entity': 'en',\n",
" 'score': 0.9997099,\n",
" 'index': 123,\n",
" 'word': 'Ph',\n",
" 'start': 505,\n",
" 'end': 507},\n",
" {'entity': 'en',\n",
" 'score': 0.9998056,\n",
" 'index': 124,\n",
" 'word': '##ara',\n",
" 'start': 507,\n",
" 'end': 510},\n",
" {'entity': 'en',\n",
" 'score': 0.99973387,\n",
" 'index': 125,\n",
" 'word': '##oh',\n",
" 'start': 510,\n",
" 'end': 512},\n",
" {'entity': 'other',\n",
" 'score': 0.9999268,\n",
" 'index': 126,\n",
" 'word': '.',\n",
" 'start': 512,\n",
" 'end': 513},\n",
" {'entity': 'en',\n",
" 'score': 0.99796546,\n",
" 'index': 127,\n",
" 'word': 'Very',\n",
" 'start': 514,\n",
" 'end': 518},\n",
" {'entity': 'en',\n",
" 'score': 0.9990651,\n",
" 'index': 128,\n",
" 'word': 'few',\n",
" 'start': 519,\n",
" 'end': 522},\n",
" {'entity': 'en',\n",
" 'score': 0.9994499,\n",
" 'index': 129,\n",
" 'word': 'days',\n",
" 'start': 523,\n",
" 'end': 527},\n",
" {'entity': 'en',\n",
" 'score': 0.9995864,\n",
" 'index': 130,\n",
" 'word': 'later',\n",
" 'start': 528,\n",
" 'end': 533},\n",
" {'entity': 'other',\n",
" 'score': 0.99990845,\n",
" 'index': 131,\n",
" 'word': ',',\n",
" 'start': 533,\n",
" 'end': 534},\n",
" {'entity': 'en',\n",
" 'score': 0.99973947,\n",
" 'index': 132,\n",
" 'word': 'we',\n",
" 'start': 535,\n",
" 'end': 537},\n",
" {'entity': 'en',\n",
" 'score': 0.99979526,\n",
" 'index': 133,\n",
" 'word': 'revealed',\n",
" 'start': 538,\n",
" 'end': 546},\n",
" {'entity': 'en',\n",
" 'score': 0.9997683,\n",
" 'index': 134,\n",
" 'word': 'the',\n",
" 'start': 547,\n",
" 'end': 550},\n",
" {'entity': 'en',\n",
" 'score': 0.999793,\n",
" 'index': 135,\n",
" 'word': 'image',\n",
" 'start': 551,\n",
" 'end': 556},\n",
" {'entity': 'en',\n",
" 'score': 0.99985754,\n",
" 'index': 136,\n",
" 'word': 'for',\n",
" 'start': 557,\n",
" 'end': 560},\n",
" {'entity': 'en',\n",
" 'score': 0.999843,\n",
" 'index': 137,\n",
" 'word': 'all',\n",
" 'start': 561,\n",
" 'end': 564},\n",
" {'entity': 'en',\n",
" 'score': 0.99983895,\n",
" 'index': 138,\n",
" 'word': 'to',\n",
" 'start': 565,\n",
" 'end': 567},\n",
" {'entity': 'en',\n",
" 'score': 0.9998221,\n",
" 'index': 139,\n",
" 'word': 'see',\n",
" 'start': 568,\n",
" 'end': 571},\n",
" {'entity': 'other',\n",
" 'score': 0.99992704,\n",
" 'index': 140,\n",
" 'word': ',',\n",
" 'start': 571,\n",
" 'end': 572},\n",
" {'entity': 'en',\n",
" 'score': 0.9997582,\n",
" 'index': 141,\n",
" 'word': 'and',\n",
" 'start': 573,\n",
" 'end': 576},\n",
" {'entity': 'en',\n",
" 'score': 0.99974746,\n",
" 'index': 142,\n",
" 'word': 'we',\n",
" 'start': 577,\n",
" 'end': 579},\n",
" {'entity': 'en',\n",
" 'score': 0.9997589,\n",
" 'index': 143,\n",
" 'word': 'made',\n",
" 'start': 580,\n",
" 'end': 584},\n",
" {'entity': 'en',\n",
" 'score': 0.9997596,\n",
" 'index': 144,\n",
" 'word': 'sure',\n",
" 'start': 585,\n",
" 'end': 589},\n",
" {'entity': 'en',\n",
" 'score': 0.9997719,\n",
" 'index': 145,\n",
" 'word': 'to',\n",
" 'start': 590,\n",
" 'end': 592},\n",
" {'entity': 'en',\n",
" 'score': 0.99976,\n",
" 'index': 146,\n",
" 'word': 'note',\n",
" 'start': 593,\n",
" 'end': 597},\n",
" {'entity': 'en',\n",
" 'score': 0.9997385,\n",
" 'index': 147,\n",
" 'word': 'that',\n",
" 'start': 598,\n",
" 'end': 602},\n",
" {'entity': 'en',\n",
" 'score': 0.999699,\n",
" 'index': 148,\n",
" 'word': 'it',\n",
" 'start': 603,\n",
" 'end': 605},\n",
" {'entity': 'en',\n",
" 'score': 0.9996177,\n",
" 'index': 149,\n",
" 'word': 'was',\n",
" 'start': 606,\n",
" 'end': 609},\n",
" {'entity': 'en',\n",
" 'score': 0.99906355,\n",
" 'index': 150,\n",
" 'word': 'a',\n",
" 'start': 610,\n",
" 'end': 611},\n",
" {'entity': 'en',\n",
" 'score': 0.9993754,\n",
" 'index': 151,\n",
" 'word': 'huge',\n",
" 'start': 612,\n",
" 'end': 616},\n",
" {'entity': 'en',\n",
" 'score': 0.99958175,\n",
" 'index': 152,\n",
" 'word': 'rock',\n",
" 'start': 617,\n",
" 'end': 621},\n",
" {'entity': 'en',\n",
" 'score': 0.9996152,\n",
" 'index': 153,\n",
" 'word': 'formation',\n",
" 'start': 622,\n",
" 'end': 631},\n",
" {'entity': 'en',\n",
" 'score': 0.9995758,\n",
" 'index': 154,\n",
" 'word': 'that',\n",
" 'start': 632,\n",
" 'end': 636},\n",
" {'entity': 'en',\n",
" 'score': 0.9995844,\n",
" 'index': 155,\n",
" 'word': 'just',\n",
" 'start': 637,\n",
" 'end': 641},\n",
" {'entity': 'en',\n",
" 'score': 0.99950886,\n",
" 'index': 156,\n",
" 'word': 'res',\n",
" 'start': 642,\n",
" 'end': 645},\n",
" {'entity': 'en',\n",
" 'score': 0.9995715,\n",
" 'index': 157,\n",
" 'word': '##emble',\n",
" 'start': 645,\n",
" 'end': 650},\n",
" {'entity': 'en',\n",
" 'score': 0.99958056,\n",
" 'index': 158,\n",
" 'word': '##d',\n",
" 'start': 650,\n",
" 'end': 651},\n",
" {'entity': 'en',\n",
" 'score': 0.99861777,\n",
" 'index': 159,\n",
" 'word': 'a',\n",
" 'start': 652,\n",
" 'end': 653},\n",
" {'entity': 'en',\n",
" 'score': 0.99934405,\n",
" 'index': 160,\n",
" 'word': 'human',\n",
" 'start': 654,\n",
" 'end': 659},\n",
" {'entity': 'en',\n",
" 'score': 0.9995927,\n",
" 'index': 161,\n",
" 'word': 'head',\n",
" 'start': 660,\n",
" 'end': 664},\n",
" {'entity': 'en',\n",
" 'score': 0.9994898,\n",
" 'index': 162,\n",
" 'word': 'and',\n",
" 'start': 665,\n",
" 'end': 668},\n",
" {'entity': 'en',\n",
" 'score': 0.9995479,\n",
" 'index': 163,\n",
" 'word': 'face',\n",
" 'start': 669,\n",
" 'end': 673},\n",
" {'entity': 'other',\n",
" 'score': 0.9999279,\n",
" 'index': 164,\n",
" 'word': ',',\n",
" 'start': 673,\n",
" 'end': 674},\n",
" {'entity': 'en',\n",
" 'score': 0.9997607,\n",
" 'index': 165,\n",
" 'word': 'but',\n",
" 'start': 675,\n",
" 'end': 678},\n",
" {'entity': 'en',\n",
" 'score': 0.99969375,\n",
" 'index': 166,\n",
" 'word': 'all',\n",
" 'start': 679,\n",
" 'end': 682},\n",
" {'entity': 'en',\n",
" 'score': 0.99976605,\n",
" 'index': 167,\n",
" 'word': 'of',\n",
" 'start': 683,\n",
" 'end': 685},\n",
" {'entity': 'en',\n",
" 'score': 0.99975795,\n",
" 'index': 168,\n",
" 'word': 'it',\n",
" 'start': 686,\n",
" 'end': 688},\n",
" {'entity': 'en',\n",
" 'score': 0.99969816,\n",
" 'index': 169,\n",
" 'word': 'was',\n",
" 'start': 689,\n",
" 'end': 692},\n",
" {'entity': 'en',\n",
" 'score': 0.9996636,\n",
" 'index': 170,\n",
" 'word': 'formed',\n",
" 'start': 693,\n",
" 'end': 699},\n",
" {'entity': 'en',\n",
" 'score': 0.9995894,\n",
" 'index': 171,\n",
" 'word': 'by',\n",
" 'start': 700,\n",
" 'end': 702},\n",
" {'entity': 'en',\n",
" 'score': 0.9996039,\n",
" 'index': 172,\n",
" 'word': 'sh',\n",
" 'start': 703,\n",
" 'end': 705},\n",
" {'entity': 'en',\n",
" 'score': 0.9997403,\n",
" 'index': 173,\n",
" 'word': '##adow',\n",
" 'start': 705,\n",
" 'end': 709},\n",
" {'entity': 'en',\n",
" 'score': 0.9997271,\n",
" 'index': 174,\n",
" 'word': '##s',\n",
" 'start': 709,\n",
" 'end': 710},\n",
" {'entity': 'other',\n",
" 'score': 0.9999298,\n",
" 'index': 175,\n",
" 'word': '.',\n",
" 'start': 710,\n",
" 'end': 711},\n",
" {'entity': 'en',\n",
" 'score': 0.9996859,\n",
" 'index': 176,\n",
" 'word': 'We',\n",
" 'start': 712,\n",
" 'end': 714},\n",
" {'entity': 'en',\n",
" 'score': 0.99971765,\n",
" 'index': 177,\n",
" 'word': 'only',\n",
" 'start': 715,\n",
" 'end': 719},\n",
" {'entity': 'en',\n",
" 'score': 0.99975497,\n",
" 'index': 178,\n",
" 'word': 'announced',\n",
" 'start': 720,\n",
" 'end': 729},\n",
" {'entity': 'en',\n",
" 'score': 0.9997633,\n",
" 'index': 179,\n",
" 'word': 'it',\n",
" 'start': 730,\n",
" 'end': 732},\n",
" {'entity': 'en',\n",
" 'score': 0.9997826,\n",
" 'index': 180,\n",
" 'word': 'because',\n",
" 'start': 733,\n",
" 'end': 740},\n",
" {'entity': 'en',\n",
" 'score': 0.99974495,\n",
" 'index': 181,\n",
" 'word': 'we',\n",
" 'start': 741,\n",
" 'end': 743},\n",
" {'entity': 'en',\n",
" 'score': 0.9997533,\n",
" 'index': 182,\n",
" 'word': 'thought',\n",
" 'start': 744,\n",
" 'end': 751},\n",
" {'entity': 'en',\n",
" 'score': 0.9997551,\n",
" 'index': 183,\n",
" 'word': 'it',\n",
" 'start': 752,\n",
" 'end': 754},\n",
" {'entity': 'en',\n",
" 'score': 0.99974245,\n",
" 'index': 184,\n",
" 'word': 'would',\n",
" 'start': 755,\n",
" 'end': 760},\n",
" {'entity': 'en',\n",
" 'score': 0.9996723,\n",
" 'index': 185,\n",
" 'word': 'be',\n",
" 'start': 761,\n",
" 'end': 763},\n",
" {'entity': 'en',\n",
" 'score': 0.9992391,\n",
" 'index': 186,\n",
" 'word': 'a',\n",
" 'start': 764,\n",
" 'end': 765},\n",
" {'entity': 'en',\n",
" 'score': 0.9996873,\n",
" 'index': 187,\n",
" 'word': 'good',\n",
" 'start': 766,\n",
" 'end': 770},\n",
" {'entity': 'en',\n",
" 'score': 0.9997737,\n",
" 'index': 188,\n",
" 'word': 'way',\n",
" 'start': 771,\n",
" 'end': 774},\n",
" {'entity': 'en',\n",
" 'score': 0.99975866,\n",
" 'index': 189,\n",
" 'word': 'to',\n",
" 'start': 775,\n",
" 'end': 777},\n",
" {'entity': 'en',\n",
" 'score': 0.99975985,\n",
" 'index': 190,\n",
" 'word': 'engage',\n",
" 'start': 778,\n",
" 'end': 784},\n",
" {'entity': 'en',\n",
" 'score': 0.9996723,\n",
" 'index': 191,\n",
" 'word': 'the',\n",
" 'start': 785,\n",
" 'end': 788},\n",
" {'entity': 'en',\n",
" 'score': 0.9997609,\n",
" 'index': 192,\n",
" 'word': 'public',\n",
" 'start': 789,\n",
" 'end': 795},\n",
" {'entity': 'en',\n",
" 'score': 0.9997042,\n",
" 'index': 193,\n",
" 'word': 'with',\n",
" 'start': 796,\n",
" 'end': 800},\n",
" {'entity': 'ne',\n",
" 'score': 0.99519366,\n",
" 'index': 194,\n",
" 'word': 'NASA',\n",
" 'start': 801,\n",
" 'end': 805},\n",
" {'entity': 'en',\n",
" 'score': 0.99966383,\n",
" 'index': 195,\n",
" 'word': \"'\",\n",
" 'start': 805,\n",
" 'end': 806},\n",
" {'entity': 'en',\n",
" 'score': 0.99948585,\n",
" 'index': 196,\n",
" 'word': 's',\n",
" 'start': 806,\n",
" 'end': 807},\n",
" {'entity': 'en',\n",
" 'score': 0.9994748,\n",
" 'index': 197,\n",
" 'word': 'findings',\n",
" 'start': 808,\n",
" 'end': 816},\n",
" {'entity': 'other',\n",
" 'score': 0.99991834,\n",
" 'index': 198,\n",
" 'word': ',',\n",
" 'start': 816,\n",
" 'end': 817},\n",
" {'entity': 'en',\n",
" 'score': 0.99959546,\n",
" 'index': 199,\n",
" 'word': 'and',\n",
" 'start': 818,\n",
" 'end': 821},\n",
" {'entity': 'en',\n",
" 'score': 0.9996618,\n",
" 'index': 200,\n",
" 'word': 'at',\n",
" 'start': 822,\n",
" 'end': 824},\n",
" {'entity': 'en',\n",
" 'score': 0.99977297,\n",
" 'index': 201,\n",
" 'word': '##rra',\n",
" 'start': 824,\n",
" 'end': 827},\n",
" {'entity': 'en',\n",
" 'score': 0.9997421,\n",
" 'index': 202,\n",
" 'word': '##ct',\n",
" 'start': 827,\n",
" 'end': 829},\n",
" {'entity': 'en',\n",
" 'score': 0.9997882,\n",
" 'index': 203,\n",
" 'word': 'attention',\n",
" 'start': 830,\n",
" 'end': 839},\n",
" {'entity': 'en',\n",
" 'score': 0.99970275,\n",
" 'index': 204,\n",
" 'word': 'to',\n",
" 'start': 840,\n",
" 'end': 842},\n",
" {'entity': 'ne',\n",
" 'score': 0.9933374,\n",
" 'index': 205,\n",
" 'word': 'Mars',\n",
" 'start': 843,\n",
" 'end': 847},\n",
" {'entity': 'other',\n",
" 'score': 0.99984145,\n",
" 'index': 206,\n",
" 'word': '-',\n",
" 'start': 847,\n",
" 'end': 848},\n",
" {'entity': 'other',\n",
" 'score': 0.9995536,\n",
" 'index': 207,\n",
" 'word': '-',\n",
" 'start': 848,\n",
" 'end': 849},\n",
" {'entity': 'en',\n",
" 'score': 0.999739,\n",
" 'index': 208,\n",
" 'word': 'and',\n",
" 'start': 850,\n",
" 'end': 853},\n",
" {'entity': 'en',\n",
" 'score': 0.9997385,\n",
" 'index': 209,\n",
" 'word': 'it',\n",
" 'start': 854,\n",
" 'end': 856},\n",
" {'entity': 'en',\n",
" 'score': 0.9997167,\n",
" 'index': 210,\n",
" 'word': 'did',\n",
" 'start': 857,\n",
" 'end': 860},\n",
" {'entity': 'other',\n",
" 'score': 0.99991965,\n",
" 'index': 211,\n",
" 'word': '.',\n",
" 'start': 860,\n",
" 'end': 861},\n",
" {'entity': 'en',\n",
" 'score': 0.9994844,\n",
" 'index': 212,\n",
" 'word': 'The',\n",
" 'start': 863,\n",
" 'end': 866},\n",
" {'entity': 'en',\n",
" 'score': 0.9995809,\n",
" 'index': 213,\n",
" 'word': 'face',\n",
" 'start': 867,\n",
" 'end': 871},\n",
" {'entity': 'en',\n",
" 'score': 0.9993393,\n",
" 'index': 214,\n",
" 'word': 'on',\n",
" 'start': 872,\n",
" 'end': 874},\n",
" {'entity': 'ne',\n",
" 'score': 0.9903474,\n",
" 'index': 215,\n",
" 'word': 'Mars',\n",
" 'start': 875,\n",
" 'end': 879},\n",
" {'entity': 'en',\n",
" 'score': 0.99960655,\n",
" 'index': 216,\n",
" 'word': 'soon',\n",
" 'start': 880,\n",
" 'end': 884},\n",
" {'entity': 'en',\n",
" 'score': 0.9992306,\n",
" 'index': 217,\n",
" 'word': 'became',\n",
" 'start': 885,\n",
" 'end': 891},\n",
" {'entity': 'en',\n",
" 'score': 0.9976654,\n",
" 'index': 218,\n",
" 'word': 'a',\n",
" 'start': 892,\n",
" 'end': 893},\n",
" {'entity': 'en',\n",
" 'score': 0.99933064,\n",
" 'index': 219,\n",
" 'word': 'pop',\n",
" 'start': 894,\n",
" 'end': 897},\n",
" {'entity': 'en',\n",
" 'score': 0.99950266,\n",
" 'index': 220,\n",
" 'word': 'i',\n",
" 'start': 898,\n",
" 'end': 899},\n",
" {'entity': 'en',\n",
" 'score': 0.99954295,\n",
" 'index': 221,\n",
" 'word': '##con',\n",
" 'start': 899,\n",
" 'end': 902},\n",
" {'entity': 'other',\n",
" 'score': 0.99992657,\n",
" 'index': 222,\n",
" 'word': ';',\n",
" 'start': 902,\n",
" 'end': 903},\n",
" {'entity': 'en',\n",
" 'score': 0.9994814,\n",
" 'index': 223,\n",
" 'word': 'shot',\n",
" 'start': 904,\n",
" 'end': 908},\n",
" {'entity': 'en',\n",
" 'score': 0.9994511,\n",
" 'index': 224,\n",
" 'word': 'in',\n",
" 'start': 909,\n",
" 'end': 911},\n",
" {'entity': 'en',\n",
" 'score': 0.9992155,\n",
" 'index': 225,\n",
" 'word': 'movies',\n",
" 'start': 912,\n",
" 'end': 918},\n",
" {'entity': 'other',\n",
" 'score': 0.99992335,\n",
" 'index': 226,\n",
" 'word': ',',\n",
" 'start': 918,\n",
" 'end': 919},\n",
" {'entity': 'en',\n",
" 'score': 0.9993299,\n",
" 'index': 227,\n",
" 'word': 'appeared',\n",
" 'start': 920,\n",
" 'end': 928},\n",
" {'entity': 'en',\n",
" 'score': 0.9994456,\n",
" 'index': 228,\n",
" 'word': 'in',\n",
" 'start': 929,\n",
" 'end': 931},\n",
" {'entity': 'en',\n",
" 'score': 0.9991399,\n",
" 'index': 229,\n",
" 'word': 'books',\n",
" 'start': 932,\n",
" 'end': 937},\n",
" {'entity': 'other',\n",
" 'score': 0.9999262,\n",
" 'index': 230,\n",
" 'word': ',',\n",
" 'start': 937,\n",
" 'end': 938},\n",
" {'entity': 'en',\n",
" 'score': 0.99925786,\n",
" 'index': 231,\n",
" 'word': 'magazines',\n",
" 'start': 939,\n",
" 'end': 948},\n",
" {'entity': 'other',\n",
" 'score': 0.99992716,\n",
" 'index': 232,\n",
" 'word': ',',\n",
" 'start': 948,\n",
" 'end': 949},\n",
" {'entity': 'en',\n",
" 'score': 0.99942744,\n",
" 'index': 233,\n",
" 'word': 'radio',\n",
" 'start': 950,\n",
" 'end': 955},\n",
" {'entity': 'en',\n",
" 'score': 0.9996457,\n",
" 'index': 234,\n",
" 'word': 'talk',\n",
" 'start': 956,\n",
" 'end': 960},\n",
" {'entity': 'en',\n",
" 'score': 0.9996043,\n",
" 'index': 235,\n",
" 'word': 'shows',\n",
" 'start': 961,\n",
" 'end': 966},\n",
" {'entity': 'other',\n",
" 'score': 0.9999255,\n",
" 'index': 236,\n",
" 'word': ',',\n",
" 'start': 966,\n",
" 'end': 967},\n",
" {'entity': 'en',\n",
" 'score': 0.99946564,\n",
" 'index': 237,\n",
" 'word': 'and',\n",
" 'start': 968,\n",
" 'end': 971},\n",
" {'entity': 'en',\n",
" 'score': 0.9997336,\n",
" 'index': 238,\n",
" 'word': 'hau',\n",
" 'start': 972,\n",
" 'end': 975},\n",
" {'entity': 'en',\n",
" 'score': 0.99972874,\n",
" 'index': 239,\n",
" 'word': '##nted',\n",
" 'start': 975,\n",
" 'end': 979},\n",
" {'entity': 'en',\n",
" 'score': 0.9996462,\n",
" 'index': 240,\n",
" 'word': 'gr',\n",
" 'start': 980,\n",
" 'end': 982},\n",
" {'entity': 'en',\n",
" 'score': 0.9997458,\n",
" 'index': 241,\n",
" 'word': '##oce',\n",
" 'start': 982,\n",
" 'end': 985},\n",
" {'entity': 'en',\n",
" 'score': 0.9996829,\n",
" 'index': 242,\n",
" 'word': '##ry',\n",
" 'start': 985,\n",
" 'end': 987},\n",
" {'entity': 'en',\n",
" 'score': 0.99970573,\n",
" 'index': 243,\n",
" 'word': 'store',\n",
" 'start': 988,\n",
" 'end': 993},\n",
" {'entity': 'en',\n",
" 'score': 0.9997999,\n",
" 'index': 244,\n",
" 'word': 'check',\n",
" 'start': 994,\n",
" 'end': 999},\n",
" {'entity': 'en',\n",
" 'score': 0.9997851,\n",
" 'index': 245,\n",
" 'word': '##out',\n",
" 'start': 999,\n",
" 'end': 1002},\n",
" {'entity': 'en',\n",
" 'score': 0.9997584,\n",
" 'index': 246,\n",
" 'word': 'lines',\n",
" 'start': 1003,\n",
" 'end': 1008},\n",
" {'entity': 'en',\n",
" 'score': 0.99963033,\n",
" 'index': 247,\n",
" 'word': 'for',\n",
" 'start': 1009,\n",
" 'end': 1012},\n",
" {'entity': 'other',\n",
" 'score': 0.90661347,\n",
" 'index': 248,\n",
" 'word': '25',\n",
" 'start': 1013,\n",
" 'end': 1015},\n",
" {'entity': 'en',\n",
" 'score': 0.99967,\n",
" 'index': 249,\n",
" 'word': 'years',\n",
" 'start': 1016,\n",
" 'end': 1021},\n",
" {'entity': 'other',\n",
" 'score': 0.99992573,\n",
" 'index': 250,\n",
" 'word': '.',\n",
" 'start': 1021,\n",
" 'end': 1022},\n",
" {'entity': 'en',\n",
" 'score': 0.9997687,\n",
" 'index': 251,\n",
" 'word': 'Some',\n",
" 'start': 1023,\n",
" 'end': 1027},\n",
" {'entity': 'en',\n",
" 'score': 0.9997943,\n",
" 'index': 252,\n",
" 'word': 'people',\n",
" 'start': 1028,\n",
" 'end': 1034},\n",
" {'entity': 'en',\n",
" 'score': 0.9997925,\n",
" 'index': 253,\n",
" 'word': 'thought',\n",
" 'start': 1035,\n",
" 'end': 1042},\n",
" {'entity': 'en',\n",
" 'score': 0.9996296,\n",
" 'index': 254,\n",
" 'word': 'the',\n",
" 'start': 1043,\n",
" 'end': 1046},\n",
" {'entity': 'en',\n",
" 'score': 0.99950707,\n",
" 'index': 255,\n",
" 'word': 'natural',\n",
" 'start': 1047,\n",
" 'end': 1054},\n",
" {'entity': 'en',\n",
" 'score': 0.9997507,\n",
" 'index': 256,\n",
" 'word': 'land',\n",
" 'start': 1055,\n",
" 'end': 1059},\n",
" {'entity': 'en',\n",
" 'score': 0.99979764,\n",
" 'index': 257,\n",
" 'word': '##form',\n",
" 'start': 1059,\n",
" 'end': 1063},\n",
" {'entity': 'en',\n",
" 'score': 0.99975556,\n",
" 'index': 258,\n",
" 'word': 'was',\n",
" 'start': 1064,\n",
" 'end': 1067},\n",
" {'entity': 'en',\n",
" 'score': 0.99978846,\n",
" 'index': 259,\n",
" 'word': 'evidence',\n",
" 'start': 1068,\n",
" 'end': 1076},\n",
" {'entity': 'en',\n",
" 'score': 0.9998209,\n",
" 'index': 260,\n",
" 'word': 'of',\n",
" 'start': 1077,\n",
" 'end': 1079},\n",
" {'entity': 'en',\n",
" 'score': 0.99982053,\n",
" 'index': 261,\n",
" 'word': 'life',\n",
" 'start': 1080,\n",
" 'end': 1084},\n",
" {'entity': 'en',\n",
" 'score': 0.999806,\n",
" 'index': 262,\n",
" 'word': 'on',\n",
" 'start': 1085,\n",
" 'end': 1087},\n",
" {'entity': 'ne',\n",
" 'score': 0.9924434,\n",
" 'index': 263,\n",
" 'word': 'Mars',\n",
" 'start': 1088,\n",
" 'end': 1092},\n",
" {'entity': 'other',\n",
" 'score': 0.9999304,\n",
" 'index': 264,\n",
" 'word': ',',\n",
" 'start': 1092,\n",
" 'end': 1093},\n",
" {'entity': 'en',\n",
" 'score': 0.9998184,\n",
" 'index': 265,\n",
" 'word': 'and',\n",
" 'start': 1094,\n",
" 'end': 1097},\n",
" {'entity': 'en',\n",
" 'score': 0.9998319,\n",
" 'index': 266,\n",
" 'word': 'that',\n",
" 'start': 1098,\n",
" 'end': 1102},\n",
" {'entity': 'en',\n",
" 'score': 0.9995714,\n",
" 'index': 267,\n",
" 'word': 'us',\n",
" 'start': 1103,\n",
" 'end': 1105},\n",
" {'entity': 'en',\n",
" 'score': 0.9997626,\n",
" 'index': 268,\n",
" 'word': 'scientists',\n",
" 'start': 1106,\n",
" 'end': 1116},\n",
" {'entity': 'en',\n",
" 'score': 0.9998241,\n",
" 'index': 269,\n",
" 'word': 'wanted',\n",
" 'start': 1117,\n",
" 'end': 1123},\n",
" {'entity': 'en',\n",
" 'score': 0.9998323,\n",
" 'index': 270,\n",
" 'word': 'to',\n",
" 'start': 1124,\n",
" 'end': 1126},\n",
" {'entity': 'en',\n",
" 'score': 0.99986374,\n",
" 'index': 271,\n",
" 'word': 'hide',\n",
" 'start': 1127,\n",
" 'end': 1131},\n",
" {'entity': 'en',\n",
" 'score': 0.99984705,\n",
" 'index': 272,\n",
" 'word': 'it',\n",
" 'start': 1132,\n",
" 'end': 1134},\n",
" {'entity': 'other',\n",
" 'score': 0.99992883,\n",
" 'index': 273,\n",
" 'word': ',',\n",
" 'start': 1134,\n",
" 'end': 1135},\n",
" {'entity': 'en',\n",
" 'score': 0.99986625,\n",
" 'index': 274,\n",
" 'word': 'but',\n",
" 'start': 1136,\n",
" 'end': 1139},\n",
" {'entity': 'en',\n",
" 'score': 0.99982065,\n",
" 'index': 275,\n",
" 'word': 'really',\n",
" 'start': 1140,\n",
" 'end': 1146},\n",
" {'entity': 'other',\n",
" 'score': 0.9999231,\n",
" 'index': 276,\n",
" 'word': ',',\n",
" 'start': 1146,\n",
" 'end': 1147},\n",
" {'entity': 'en',\n",
" 'score': 0.99983907,\n",
" 'index': 277,\n",
" 'word': 'the',\n",
" 'start': 1148,\n",
" 'end': 1151},\n",
" {'entity': 'en',\n",
" 'score': 0.9998418,\n",
" 'index': 278,\n",
" 'word': 'defender',\n",
" 'start': 1152,\n",
" 'end': 1160},\n",
" {'entity': 'en',\n",
" 'score': 0.9998036,\n",
" 'index': 279,\n",
" 'word': '##s',\n",
" 'start': 1160,\n",
" 'end': 1161},\n",
" {'entity': 'en',\n",
" 'score': 0.9998399,\n",
" 'index': 280,\n",
" 'word': 'of',\n",
" 'start': 1162,\n",
" 'end': 1164},\n",
" {'entity': 'en',\n",
" 'score': 0.9997856,\n",
" 'index': 281,\n",
" 'word': 'the',\n",
" 'start': 1165,\n",
" 'end': 1168},\n",
" {'entity': 'ne',\n",
" 'score': 0.9955486,\n",
" 'index': 282,\n",
" 'word': 'NASA',\n",
" 'start': 1169,\n",
" 'end': 1173},\n",
" {'entity': 'en',\n",
" 'score': 0.9998142,\n",
" 'index': 283,\n",
" 'word': 'budget',\n",
" 'start': 1174,\n",
" 'end': 1180},\n",
" {'entity': 'en',\n",
" 'score': 0.9998621,\n",
" 'index': 284,\n",
" 'word': 'wish',\n",
" 'start': 1181,\n",
" 'end': 1185},\n",
" {'entity': 'en',\n",
" 'score': 0.99985635,\n",
" 'index': 285,\n",
" 'word': 'there',\n",
" 'start': 1186,\n",
" 'end': 1191},\n",
" {'entity': 'en',\n",
" 'score': 0.9998217,\n",
" 'index': 286,\n",
" 'word': 'was',\n",
" 'start': 1192,\n",
" 'end': 1195},\n",
" {'entity': 'en',\n",
" 'score': 0.9998242,\n",
" 'index': 287,\n",
" 'word': 'ancient',\n",
" 'start': 1196,\n",
" 'end': 1203},\n",
" {'entity': 'en',\n",
" 'score': 0.9998467,\n",
" 'index': 288,\n",
" 'word': 'civili',\n",
" 'start': 1204,\n",
" 'end': 1210},\n",
" {'entity': 'en',\n",
" 'score': 0.9998492,\n",
" 'index': 289,\n",
" 'word': '##zation',\n",
" 'start': 1210,\n",
" 'end': 1216},\n",
" {'entity': 'en',\n",
" 'score': 0.99985075,\n",
" 'index': 290,\n",
" 'word': 'on',\n",
" 'start': 1217,\n",
" 'end': 1219},\n",
" {'entity': 'ne',\n",
" 'score': 0.9934476,\n",
" 'index': 291,\n",
" 'word': 'Mars',\n",
" 'start': 1220,\n",
" 'end': 1224},\n",
" {'entity': 'other',\n",
" 'score': 0.9999279,\n",
" 'index': 292,\n",
" 'word': '.',\n",
" 'start': 1224,\n",
" 'end': 1225},\n",
" {'entity': 'en',\n",
" 'score': 0.9996929,\n",
" 'index': 293,\n",
" 'word': 'We',\n",
" 'start': 1226,\n",
" 'end': 1228},\n",
" {'entity': 'en',\n",
" 'score': 0.9997583,\n",
" 'index': 294,\n",
" 'word': 'decided',\n",
" 'start': 1229,\n",
" 'end': 1236},\n",
" {'entity': 'en',\n",
" 'score': 0.99972254,\n",
" 'index': 295,\n",
" 'word': 'to',\n",
" 'start': 1237,\n",
" 'end': 1239},\n",
" {'entity': 'en',\n",
" 'score': 0.99963474,\n",
" 'index': 296,\n",
" 'word': 'take',\n",
" 'start': 1240,\n",
" 'end': 1244},\n",
" {'entity': 'en',\n",
" 'score': 0.99941206,\n",
" 'index': 297,\n",
" 'word': 'another',\n",
" 'start': 1245,\n",
" 'end': 1252},\n",
" {'entity': 'en',\n",
" 'score': 0.99968994,\n",
" 'index': 298,\n",
" 'word': 'shot',\n",
" 'start': 1253,\n",
" 'end': 1257},\n",
" {'entity': 'en',\n",
" 'score': 0.9997118,\n",
" 'index': 299,\n",
" 'word': 'just',\n",
" 'start': 1258,\n",
" 'end': 1262},\n",
" {'entity': 'en',\n",
" 'score': 0.9997631,\n",
" 'index': 300,\n",
" 'word': 'to',\n",
" 'start': 1263,\n",
" 'end': 1265},\n",
" {'entity': 'en',\n",
" 'score': 0.99978083,\n",
" 'index': 301,\n",
" 'word': 'make',\n",
" 'start': 1266,\n",
" 'end': 1270},\n",
" {'entity': 'en',\n",
" 'score': 0.9998004,\n",
" 'index': 302,\n",
" 'word': 'sure',\n",
" 'start': 1271,\n",
" 'end': 1275},\n",
" {'entity': 'en',\n",
" 'score': 0.99981683,\n",
" 'index': 303,\n",
" 'word': 'we',\n",
" 'start': 1276,\n",
" 'end': 1278},\n",
" {'entity': 'en',\n",
" 'score': 0.99981815,\n",
" 'index': 304,\n",
" 'word': 'were',\n",
" 'start': 1279,\n",
" 'end': 1283},\n",
" {'entity': 'en',\n",
" 'score': 0.99982005,\n",
" 'index': 305,\n",
" 'word': '##n',\n",
" 'start': 1283,\n",
" 'end': 1284},\n",
" {'entity': 'en',\n",
" 'score': 0.9997904,\n",
" 'index': 306,\n",
" 'word': \"'\",\n",
" 'start': 1284,\n",
" 'end': 1285},\n",
" {'entity': 'en',\n",
" 'score': 0.999744,\n",
" 'index': 307,\n",
" 'word': 't',\n",
" 'start': 1285,\n",
" 'end': 1286},\n",
" {'entity': 'en',\n",
" 'score': 0.9998004,\n",
" 'index': 308,\n",
" 'word': 'wrong',\n",
" 'start': 1287,\n",
" 'end': 1292},\n",
" {'entity': 'other',\n",
" 'score': 0.99991894,\n",
" 'index': 309,\n",
" 'word': ',',\n",
" 'start': 1292,\n",
" 'end': 1293},\n",
" {'entity': 'en',\n",
" 'score': 0.9986118,\n",
" 'index': 310,\n",
" 'word': 'on',\n",
" 'start': 1294,\n",
" 'end': 1296},\n",
" {'entity': 'other',\n",
" 'score': 0.53777504,\n",
" 'index': 311,\n",
" 'word': 'April',\n",
" 'start': 1297,\n",
" 'end': 1302},\n",
" {'entity': 'other',\n",
" 'score': 0.9973557,\n",
" 'index': 312,\n",
" 'word': '5',\n",
" 'start': 1303,\n",
" 'end': 1304},\n",
" {'entity': 'other',\n",
" 'score': 0.9998646,\n",
" 'index': 313,\n",
" 'word': ',',\n",
" 'start': 1304,\n",
" 'end': 1305},\n",
" {'entity': 'other',\n",
" 'score': 0.99943346,\n",
" 'index': 314,\n",
" 'word': '1998',\n",
" 'start': 1306,\n",
" 'end': 1310},\n",
" {'entity': 'other',\n",
" 'score': 0.9999175,\n",
" 'index': 315,\n",
" 'word': '.',\n",
" 'start': 1310,\n",
" 'end': 1311},\n",
" {'entity': 'ne',\n",
" 'score': 0.99906963,\n",
" 'index': 316,\n",
" 'word': 'Michael',\n",
" 'start': 1312,\n",
" 'end': 1319},\n",
" {'entity': 'ne',\n",
" 'score': 0.9983594,\n",
" 'index': 317,\n",
" 'word': 'Mali',\n",
" 'start': 1320,\n",
" 'end': 1324},\n",
" {'entity': 'ne',\n",
" 'score': 0.9985464,\n",
" 'index': 318,\n",
" 'word': '##n',\n",
" 'start': 1324,\n",
" 'end': 1325},\n",
" {'entity': 'en',\n",
" 'score': 0.99965477,\n",
" 'index': 319,\n",
" 'word': 'and',\n",
" 'start': 1326,\n",
" 'end': 1329},\n",
" {'entity': 'en',\n",
" 'score': 0.9990552,\n",
" 'index': 320,\n",
" 'word': 'his',\n",
" 'start': 1330,\n",
" 'end': 1333},\n",
" {'entity': 'ne',\n",
" 'score': 0.97999007,\n",
" 'index': 321,\n",
" 'word': 'Mars',\n",
" 'start': 1334,\n",
" 'end': 1338},\n",
" {'entity': 'ne',\n",
" 'score': 0.6277367,\n",
" 'index': 322,\n",
" 'word': 'Or',\n",
" 'start': 1339,\n",
" 'end': 1341},\n",
" {'entity': 'ne',\n",
" 'score': 0.5293862,\n",
" 'index': 323,\n",
" 'word': '##biter',\n",
" 'start': 1341,\n",
" 'end': 1346},\n",
" {'entity': 'en',\n",
" 'score': 0.9995109,\n",
" 'index': 324,\n",
" 'word': 'camera',\n",
" 'start': 1347,\n",
" 'end': 1353},\n",
" {'entity': 'en',\n",
" 'score': 0.9995858,\n",
" 'index': 325,\n",
" 'word': 'team',\n",
" 'start': 1354,\n",
" 'end': 1358},\n",
" {'entity': 'en',\n",
" 'score': 0.99904686,\n",
" 'index': 326,\n",
" 'word': 'took',\n",
" 'start': 1359,\n",
" 'end': 1363},\n",
" {'entity': 'en',\n",
" 'score': 0.99744225,\n",
" 'index': 327,\n",
" 'word': 'a',\n",
" 'start': 1364,\n",
" 'end': 1365},\n",
" {'entity': 'en',\n",
" 'score': 0.99924374,\n",
" 'index': 328,\n",
" 'word': 'picture',\n",
" 'start': 1366,\n",
" 'end': 1373},\n",
" {'entity': 'en',\n",
" 'score': 0.99940455,\n",
" 'index': 329,\n",
" 'word': 'that',\n",
" 'start': 1374,\n",
" 'end': 1378},\n",
" {'entity': 'en',\n",
" 'score': 0.9994288,\n",
" 'index': 330,\n",
" 'word': 'was',\n",
" 'start': 1379,\n",
" 'end': 1382},\n",
" {'entity': 'en',\n",
" 'score': 0.99890625,\n",
" 'index': 331,\n",
" 'word': 'ten',\n",
" 'start': 1383,\n",
" 'end': 1386},\n",
" {'entity': 'en',\n",
" 'score': 0.9993057,\n",
" 'index': 332,\n",
" 'word': 'times',\n",
" 'start': 1387,\n",
" 'end': 1392},\n",
" {'entity': 'en',\n",
" 'score': 0.99956423,\n",
" 'index': 333,\n",
" 'word': 'sharp',\n",
" 'start': 1393,\n",
" 'end': 1398},\n",
" {'entity': 'en',\n",
" 'score': 0.99929535,\n",
" 'index': 334,\n",
" 'word': '##er',\n",
" 'start': 1398,\n",
" 'end': 1400},\n",
" {'entity': 'en',\n",
" 'score': 0.9995135,\n",
" 'index': 335,\n",
" 'word': 'than',\n",
" 'start': 1401,\n",
" 'end': 1405},\n",
" {'entity': 'en',\n",
" 'score': 0.99948967,\n",
" 'index': 336,\n",
" 'word': 'the',\n",
" 'start': 1406,\n",
" 'end': 1409},\n",
" {'entity': 'en',\n",
" 'score': 0.99922657,\n",
" 'index': 337,\n",
" 'word': 'original',\n",
" 'start': 1410,\n",
" 'end': 1418},\n",
" {'entity': 'en',\n",
" 'score': 0.54231477,\n",
" 'index': 338,\n",
" 'word': 'Viking',\n",
" 'start': 1419,\n",
" 'end': 1425},\n",
" {'entity': 'en',\n",
" 'score': 0.9983171,\n",
" 'index': 339,\n",
" 'word': 'photos',\n",
" 'start': 1426,\n",
" 'end': 1432},\n",
" {'entity': 'other',\n",
" 'score': 0.9999213,\n",
" 'index': 340,\n",
" 'word': ',',\n",
" 'start': 1432,\n",
" 'end': 1433},\n",
" {'entity': 'en',\n",
" 'score': 0.9995522,\n",
" 'index': 341,\n",
" 'word': 'reveal',\n",
" 'start': 1434,\n",
" 'end': 1440},\n",
" {'entity': 'en',\n",
" 'score': 0.9996402,\n",
" 'index': 342,\n",
" 'word': '##ing',\n",
" 'start': 1440,\n",
" 'end': 1443},\n",
" {'entity': 'en',\n",
" 'score': 0.9978428,\n",
" 'index': 343,\n",
" 'word': 'a',\n",
" 'start': 1444,\n",
" 'end': 1445},\n",
" {'entity': 'en',\n",
" 'score': 0.99874496,\n",
" 'index': 344,\n",
" 'word': 'natural',\n",
" 'start': 1446,\n",
" 'end': 1453},\n",
" {'entity': 'en',\n",
" 'score': 0.99962974,\n",
" 'index': 345,\n",
" 'word': 'land',\n",
" 'start': 1454,\n",
" 'end': 1458},\n",
" {'entity': 'en',\n",
" 'score': 0.99966383,\n",
" 'index': 346,\n",
" 'word': '##form',\n",
" 'start': 1458,\n",
" 'end': 1462},\n",
" {'entity': 'other',\n",
" 'score': 0.99993026,\n",
" 'index': 347,\n",
" 'word': ',',\n",
" 'start': 1462,\n",
" 'end': 1463},\n",
" {'entity': 'en',\n",
" 'score': 0.99916697,\n",
" 'index': 348,\n",
" 'word': 'which',\n",
" 'start': 1464,\n",
" 'end': 1469},\n",
" {'entity': 'en',\n",
" 'score': 0.9983835,\n",
" 'index': 349,\n",
" 'word': 'meant',\n",
" 'start': 1470,\n",
" 'end': 1475},\n",
" {'entity': 'en',\n",
" 'score': 0.9928202,\n",
" 'index': 350,\n",
" 'word': 'no',\n",
" 'start': 1476,\n",
" 'end': 1478},\n",
" {'entity': 'en',\n",
" 'score': 0.9962263,\n",
" 'index': 351,\n",
" 'word': 'alien',\n",
" 'start': 1479,\n",
" 'end': 1484},\n",
" {'entity': 'en',\n",
" 'score': 0.9987729,\n",
" 'index': 352,\n",
" 'word': 'monument',\n",
" 'start': 1485,\n",
" 'end': 1493},\n",
" {'entity': 'other',\n",
" 'score': 0.9999304,\n",
" 'index': 353,\n",
" 'word': '.',\n",
" 'start': 1493,\n",
" 'end': 1494},\n",
" {'entity': 'other',\n",
" 'score': 0.99991786,\n",
" 'index': 354,\n",
" 'word': '\"',\n",
" 'start': 1495,\n",
" 'end': 1496},\n",
" {'entity': 'en',\n",
" 'score': 0.99973565,\n",
" 'index': 355,\n",
" 'word': 'But',\n",
" 'start': 1496,\n",
" 'end': 1499},\n",
" {'entity': 'en',\n",
" 'score': 0.9994904,\n",
" 'index': 356,\n",
" 'word': 'that',\n",
" 'start': 1500,\n",
" 'end': 1504},\n",
" {'entity': 'en',\n",
" 'score': 0.9995603,\n",
" 'index': 357,\n",
" 'word': 'picture',\n",
" 'start': 1505,\n",
" 'end': 1512},\n",
" {'entity': 'en',\n",
" 'score': 0.99966073,\n",
" 'index': 358,\n",
" 'word': 'wasn',\n",
" 'start': 1513,\n",
" 'end': 1517},\n",
" {'entity': 'en',\n",
" 'score': 0.99970895,\n",
" 'index': 359,\n",
" 'word': \"'\",\n",
" 'start': 1517,\n",
" 'end': 1518},\n",
" {'entity': 'en',\n",
" 'score': 0.99959975,\n",
" 'index': 360,\n",
" 'word': 't',\n",
" 'start': 1518,\n",
" 'end': 1519},\n",
" {'entity': 'en',\n",
" 'score': 0.9995778,\n",
" 'index': 361,\n",
" 'word': 'very',\n",
" 'start': 1520,\n",
" 'end': 1524},\n",
" {'entity': 'en',\n",
" 'score': 0.99971443,\n",
" 'index': 362,\n",
" 'word': 'clear',\n",
" 'start': 1525,\n",
" 'end': 1530},\n",
" {'entity': 'en',\n",
" 'score': 0.9997217,\n",
" 'index': 363,\n",
" 'word': 'at',\n",
" 'start': 1531,\n",
" 'end': 1533},\n",
" {'entity': 'en',\n",
" 'score': 0.99960893,\n",
" 'index': 364,\n",
" 'word': 'all',\n",
" 'start': 1534,\n",
" 'end': 1537},\n",
" {'entity': 'other',\n",
" 'score': 0.99993,\n",
" 'index': 365,\n",
" 'word': ',',\n",
" 'start': 1537,\n",
" 'end': 1538},\n",
" {'entity': 'en',\n",
" 'score': 0.99955565,\n",
" 'index': 366,\n",
" 'word': 'which',\n",
" 'start': 1539,\n",
" 'end': 1544},\n",
" {'entity': 'en',\n",
" 'score': 0.99938047,\n",
" 'index': 367,\n",
" 'word': 'could',\n",
" 'start': 1545,\n",
" 'end': 1550},\n",
" {'entity': 'en',\n",
" 'score': 0.99928325,\n",
" 'index': 368,\n",
" 'word': 'mean',\n",
" 'start': 1551,\n",
" 'end': 1555},\n",
" {'entity': 'en',\n",
" 'score': 0.9982596,\n",
" 'index': 369,\n",
" 'word': 'alien',\n",
" 'start': 1556,\n",
" 'end': 1561},\n",
" {'entity': 'en',\n",
" 'score': 0.99910384,\n",
" 'index': 370,\n",
" 'word': 'marking',\n",
" 'start': 1562,\n",
" 'end': 1569},\n",
" {'entity': 'en',\n",
" 'score': 0.99916995,\n",
" 'index': 371,\n",
" 'word': '##s',\n",
" 'start': 1569,\n",
" 'end': 1570},\n",
" {'entity': 'en',\n",
" 'score': 0.99937785,\n",
" 'index': 372,\n",
" 'word': 'were',\n",
" 'start': 1571,\n",
" 'end': 1575},\n",
" {'entity': 'en',\n",
" 'score': 0.99948174,\n",
" 'index': 373,\n",
" 'word': 'hidden',\n",
" 'start': 1576,\n",
" 'end': 1582},\n",
" {'entity': 'en',\n",
" 'score': 0.99941933,\n",
" 'index': 374,\n",
" 'word': 'by',\n",
" 'start': 1583,\n",
" 'end': 1585},\n",
" {'entity': 'en',\n",
" 'score': 0.99944514,\n",
" 'index': 375,\n",
" 'word': 'ha',\n",
" 'start': 1586,\n",
" 'end': 1588},\n",
" {'entity': 'en',\n",
" 'score': 0.9994628,\n",
" 'index': 376,\n",
" 'word': '##ze',\n",
" 'start': 1588,\n",
" 'end': 1590},\n",
" {'entity': 'other',\n",
" 'score': 0.9999194,\n",
" 'index': 377,\n",
" 'word': '\"',\n",
" 'start': 1590,\n",
" 'end': 1591},\n",
" {'entity': 'en',\n",
" 'score': 0.9996394,\n",
" 'index': 378,\n",
" 'word': 'Well',\n",
" 'start': 1592,\n",
" 'end': 1596},\n",
" {'entity': 'en',\n",
" 'score': 0.99759346,\n",
" 'index': 379,\n",
" 'word': 'no',\n",
" 'start': 1597,\n",
" 'end': 1599},\n",
" {'entity': 'other',\n",
" 'score': 0.99993074,\n",
" 'index': 380,\n",
" 'word': ',',\n",
" 'start': 1599,\n",
" 'end': 1600},\n",
" {'entity': 'en',\n",
" 'score': 0.9995228,\n",
" 'index': 381,\n",
" 'word': 'ye',\n",
" 'start': 1601,\n",
" 'end': 1603},\n",
" {'entity': 'en',\n",
" 'score': 0.99969923,\n",
" 'index': 382,\n",
" 'word': '##s',\n",
" 'start': 1603,\n",
" 'end': 1604},\n",
" {'entity': 'en',\n",
" 'score': 0.99970835,\n",
" 'index': 383,\n",
" 'word': 'that',\n",
" 'start': 1605,\n",
" 'end': 1609},\n",
" {'entity': 'en',\n",
" 'score': 0.999665,\n",
" 'index': 384,\n",
" 'word': 'rum',\n",
" 'start': 1610,\n",
" 'end': 1613},\n",
" {'entity': 'en',\n",
" 'score': 0.99970514,\n",
" 'index': 385,\n",
" 'word': '##or',\n",
" 'start': 1613,\n",
" 'end': 1615},\n",
" {'entity': 'en',\n",
" 'score': 0.99943703,\n",
" 'index': 386,\n",
" 'word': 'started',\n",
" 'start': 1616,\n",
" 'end': 1623},\n",
" {'entity': 'other',\n",
" 'score': 0.9999294,\n",
" 'index': 387,\n",
" 'word': ',',\n",
" 'start': 1623,\n",
" 'end': 1624},\n",
" {'entity': 'en',\n",
" 'score': 0.99961936,\n",
" 'index': 388,\n",
" 'word': 'but',\n",
" 'start': 1625,\n",
" 'end': 1628},\n",
" {'entity': 'en',\n",
" 'score': 0.99959284,\n",
" 'index': 389,\n",
" 'word': 'to',\n",
" 'start': 1629,\n",
" 'end': 1631},\n",
" {'entity': 'en',\n",
" 'score': 0.99967396,\n",
" 'index': 390,\n",
" 'word': 'prove',\n",
" 'start': 1632,\n",
" 'end': 1637},\n",
" {'entity': 'en',\n",
" 'score': 0.9996499,\n",
" 'index': 391,\n",
" 'word': 'them',\n",
" 'start': 1638,\n",
" 'end': 1642},\n",
" {'entity': 'en',\n",
" 'score': 0.9996276,\n",
" 'index': 392,\n",
" 'word': 'wrong',\n",
" 'start': 1643,\n",
" 'end': 1648},\n",
" {'entity': 'en',\n",
" 'score': 0.99944097,\n",
" 'index': 393,\n",
" 'word': 'on',\n",
" 'start': 1649,\n",
" 'end': 1651},\n",
" {'entity': 'en',\n",
" 'score': 0.61485404,\n",
" 'index': 394,\n",
" 'word': 'April',\n",
" 'start': 1652,\n",
" 'end': 1657},\n",
" {'entity': 'other',\n",
" 'score': 0.99845207,\n",
" 'index': 395,\n",
" 'word': '8',\n",
" 'start': 1658,\n",
" 'end': 1659},\n",
" {'entity': 'other',\n",
" 'score': 0.9998248,\n",
" 'index': 396,\n",
" 'word': ',',\n",
" 'start': 1659,\n",
" 'end': 1660},\n",
" {'entity': 'other',\n",
" 'score': 0.967133,\n",
" 'index': 397,\n",
" 'word': '2001',\n",
" 'start': 1661,\n",
" 'end': 1665},\n",
" {'entity': 'en',\n",
" 'score': 0.99961036,\n",
" 'index': 398,\n",
" 'word': 'we',\n",
" 'start': 1666,\n",
" 'end': 1668},\n",
" {'entity': 'en',\n",
" 'score': 0.9996408,\n",
" 'index': 399,\n",
" 'word': 'decided',\n",
" 'start': 1669,\n",
" 'end': 1676},\n",
" {'entity': 'en',\n",
" 'score': 0.99961096,\n",
" 'index': 400,\n",
" 'word': 'to',\n",
" 'start': 1677,\n",
" 'end': 1679},\n",
" {'entity': 'en',\n",
" 'score': 0.9994717,\n",
" 'index': 401,\n",
" 'word': 'take',\n",
" 'start': 1680,\n",
" 'end': 1684},\n",
" {'entity': 'en',\n",
" 'score': 0.99907696,\n",
" 'index': 402,\n",
" 'word': 'another',\n",
" 'start': 1685,\n",
" 'end': 1692},\n",
" {'entity': 'en',\n",
" 'score': 0.9993956,\n",
" 'index': 403,\n",
" 'word': 'picture',\n",
" 'start': 1693,\n",
" 'end': 1700},\n",
" {'entity': 'other',\n",
" 'score': 0.9999089,\n",
" 'index': 404,\n",
" 'word': ',',\n",
" 'start': 1700,\n",
" 'end': 1701},\n",
" {'entity': 'en',\n",
" 'score': 0.9996431,\n",
" 'index': 405,\n",
" 'word': 'making',\n",
" 'start': 1702,\n",
" 'end': 1708},\n",
" {'entity': 'en',\n",
" 'score': 0.9997148,\n",
" 'index': 406,\n",
" 'word': 'sure',\n",
" 'start': 1709,\n",
" 'end': 1713},\n",
" {'entity': 'en',\n",
" 'score': 0.9997253,\n",
" 'index': 407,\n",
" 'word': 'it',\n",
" 'start': 1714,\n",
" 'end': 1716},\n",
" {'entity': 'en',\n",
" 'score': 0.99960643,\n",
" 'index': 408,\n",
" 'word': 'was',\n",
" 'start': 1717,\n",
" 'end': 1720},\n",
" {'entity': 'en',\n",
" 'score': 0.9989556,\n",
" 'index': 409,\n",
" 'word': 'a',\n",
" 'start': 1721,\n",
" 'end': 1722},\n",
" {'entity': 'en',\n",
" 'score': 0.99975353,\n",
" 'index': 410,\n",
" 'word': 'cloud',\n",
" 'start': 1723,\n",
" 'end': 1728},\n",
" {'entity': 'en',\n",
" 'score': 0.999777,\n",
" 'index': 411,\n",
" 'word': '##less',\n",
" 'start': 1728,\n",
" 'end': 1732},\n",
" {'entity': 'en',\n",
" 'score': 0.9997631,\n",
" 'index': 412,\n",
" 'word': 'summer',\n",
" 'start': 1733,\n",
" 'end': 1739},\n",
" {'entity': 'en',\n",
" 'score': 0.99969566,\n",
" 'index': 413,\n",
" 'word': 'day',\n",
" 'start': 1740,\n",
" 'end': 1743},\n",
" {'entity': 'other',\n",
" 'score': 0.9999254,\n",
" 'index': 414,\n",
" 'word': '.',\n",
" 'start': 1743,\n",
" 'end': 1744},\n",
" {'entity': 'ne',\n",
" 'score': 0.9956275,\n",
" 'index': 415,\n",
" 'word': 'Mali',\n",
" 'start': 1745,\n",
" 'end': 1749},\n",
" {'entity': 'ne',\n",
" 'score': 0.9934854,\n",
" 'index': 416,\n",
" 'word': '##n',\n",
" 'start': 1749,\n",
" 'end': 1750},\n",
" {'entity': 'en',\n",
" 'score': 0.9940276,\n",
" 'index': 417,\n",
" 'word': \"'\",\n",
" 'start': 1750,\n",
" 'end': 1751},\n",
" {'entity': 'en',\n",
" 'score': 0.9994653,\n",
" 'index': 418,\n",
" 'word': 's',\n",
" 'start': 1751,\n",
" 'end': 1752},\n",
" {'entity': 'en',\n",
" 'score': 0.99972516,\n",
" 'index': 419,\n",
" 'word': 'team',\n",
" 'start': 1753,\n",
" 'end': 1757},\n",
" {'entity': 'en',\n",
" 'score': 0.9996685,\n",
" 'index': 420,\n",
" 'word': 'captured',\n",
" 'start': 1758,\n",
" 'end': 1766},\n",
" {'entity': 'en',\n",
" 'score': 0.99941444,\n",
" 'index': 421,\n",
" 'word': 'an',\n",
" 'start': 1767,\n",
" 'end': 1769},\n",
" {'entity': 'en',\n",
" 'score': 0.99968505,\n",
" 'index': 422,\n",
" 'word': 'ama',\n",
" 'start': 1770,\n",
" 'end': 1773},\n",
" {'entity': 'en',\n",
" 'score': 0.99972624,\n",
" 'index': 423,\n",
" 'word': '##zing',\n",
" 'start': 1773,\n",
" 'end': 1777},\n",
" {'entity': 'en',\n",
" 'score': 0.9996338,\n",
" 'index': 424,\n",
" 'word': 'photo',\n",
" 'start': 1778,\n",
" 'end': 1783},\n",
" {'entity': 'en',\n",
" 'score': 0.9996408,\n",
" 'index': 425,\n",
" 'word': 'using',\n",
" 'start': 1784,\n",
" 'end': 1789},\n",
" {'entity': 'en',\n",
" 'score': 0.9994491,\n",
" 'index': 426,\n",
" 'word': 'the',\n",
" 'start': 1790,\n",
" 'end': 1793},\n",
" {'entity': 'en',\n",
" 'score': 0.99961936,\n",
" 'index': 427,\n",
" 'word': 'camera',\n",
" 'start': 1794,\n",
" 'end': 1800},\n",
" {'entity': 'en',\n",
" 'score': 0.9996668,\n",
" 'index': 428,\n",
" 'word': \"'\",\n",
" 'start': 1800,\n",
" 'end': 1801},\n",
" {'entity': 'en',\n",
" 'score': 0.9995703,\n",
" 'index': 429,\n",
" 'word': 's',\n",
" 'start': 1801,\n",
" 'end': 1802},\n",
" {'entity': 'en',\n",
" 'score': 0.99967444,\n",
" 'index': 430,\n",
" 'word': 'absolute',\n",
" 'start': 1803,\n",
" 'end': 1811},\n",
" {'entity': 'en',\n",
" 'score': 0.99970645,\n",
" 'index': 431,\n",
" 'word': 'maximum',\n",
" 'start': 1812,\n",
" 'end': 1819},\n",
" {'entity': 'en',\n",
" 'score': 0.9996973,\n",
" 'index': 432,\n",
" 'word': 'revolution',\n",
" 'start': 1820,\n",
" 'end': 1830},\n",
" {'entity': 'other',\n",
" 'score': 0.99992955,\n",
" 'index': 433,\n",
" 'word': '.',\n",
" 'start': 1830,\n",
" 'end': 1831},\n",
" {'entity': 'en',\n",
" 'score': 0.9997081,\n",
" 'index': 434,\n",
" 'word': 'With',\n",
" 'start': 1832,\n",
" 'end': 1836},\n",
" {'entity': 'en',\n",
" 'score': 0.99963665,\n",
" 'index': 435,\n",
" 'word': 'this',\n",
" 'start': 1837,\n",
" 'end': 1841},\n",
" {'entity': 'en',\n",
" 'score': 0.99965286,\n",
" 'index': 436,\n",
" 'word': 'camera',\n",
" 'start': 1842,\n",
" 'end': 1848},\n",
" {'entity': 'en',\n",
" 'score': 0.9997528,\n",
" 'index': 437,\n",
" 'word': 'you',\n",
" 'start': 1849,\n",
" 'end': 1852},\n",
" {'entity': 'en',\n",
" 'score': 0.9997576,\n",
" 'index': 438,\n",
" 'word': 'can',\n",
" 'start': 1853,\n",
" 'end': 1856},\n",
" {'entity': 'en',\n",
" 'score': 0.9997657,\n",
" 'index': 439,\n",
" 'word': 'disc',\n",
" 'start': 1857,\n",
" 'end': 1861},\n",
" {'entity': 'en',\n",
" 'score': 0.9997956,\n",
" 'index': 440,\n",
" 'word': '##ern',\n",
" 'start': 1861,\n",
" 'end': 1864},\n",
" {'entity': 'en',\n",
" 'score': 0.9997335,\n",
" 'index': 441,\n",
" 'word': 'things',\n",
" 'start': 1865,\n",
" 'end': 1871},\n",
" {'entity': 'en',\n",
" 'score': 0.999653,\n",
" 'index': 442,\n",
" 'word': 'in',\n",
" 'start': 1872,\n",
" 'end': 1874},\n",
" {'entity': 'en',\n",
" 'score': 0.9991596,\n",
" 'index': 443,\n",
" 'word': 'a',\n",
" 'start': 1875,\n",
" 'end': 1876},\n",
" {'entity': 'en',\n",
" 'score': 0.99945194,\n",
" 'index': 444,\n",
" 'word': 'digital',\n",
" 'start': 1877,\n",
" 'end': 1884},\n",
" {'entity': 'en',\n",
" 'score': 0.9995933,\n",
" 'index': 445,\n",
" 'word': 'image',\n",
" 'start': 1885,\n",
" 'end': 1890},\n",
" {'entity': 'other',\n",
" 'score': 0.9999232,\n",
" 'index': 446,\n",
" 'word': ',',\n",
" 'start': 1890,\n",
" 'end': 1891},\n",
" {'entity': 'other',\n",
" 'score': 0.99489,\n",
" 'index': 447,\n",
" 'word': '3',\n",
" 'start': 1892,\n",
" 'end': 1893},\n",
" {'entity': 'en',\n",
" 'score': 0.9991105,\n",
" 'index': 448,\n",
" 'word': 'times',\n",
" 'start': 1894,\n",
" 'end': 1899},\n",
" {'entity': 'en',\n",
" 'score': 0.9994937,\n",
" 'index': 449,\n",
" 'word': 'bigger',\n",
" 'start': 1900,\n",
" 'end': 1906},\n",
" {'entity': 'en',\n",
" 'score': 0.9995734,\n",
" 'index': 450,\n",
" 'word': 'than',\n",
" 'start': 1907,\n",
" 'end': 1911},\n",
" {'entity': 'en',\n",
" 'score': 0.9995715,\n",
" 'index': 451,\n",
" 'word': 'the',\n",
" 'start': 1912,\n",
" 'end': 1915},\n",
" {'entity': 'en',\n",
" 'score': 0.9996369,\n",
" 'index': 452,\n",
" 'word': 'pi',\n",
" 'start': 1916,\n",
" 'end': 1918},\n",
" {'entity': 'en',\n",
" 'score': 0.9996636,\n",
" 'index': 453,\n",
" 'word': '##xel',\n",
" 'start': 1918,\n",
" 'end': 1921},\n",
" {'entity': 'en',\n",
" 'score': 0.99961394,\n",
" 'index': 454,\n",
" 'word': 'size',\n",
" 'start': 1922,\n",
" 'end': 1926},\n",
" {'entity': 'en',\n",
" 'score': 0.9996786,\n",
" 'index': 455,\n",
" 'word': 'which',\n",
" 'start': 1927,\n",
" 'end': 1932},\n",
" {'entity': 'en',\n",
" 'score': 0.9996667,\n",
" 'index': 456,\n",
" 'word': 'means',\n",
" 'start': 1933,\n",
" 'end': 1938},\n",
" {'entity': 'en',\n",
" 'score': 0.9997652,\n",
" 'index': 457,\n",
" 'word': 'if',\n",
" 'start': 1939,\n",
" 'end': 1941},\n",
" {'entity': 'en',\n",
" 'score': 0.99967515,\n",
" 'index': 458,\n",
" 'word': 'there',\n",
" 'start': 1942,\n",
" 'end': 1947},\n",
" {'entity': 'en',\n",
" 'score': 0.9996507,\n",
" 'index': 459,\n",
" 'word': 'were',\n",
" 'start': 1948,\n",
" 'end': 1952},\n",
" {'entity': 'en',\n",
" 'score': 0.99955803,\n",
" 'index': 460,\n",
" 'word': 'any',\n",
" 'start': 1953,\n",
" 'end': 1956},\n",
" {'entity': 'en',\n",
" 'score': 0.99953544,\n",
" 'index': 461,\n",
" 'word': 'signs',\n",
" 'start': 1957,\n",
" 'end': 1962},\n",
" {'entity': 'en',\n",
" 'score': 0.99960965,\n",
" 'index': 462,\n",
" 'word': 'of',\n",
" 'start': 1963,\n",
" 'end': 1965},\n",
" {'entity': 'en',\n",
" 'score': 0.99961805,\n",
" 'index': 463,\n",
" 'word': 'life',\n",
" 'start': 1966,\n",
" 'end': 1970},\n",
" {'entity': 'other',\n",
" 'score': 0.9999225,\n",
" 'index': 464,\n",
" 'word': ',',\n",
" 'start': 1970,\n",
" 'end': 1971},\n",
" {'entity': 'en',\n",
" 'score': 0.9997342,\n",
" 'index': 465,\n",
" 'word': 'you',\n",
" 'start': 1972,\n",
" 'end': 1975},\n",
" {'entity': 'en',\n",
" 'score': 0.99973625,\n",
" 'index': 466,\n",
" 'word': 'could',\n",
" 'start': 1976,\n",
" 'end': 1981},\n",
" {'entity': 'en',\n",
" 'score': 0.9997397,\n",
" 'index': 467,\n",
" 'word': 'easily',\n",
" 'start': 1982,\n",
" 'end': 1988},\n",
" {'entity': 'en',\n",
" 'score': 0.99969923,\n",
" 'index': 468,\n",
" 'word': 'see',\n",
" 'start': 1989,\n",
" 'end': 1992},\n",
" {'entity': 'en',\n",
" 'score': 0.999673,\n",
" 'index': 469,\n",
" 'word': 'what',\n",
" 'start': 1993,\n",
" 'end': 1997},\n",
" {'entity': 'en',\n",
" 'score': 0.9996773,\n",
" 'index': 470,\n",
" 'word': 'they',\n",
" 'start': 1998,\n",
" 'end': 2002},\n",
" {'entity': 'en',\n",
" 'score': 0.9995364,\n",
" 'index': 471,\n",
" 'word': 'were',\n",
" 'start': 2003,\n",
" 'end': 2007},\n",
" {'entity': 'other',\n",
" 'score': 0.99993205,\n",
" 'index': 472,\n",
" 'word': '.',\n",
" 'start': 2007,\n",
" 'end': 2008},\n",
" {'entity': 'en',\n",
" 'score': 0.99950016,\n",
" 'index': 473,\n",
" 'word': 'What',\n",
" 'start': 2009,\n",
" 'end': 2013},\n",
" {'entity': 'en',\n",
" 'score': 0.999288,\n",
" 'index': 474,\n",
" 'word': 'the',\n",
" 'start': 2014,\n",
" 'end': 2017},\n",
" {'entity': 'en',\n",
" 'score': 0.99938726,\n",
" 'index': 475,\n",
" 'word': 'picture',\n",
" 'start': 2018,\n",
" 'end': 2025},\n",
" {'entity': 'en',\n",
" 'score': 0.9994843,\n",
" 'index': 476,\n",
" 'word': 'showed',\n",
" 'start': 2026,\n",
" 'end': 2032},\n",
" {'entity': 'en',\n",
" 'score': 0.99938786,\n",
" 'index': 477,\n",
" 'word': 'was',\n",
" 'start': 2033,\n",
" 'end': 2036},\n",
" {'entity': 'en',\n",
" 'score': 0.99811554,\n",
" 'index': 478,\n",
" 'word': 'the',\n",
" 'start': 2037,\n",
" 'end': 2040},\n",
" {'entity': 'en',\n",
" 'score': 0.9786213,\n",
" 'index': 479,\n",
" 'word': 'but',\n",
" 'start': 2041,\n",
" 'end': 2044},\n",
" {'entity': 'en',\n",
" 'score': 0.99004257,\n",
" 'index': 480,\n",
" 'word': '##te',\n",
" 'start': 2044,\n",
" 'end': 2046},\n",
" {'entity': 'en',\n",
" 'score': 0.99612516,\n",
" 'index': 481,\n",
" 'word': 'or',\n",
" 'start': 2047,\n",
" 'end': 2049},\n",
" {'entity': 'en',\n",
" 'score': 0.87278074,\n",
" 'index': 482,\n",
" 'word': 'mesa',\n",
" 'start': 2050,\n",
" 'end': 2054},\n",
" {'entity': 'other',\n",
" 'score': 0.9999268,\n",
" 'index': 483,\n",
" 'word': ',',\n",
" 'start': 2054,\n",
" 'end': 2055},\n",
" {'entity': 'en',\n",
" 'score': 0.99910945,\n",
" 'index': 484,\n",
" 'word': 'which',\n",
" 'start': 2056,\n",
" 'end': 2061},\n",
" {'entity': 'en',\n",
" 'score': 0.9985214,\n",
" 'index': 485,\n",
" 'word': 'are',\n",
" 'start': 2062,\n",
" 'end': 2065},\n",
" {'entity': 'en',\n",
" 'score': 0.9983,\n",
" 'index': 486,\n",
" 'word': 'land',\n",
" 'start': 2066,\n",
" 'end': 2070},\n",
" {'entity': 'en',\n",
" 'score': 0.9991359,\n",
" 'index': 487,\n",
" 'word': '##form',\n",
" 'start': 2070,\n",
" 'end': 2074},\n",
" {'entity': 'en',\n",
" 'score': 0.99933463,\n",
" 'index': 488,\n",
" 'word': '##s',\n",
" 'start': 2074,\n",
" 'end': 2075},\n",
" {'entity': 'en',\n",
" 'score': 0.9994253,\n",
" 'index': 489,\n",
" 'word': 'common',\n",
" 'start': 2076,\n",
" 'end': 2082},\n",
" {'entity': 'en',\n",
" 'score': 0.9994831,\n",
" 'index': 490,\n",
" 'word': 'around',\n",
" 'start': 2083,\n",
" 'end': 2089},\n",
" {'entity': 'en',\n",
" 'score': 0.99935335,\n",
" 'index': 491,\n",
" 'word': 'the',\n",
" 'start': 2090,\n",
" 'end': 2093},\n",
" {'entity': 'en',\n",
" 'score': 0.9982742,\n",
" 'index': 492,\n",
" 'word': 'American',\n",
" 'start': 2094,\n",
" 'end': 2102},\n",
" {'entity': 'en',\n",
" 'score': 0.99908066,\n",
" 'index': 493,\n",
" 'word': 'West',\n",
" 'start': 2103,\n",
" 'end': 2107},\n",
" {'entity': 'other',\n",
" 'score': 0.9999201,\n",
" 'index': 494,\n",
" 'word': '.',\n",
" 'start': 2107,\n",
" 'end': 2108}]"
]
},
"execution_count": 127,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from transformers import AutoTokenizer, AutoModelForTokenClassification, pipeline\n",
"\n",
"tokenizer = AutoTokenizer.from_pretrained(\"sagorsarker/codeswitch-spaeng-lid-lince\")\n",
"\n",
"model = AutoModelForTokenClassification.from_pretrained(\"sagorsarker/codeswitch-spaeng-lid-lince\")\n",
"lid_model = pipeline('ner', model=model, tokenizer=tokenizer)\n",
"\n",
"lid_model(text)"
]
},
{
"cell_type": "code",
"execution_count": 130,
"id": "867c77f9-b9d8-4f29-b07f-cab4ee55a8f3",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"entity\n",
"en 411\n",
"ne 22\n",
"other 61\n",
"dtype: int64\n"
]
},
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" entity | \n",
" score | \n",
"
\n",
" \n",
" | entity | \n",
" | \n",
" | \n",
"
\n",
" \n",
" \n",
" \n",
" | en | \n",
" 411 | \n",
" 0.996325 | \n",
"
\n",
" \n",
" | ne | \n",
" 22 | \n",
" 0.925739 | \n",
"
\n",
" \n",
" | other | \n",
" 61 | \n",
" 0.990110 | \n",
"
\n",
" \n",
"
\n",
"
\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" entity | \n",
" score | \n",
"
\n",
" \n",
" | entity | \n",
" | \n",
" | \n",
"
\n",
" \n",
" \n",
" \n",
" | I-EN | \n",
" 493 | \n",
" 0.999995 | \n",
"
\n",
" \n",
"
\n",
"
\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" entity | \n",
" score | \n",
"
\n",
" \n",
" | entity | \n",
" | \n",
" | \n",
"
\n",
" \n",
" \n",
" \n",
" | ar | \n",
" 8 | \n",
" 0.084216 | \n",
"
\n",
" \n",
" | el | \n",
" 305 | \n",
" 0.120293 | \n",
"
\n",
" \n",
" | en | \n",
" 178 | \n",
" 0.143087 | \n",
"
\n",
" \n",
" | ja | \n",
" 1 | \n",
" 0.077624 | \n",
"
\n",
" \n",
" | pl | \n",
" 1 | \n",
" 0.075424 | \n",
"
\n",
" \n",
"
\n",
"