End of training
Browse files- README.md +116 -0
- all_results.json +6 -0
- config.json +236 -0
- model.safetensors +3 -0
- preprocessor_config.json +27 -0
- test_results.json +6 -0
- trainer_state.json +1177 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,116 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: mit
|
| 4 |
+
base_model: google/vivit-b-16x2-kinetics400
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
metrics:
|
| 8 |
+
- accuracy
|
| 9 |
+
- precision
|
| 10 |
+
- recall
|
| 11 |
+
- f1
|
| 12 |
+
model-index:
|
| 13 |
+
- name: ViViT_WLASL_100_SR_4_ep200_p20
|
| 14 |
+
results: []
|
| 15 |
+
---
|
| 16 |
+
|
| 17 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 18 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 19 |
+
|
| 20 |
+
# ViViT_WLASL_100_SR_4_ep200_p20
|
| 21 |
+
|
| 22 |
+
This model is a fine-tuned version of [google/vivit-b-16x2-kinetics400](https://huggingface.co/google/vivit-b-16x2-kinetics400) on an unknown dataset.
|
| 23 |
+
It achieves the following results on the evaluation set:
|
| 24 |
+
- Loss: 1.7950
|
| 25 |
+
- Accuracy: 0.6272
|
| 26 |
+
- Precision: 0.6755
|
| 27 |
+
- Recall: 0.6272
|
| 28 |
+
- F1: 0.6047
|
| 29 |
+
|
| 30 |
+
## Model description
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Intended uses & limitations
|
| 35 |
+
|
| 36 |
+
More information needed
|
| 37 |
+
|
| 38 |
+
## Training and evaluation data
|
| 39 |
+
|
| 40 |
+
More information needed
|
| 41 |
+
|
| 42 |
+
## Training procedure
|
| 43 |
+
|
| 44 |
+
### Training hyperparameters
|
| 45 |
+
|
| 46 |
+
The following hyperparameters were used during training:
|
| 47 |
+
- learning_rate: 5e-05
|
| 48 |
+
- train_batch_size: 2
|
| 49 |
+
- eval_batch_size: 2
|
| 50 |
+
- seed: 42
|
| 51 |
+
- gradient_accumulation_steps: 4
|
| 52 |
+
- total_train_batch_size: 8
|
| 53 |
+
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 54 |
+
- lr_scheduler_type: linear
|
| 55 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 56 |
+
- training_steps: 36000
|
| 57 |
+
- mixed_precision_training: Native AMP
|
| 58 |
+
|
| 59 |
+
### Training results
|
| 60 |
+
|
| 61 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|
| 62 |
+
|:-------------:|:-------:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
|
| 63 |
+
| 18.9788 | 0.005 | 180 | 4.6909 | 0.0266 | 0.0055 | 0.0266 | 0.0082 |
|
| 64 |
+
| 18.6803 | 1.0050 | 360 | 4.6180 | 0.0266 | 0.0034 | 0.0266 | 0.0058 |
|
| 65 |
+
| 18.0341 | 2.0050 | 540 | 4.5181 | 0.0473 | 0.0205 | 0.0473 | 0.0238 |
|
| 66 |
+
| 17.0617 | 3.0050 | 721 | 4.3376 | 0.0740 | 0.0541 | 0.0740 | 0.0478 |
|
| 67 |
+
| 15.6302 | 4.005 | 901 | 4.0709 | 0.1183 | 0.0772 | 0.1183 | 0.0815 |
|
| 68 |
+
| 14.0851 | 5.0050 | 1081 | 3.7552 | 0.1864 | 0.1536 | 0.1864 | 0.1449 |
|
| 69 |
+
| 11.953 | 6.0050 | 1261 | 3.4065 | 0.2663 | 0.2643 | 0.2663 | 0.2221 |
|
| 70 |
+
| 9.8223 | 7.0050 | 1442 | 3.1038 | 0.3373 | 0.3284 | 0.3373 | 0.2944 |
|
| 71 |
+
| 7.8126 | 8.005 | 1622 | 2.7834 | 0.4142 | 0.4201 | 0.4142 | 0.3676 |
|
| 72 |
+
| 6.0952 | 9.0050 | 1802 | 2.5129 | 0.4763 | 0.4950 | 0.4763 | 0.4368 |
|
| 73 |
+
| 4.3155 | 10.0050 | 1982 | 2.2757 | 0.5059 | 0.5594 | 0.5059 | 0.4894 |
|
| 74 |
+
| 3.0214 | 11.0050 | 2163 | 2.0460 | 0.5473 | 0.5640 | 0.5473 | 0.5178 |
|
| 75 |
+
| 2.0687 | 12.005 | 2343 | 1.8803 | 0.5917 | 0.6184 | 0.5917 | 0.5731 |
|
| 76 |
+
| 1.3523 | 13.0050 | 2523 | 1.7261 | 0.5917 | 0.6065 | 0.5917 | 0.5601 |
|
| 77 |
+
| 0.7828 | 14.0050 | 2703 | 1.6275 | 0.6036 | 0.6644 | 0.6036 | 0.5924 |
|
| 78 |
+
| 0.4222 | 15.0050 | 2884 | 1.5284 | 0.6420 | 0.6756 | 0.6420 | 0.6245 |
|
| 79 |
+
| 0.3113 | 16.005 | 3064 | 1.5459 | 0.6272 | 0.6664 | 0.6272 | 0.6092 |
|
| 80 |
+
| 0.2021 | 17.0050 | 3244 | 1.4441 | 0.6657 | 0.6991 | 0.6657 | 0.6484 |
|
| 81 |
+
| 0.1698 | 18.0050 | 3424 | 1.5340 | 0.6124 | 0.6511 | 0.6124 | 0.5942 |
|
| 82 |
+
| 0.1199 | 19.0050 | 3605 | 1.3935 | 0.6509 | 0.6746 | 0.6509 | 0.6288 |
|
| 83 |
+
| 0.0244 | 20.005 | 3785 | 1.4782 | 0.6686 | 0.7130 | 0.6686 | 0.6574 |
|
| 84 |
+
| 0.0407 | 21.0050 | 3965 | 1.3890 | 0.6686 | 0.7149 | 0.6686 | 0.6557 |
|
| 85 |
+
| 0.0719 | 22.0050 | 4145 | 1.4897 | 0.6598 | 0.7189 | 0.6598 | 0.6477 |
|
| 86 |
+
| 0.1163 | 23.0050 | 4326 | 1.3919 | 0.6716 | 0.7218 | 0.6716 | 0.6639 |
|
| 87 |
+
| 0.1167 | 24.005 | 4506 | 1.5690 | 0.6538 | 0.7189 | 0.6538 | 0.6380 |
|
| 88 |
+
| 0.0366 | 25.0050 | 4686 | 1.5032 | 0.6746 | 0.6979 | 0.6746 | 0.6541 |
|
| 89 |
+
| 0.1065 | 26.0050 | 4866 | 1.4893 | 0.6391 | 0.6475 | 0.6391 | 0.6135 |
|
| 90 |
+
| 0.0454 | 27.0050 | 5047 | 1.5013 | 0.6243 | 0.6601 | 0.6243 | 0.6022 |
|
| 91 |
+
| 0.0844 | 28.005 | 5227 | 1.5609 | 0.6598 | 0.6974 | 0.6598 | 0.6388 |
|
| 92 |
+
| 0.109 | 29.0050 | 5407 | 1.4840 | 0.6657 | 0.7151 | 0.6657 | 0.6507 |
|
| 93 |
+
| 0.1508 | 30.0050 | 5587 | 1.8017 | 0.6036 | 0.6784 | 0.6036 | 0.5903 |
|
| 94 |
+
| 0.1114 | 31.0050 | 5768 | 1.6676 | 0.6391 | 0.6721 | 0.6391 | 0.6134 |
|
| 95 |
+
| 0.0931 | 32.005 | 5948 | 1.5345 | 0.6746 | 0.7082 | 0.6746 | 0.6520 |
|
| 96 |
+
| 0.0619 | 33.0050 | 6128 | 1.7462 | 0.6302 | 0.6424 | 0.6302 | 0.6008 |
|
| 97 |
+
| 0.2698 | 34.0050 | 6308 | 1.7032 | 0.6331 | 0.6711 | 0.6331 | 0.6126 |
|
| 98 |
+
| 0.1108 | 35.0050 | 6489 | 1.7695 | 0.6538 | 0.6784 | 0.6538 | 0.6265 |
|
| 99 |
+
| 0.1006 | 36.005 | 6669 | 2.0188 | 0.5828 | 0.6289 | 0.5828 | 0.5661 |
|
| 100 |
+
| 0.0823 | 37.0050 | 6849 | 1.6487 | 0.6568 | 0.6874 | 0.6568 | 0.6425 |
|
| 101 |
+
| 0.0632 | 38.0050 | 7029 | 1.8014 | 0.6361 | 0.6917 | 0.6361 | 0.6253 |
|
| 102 |
+
| 0.1162 | 39.0050 | 7210 | 1.6741 | 0.6450 | 0.6672 | 0.6450 | 0.6196 |
|
| 103 |
+
| 0.0846 | 40.005 | 7390 | 1.8032 | 0.6361 | 0.6948 | 0.6361 | 0.6205 |
|
| 104 |
+
| 0.1528 | 41.0050 | 7570 | 1.8375 | 0.6331 | 0.6732 | 0.6331 | 0.6102 |
|
| 105 |
+
| 0.0695 | 42.0050 | 7750 | 1.6722 | 0.6568 | 0.7030 | 0.6568 | 0.6417 |
|
| 106 |
+
| 0.1516 | 43.0050 | 7931 | 1.7811 | 0.6716 | 0.7009 | 0.6716 | 0.6495 |
|
| 107 |
+
| 0.1565 | 44.005 | 8111 | 1.8077 | 0.6538 | 0.6884 | 0.6538 | 0.6347 |
|
| 108 |
+
| 0.0728 | 45.0050 | 8291 | 1.7950 | 0.6272 | 0.6755 | 0.6272 | 0.6047 |
|
| 109 |
+
|
| 110 |
+
|
| 111 |
+
### Framework versions
|
| 112 |
+
|
| 113 |
+
- Transformers 4.46.1
|
| 114 |
+
- Pytorch 2.5.1+cu124
|
| 115 |
+
- Datasets 3.1.0
|
| 116 |
+
- Tokenizers 0.20.1
|
all_results.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"accuracy": 0.6046511627906976,
|
| 3 |
+
"f1": 0.5856390379646192,
|
| 4 |
+
"precision": 0.6404392764857881,
|
| 5 |
+
"recall": 0.6046511627906976
|
| 6 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,236 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google/vivit-b-16x2-kinetics400",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"VivitForVideoClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"hidden_act": "gelu_fast",
|
| 8 |
+
"hidden_dropout_prob": 0.0,
|
| 9 |
+
"hidden_size": 768,
|
| 10 |
+
"id2label": {
|
| 11 |
+
"0": "accident",
|
| 12 |
+
"1": "africa",
|
| 13 |
+
"2": "all",
|
| 14 |
+
"3": "apple",
|
| 15 |
+
"4": "basketball",
|
| 16 |
+
"5": "bed",
|
| 17 |
+
"6": "before",
|
| 18 |
+
"7": "bird",
|
| 19 |
+
"8": "birthday",
|
| 20 |
+
"9": "black",
|
| 21 |
+
"10": "blue",
|
| 22 |
+
"11": "book",
|
| 23 |
+
"12": "bowling",
|
| 24 |
+
"13": "brown",
|
| 25 |
+
"14": "but",
|
| 26 |
+
"15": "can",
|
| 27 |
+
"16": "candy",
|
| 28 |
+
"17": "chair",
|
| 29 |
+
"18": "change",
|
| 30 |
+
"19": "cheat",
|
| 31 |
+
"20": "city",
|
| 32 |
+
"21": "clothes",
|
| 33 |
+
"22": "color",
|
| 34 |
+
"23": "computer",
|
| 35 |
+
"24": "cook",
|
| 36 |
+
"25": "cool",
|
| 37 |
+
"26": "corn",
|
| 38 |
+
"27": "cousin",
|
| 39 |
+
"28": "cow",
|
| 40 |
+
"29": "dance",
|
| 41 |
+
"30": "dark",
|
| 42 |
+
"31": "deaf",
|
| 43 |
+
"32": "decide",
|
| 44 |
+
"33": "doctor",
|
| 45 |
+
"34": "dog",
|
| 46 |
+
"35": "drink",
|
| 47 |
+
"36": "eat",
|
| 48 |
+
"37": "enjoy",
|
| 49 |
+
"38": "family",
|
| 50 |
+
"39": "fine",
|
| 51 |
+
"40": "finish",
|
| 52 |
+
"41": "fish",
|
| 53 |
+
"42": "forget",
|
| 54 |
+
"43": "full",
|
| 55 |
+
"44": "give",
|
| 56 |
+
"45": "go",
|
| 57 |
+
"46": "graduate",
|
| 58 |
+
"47": "hat",
|
| 59 |
+
"48": "hearing",
|
| 60 |
+
"49": "help",
|
| 61 |
+
"50": "hot",
|
| 62 |
+
"51": "how",
|
| 63 |
+
"52": "jacket",
|
| 64 |
+
"53": "kiss",
|
| 65 |
+
"54": "language",
|
| 66 |
+
"55": "last",
|
| 67 |
+
"56": "later",
|
| 68 |
+
"57": "letter",
|
| 69 |
+
"58": "like",
|
| 70 |
+
"59": "man",
|
| 71 |
+
"60": "many",
|
| 72 |
+
"61": "medicine",
|
| 73 |
+
"62": "meet",
|
| 74 |
+
"63": "mother",
|
| 75 |
+
"64": "need",
|
| 76 |
+
"65": "no",
|
| 77 |
+
"66": "now",
|
| 78 |
+
"67": "orange",
|
| 79 |
+
"68": "paint",
|
| 80 |
+
"69": "paper",
|
| 81 |
+
"70": "pink",
|
| 82 |
+
"71": "pizza",
|
| 83 |
+
"72": "play",
|
| 84 |
+
"73": "pull",
|
| 85 |
+
"74": "purple",
|
| 86 |
+
"75": "right",
|
| 87 |
+
"76": "same",
|
| 88 |
+
"77": "school",
|
| 89 |
+
"78": "secretary",
|
| 90 |
+
"79": "shirt",
|
| 91 |
+
"80": "short",
|
| 92 |
+
"81": "son",
|
| 93 |
+
"82": "study",
|
| 94 |
+
"83": "table",
|
| 95 |
+
"84": "tall",
|
| 96 |
+
"85": "tell",
|
| 97 |
+
"86": "thanksgiving",
|
| 98 |
+
"87": "thin",
|
| 99 |
+
"88": "thursday",
|
| 100 |
+
"89": "time",
|
| 101 |
+
"90": "walk",
|
| 102 |
+
"91": "want",
|
| 103 |
+
"92": "what",
|
| 104 |
+
"93": "white",
|
| 105 |
+
"94": "who",
|
| 106 |
+
"95": "woman",
|
| 107 |
+
"96": "work",
|
| 108 |
+
"97": "wrong",
|
| 109 |
+
"98": "year",
|
| 110 |
+
"99": "yes"
|
| 111 |
+
},
|
| 112 |
+
"image_size": 224,
|
| 113 |
+
"initializer_range": 0.02,
|
| 114 |
+
"intermediate_size": 3072,
|
| 115 |
+
"label2id": {
|
| 116 |
+
"accident": 0,
|
| 117 |
+
"africa": 1,
|
| 118 |
+
"all": 2,
|
| 119 |
+
"apple": 3,
|
| 120 |
+
"basketball": 4,
|
| 121 |
+
"bed": 5,
|
| 122 |
+
"before": 6,
|
| 123 |
+
"bird": 7,
|
| 124 |
+
"birthday": 8,
|
| 125 |
+
"black": 9,
|
| 126 |
+
"blue": 10,
|
| 127 |
+
"book": 11,
|
| 128 |
+
"bowling": 12,
|
| 129 |
+
"brown": 13,
|
| 130 |
+
"but": 14,
|
| 131 |
+
"can": 15,
|
| 132 |
+
"candy": 16,
|
| 133 |
+
"chair": 17,
|
| 134 |
+
"change": 18,
|
| 135 |
+
"cheat": 19,
|
| 136 |
+
"city": 20,
|
| 137 |
+
"clothes": 21,
|
| 138 |
+
"color": 22,
|
| 139 |
+
"computer": 23,
|
| 140 |
+
"cook": 24,
|
| 141 |
+
"cool": 25,
|
| 142 |
+
"corn": 26,
|
| 143 |
+
"cousin": 27,
|
| 144 |
+
"cow": 28,
|
| 145 |
+
"dance": 29,
|
| 146 |
+
"dark": 30,
|
| 147 |
+
"deaf": 31,
|
| 148 |
+
"decide": 32,
|
| 149 |
+
"doctor": 33,
|
| 150 |
+
"dog": 34,
|
| 151 |
+
"drink": 35,
|
| 152 |
+
"eat": 36,
|
| 153 |
+
"enjoy": 37,
|
| 154 |
+
"family": 38,
|
| 155 |
+
"fine": 39,
|
| 156 |
+
"finish": 40,
|
| 157 |
+
"fish": 41,
|
| 158 |
+
"forget": 42,
|
| 159 |
+
"full": 43,
|
| 160 |
+
"give": 44,
|
| 161 |
+
"go": 45,
|
| 162 |
+
"graduate": 46,
|
| 163 |
+
"hat": 47,
|
| 164 |
+
"hearing": 48,
|
| 165 |
+
"help": 49,
|
| 166 |
+
"hot": 50,
|
| 167 |
+
"how": 51,
|
| 168 |
+
"jacket": 52,
|
| 169 |
+
"kiss": 53,
|
| 170 |
+
"language": 54,
|
| 171 |
+
"last": 55,
|
| 172 |
+
"later": 56,
|
| 173 |
+
"letter": 57,
|
| 174 |
+
"like": 58,
|
| 175 |
+
"man": 59,
|
| 176 |
+
"many": 60,
|
| 177 |
+
"medicine": 61,
|
| 178 |
+
"meet": 62,
|
| 179 |
+
"mother": 63,
|
| 180 |
+
"need": 64,
|
| 181 |
+
"no": 65,
|
| 182 |
+
"now": 66,
|
| 183 |
+
"orange": 67,
|
| 184 |
+
"paint": 68,
|
| 185 |
+
"paper": 69,
|
| 186 |
+
"pink": 70,
|
| 187 |
+
"pizza": 71,
|
| 188 |
+
"play": 72,
|
| 189 |
+
"pull": 73,
|
| 190 |
+
"purple": 74,
|
| 191 |
+
"right": 75,
|
| 192 |
+
"same": 76,
|
| 193 |
+
"school": 77,
|
| 194 |
+
"secretary": 78,
|
| 195 |
+
"shirt": 79,
|
| 196 |
+
"short": 80,
|
| 197 |
+
"son": 81,
|
| 198 |
+
"study": 82,
|
| 199 |
+
"table": 83,
|
| 200 |
+
"tall": 84,
|
| 201 |
+
"tell": 85,
|
| 202 |
+
"thanksgiving": 86,
|
| 203 |
+
"thin": 87,
|
| 204 |
+
"thursday": 88,
|
| 205 |
+
"time": 89,
|
| 206 |
+
"walk": 90,
|
| 207 |
+
"want": 91,
|
| 208 |
+
"what": 92,
|
| 209 |
+
"white": 93,
|
| 210 |
+
"who": 94,
|
| 211 |
+
"woman": 95,
|
| 212 |
+
"work": 96,
|
| 213 |
+
"wrong": 97,
|
| 214 |
+
"year": 98,
|
| 215 |
+
"yes": 99
|
| 216 |
+
},
|
| 217 |
+
"layer_norm_eps": 1e-06,
|
| 218 |
+
"model_type": "vivit",
|
| 219 |
+
"num_attention_heads": 12,
|
| 220 |
+
"num_channels": 3,
|
| 221 |
+
"num_frames": 32,
|
| 222 |
+
"num_hidden_layers": 12,
|
| 223 |
+
"qkv_bias": true,
|
| 224 |
+
"torch_dtype": "float32",
|
| 225 |
+
"transformers_version": "4.46.1",
|
| 226 |
+
"tubelet_size": [
|
| 227 |
+
2,
|
| 228 |
+
16,
|
| 229 |
+
16
|
| 230 |
+
],
|
| 231 |
+
"video_size": [
|
| 232 |
+
32,
|
| 233 |
+
224,
|
| 234 |
+
224
|
| 235 |
+
]
|
| 236 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:546e49a0b9b98d955604a5971580c61b675616c8619f227c72a072325466cc71
|
| 3 |
+
size 354916840
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"crop_size": {
|
| 3 |
+
"height": 224,
|
| 4 |
+
"width": 224
|
| 5 |
+
},
|
| 6 |
+
"do_center_crop": true,
|
| 7 |
+
"do_normalize": true,
|
| 8 |
+
"do_rescale": true,
|
| 9 |
+
"do_resize": true,
|
| 10 |
+
"image_mean": [
|
| 11 |
+
0.5,
|
| 12 |
+
0.5,
|
| 13 |
+
0.5
|
| 14 |
+
],
|
| 15 |
+
"image_processor_type": "VivitImageProcessor",
|
| 16 |
+
"image_std": [
|
| 17 |
+
0.5,
|
| 18 |
+
0.5,
|
| 19 |
+
0.5
|
| 20 |
+
],
|
| 21 |
+
"offset": true,
|
| 22 |
+
"resample": 2,
|
| 23 |
+
"rescale_factor": 0.00784313725490196,
|
| 24 |
+
"size": {
|
| 25 |
+
"shortest_edge": 224
|
| 26 |
+
}
|
| 27 |
+
}
|
test_results.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"accuracy": 0.6046511627906976,
|
| 3 |
+
"f1": 0.5856390379646192,
|
| 4 |
+
"precision": 0.6404392764857881,
|
| 5 |
+
"recall": 0.6046511627906976
|
| 6 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1177 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.6745562130177515,
|
| 3 |
+
"best_model_checkpoint": "/media/cse/HDD/Shawon/shawon/MY DATA/ViViT_WLASL_100_SR_4_ep200_p20/checkpoint-4686",
|
| 4 |
+
"epoch": 45.00499305555556,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 8291,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.002777777777777778,
|
| 13 |
+
"grad_norm": 46.614158630371094,
|
| 14 |
+
"learning_rate": 1.3472222222222222e-06,
|
| 15 |
+
"loss": 18.9788,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.005,
|
| 20 |
+
"eval_accuracy": 0.026627218934911243,
|
| 21 |
+
"eval_f1": 0.008186647196111367,
|
| 22 |
+
"eval_loss": 4.690920829772949,
|
| 23 |
+
"eval_precision": 0.005517409748672709,
|
| 24 |
+
"eval_recall": 0.026627218934911243,
|
| 25 |
+
"eval_runtime": 19.4352,
|
| 26 |
+
"eval_samples_per_second": 17.391,
|
| 27 |
+
"eval_steps_per_second": 8.696,
|
| 28 |
+
"step": 180
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 1.000548611111111,
|
| 32 |
+
"grad_norm": 44.873966217041016,
|
| 33 |
+
"learning_rate": 2.7361111111111114e-06,
|
| 34 |
+
"loss": 18.8244,
|
| 35 |
+
"step": 200
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 1.0033263888888888,
|
| 39 |
+
"grad_norm": 43.23536682128906,
|
| 40 |
+
"learning_rate": 4.125e-06,
|
| 41 |
+
"loss": 18.6803,
|
| 42 |
+
"step": 300
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 1.0049930555555555,
|
| 46 |
+
"eval_accuracy": 0.026627218934911243,
|
| 47 |
+
"eval_f1": 0.0057761083277597,
|
| 48 |
+
"eval_loss": 4.617973327636719,
|
| 49 |
+
"eval_precision": 0.003442859602301134,
|
| 50 |
+
"eval_recall": 0.026627218934911243,
|
| 51 |
+
"eval_runtime": 18.7273,
|
| 52 |
+
"eval_samples_per_second": 18.049,
|
| 53 |
+
"eval_steps_per_second": 9.024,
|
| 54 |
+
"step": 360
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 2.001097222222222,
|
| 58 |
+
"grad_norm": 43.87873458862305,
|
| 59 |
+
"learning_rate": 5.513888888888889e-06,
|
| 60 |
+
"loss": 18.3151,
|
| 61 |
+
"step": 400
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 2.003875,
|
| 65 |
+
"grad_norm": 45.60523986816406,
|
| 66 |
+
"learning_rate": 6.902777777777777e-06,
|
| 67 |
+
"loss": 18.0341,
|
| 68 |
+
"step": 500
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 2.004986111111111,
|
| 72 |
+
"eval_accuracy": 0.047337278106508875,
|
| 73 |
+
"eval_f1": 0.023768276430998322,
|
| 74 |
+
"eval_loss": 4.518121242523193,
|
| 75 |
+
"eval_precision": 0.020485887948289476,
|
| 76 |
+
"eval_recall": 0.047337278106508875,
|
| 77 |
+
"eval_runtime": 19.5951,
|
| 78 |
+
"eval_samples_per_second": 17.249,
|
| 79 |
+
"eval_steps_per_second": 8.625,
|
| 80 |
+
"step": 540
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 3.0016458333333333,
|
| 84 |
+
"grad_norm": 49.1752815246582,
|
| 85 |
+
"learning_rate": 8.291666666666667e-06,
|
| 86 |
+
"loss": 17.4359,
|
| 87 |
+
"step": 600
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 3.004423611111111,
|
| 91 |
+
"grad_norm": 51.076786041259766,
|
| 92 |
+
"learning_rate": 9.680555555555556e-06,
|
| 93 |
+
"loss": 17.0617,
|
| 94 |
+
"step": 700
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 3.0050069444444443,
|
| 98 |
+
"eval_accuracy": 0.07396449704142012,
|
| 99 |
+
"eval_f1": 0.047821065111354005,
|
| 100 |
+
"eval_loss": 4.337636470794678,
|
| 101 |
+
"eval_precision": 0.054133340554329064,
|
| 102 |
+
"eval_recall": 0.07396449704142012,
|
| 103 |
+
"eval_runtime": 18.9343,
|
| 104 |
+
"eval_samples_per_second": 17.851,
|
| 105 |
+
"eval_steps_per_second": 8.926,
|
| 106 |
+
"step": 721
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 4.002194444444444,
|
| 110 |
+
"grad_norm": 46.15924835205078,
|
| 111 |
+
"learning_rate": 1.1069444444444445e-05,
|
| 112 |
+
"loss": 16.0988,
|
| 113 |
+
"step": 800
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 4.004972222222222,
|
| 117 |
+
"grad_norm": 51.79167938232422,
|
| 118 |
+
"learning_rate": 1.2458333333333334e-05,
|
| 119 |
+
"loss": 15.6302,
|
| 120 |
+
"step": 900
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 4.005,
|
| 124 |
+
"eval_accuracy": 0.11834319526627218,
|
| 125 |
+
"eval_f1": 0.081513142682628,
|
| 126 |
+
"eval_loss": 4.070884704589844,
|
| 127 |
+
"eval_precision": 0.07720141700404858,
|
| 128 |
+
"eval_recall": 0.11834319526627218,
|
| 129 |
+
"eval_runtime": 20.1431,
|
| 130 |
+
"eval_samples_per_second": 16.78,
|
| 131 |
+
"eval_steps_per_second": 8.39,
|
| 132 |
+
"step": 901
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 5.002743055555555,
|
| 136 |
+
"grad_norm": 55.07365798950195,
|
| 137 |
+
"learning_rate": 1.3847222222222223e-05,
|
| 138 |
+
"loss": 14.0851,
|
| 139 |
+
"step": 1000
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 5.0049930555555555,
|
| 143 |
+
"eval_accuracy": 0.1863905325443787,
|
| 144 |
+
"eval_f1": 0.14494279207764105,
|
| 145 |
+
"eval_loss": 3.755218029022217,
|
| 146 |
+
"eval_precision": 0.1535636372174834,
|
| 147 |
+
"eval_recall": 0.1863905325443787,
|
| 148 |
+
"eval_runtime": 19.7392,
|
| 149 |
+
"eval_samples_per_second": 17.123,
|
| 150 |
+
"eval_steps_per_second": 8.562,
|
| 151 |
+
"step": 1081
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 6.0005138888888885,
|
| 155 |
+
"grad_norm": 49.54998779296875,
|
| 156 |
+
"learning_rate": 1.5236111111111112e-05,
|
| 157 |
+
"loss": 13.3452,
|
| 158 |
+
"step": 1100
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 6.003291666666667,
|
| 162 |
+
"grad_norm": 54.62944793701172,
|
| 163 |
+
"learning_rate": 1.6625e-05,
|
| 164 |
+
"loss": 11.953,
|
| 165 |
+
"step": 1200
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 6.004986111111111,
|
| 169 |
+
"eval_accuracy": 0.26627218934911245,
|
| 170 |
+
"eval_f1": 0.2220989336141275,
|
| 171 |
+
"eval_loss": 3.406508684158325,
|
| 172 |
+
"eval_precision": 0.26431351743742965,
|
| 173 |
+
"eval_recall": 0.26627218934911245,
|
| 174 |
+
"eval_runtime": 19.5936,
|
| 175 |
+
"eval_samples_per_second": 17.251,
|
| 176 |
+
"eval_steps_per_second": 8.625,
|
| 177 |
+
"step": 1261
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 7.0010625,
|
| 181 |
+
"grad_norm": 56.319576263427734,
|
| 182 |
+
"learning_rate": 1.8013888888888888e-05,
|
| 183 |
+
"loss": 11.0917,
|
| 184 |
+
"step": 1300
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 7.003840277777778,
|
| 188 |
+
"grad_norm": 54.84210205078125,
|
| 189 |
+
"learning_rate": 1.940277777777778e-05,
|
| 190 |
+
"loss": 9.8223,
|
| 191 |
+
"step": 1400
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 7.005006944444444,
|
| 195 |
+
"eval_accuracy": 0.33727810650887574,
|
| 196 |
+
"eval_f1": 0.29438573252416295,
|
| 197 |
+
"eval_loss": 3.103813886642456,
|
| 198 |
+
"eval_precision": 0.3283793047846302,
|
| 199 |
+
"eval_recall": 0.33727810650887574,
|
| 200 |
+
"eval_runtime": 20.1573,
|
| 201 |
+
"eval_samples_per_second": 16.768,
|
| 202 |
+
"eval_steps_per_second": 8.384,
|
| 203 |
+
"step": 1442
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 8.001611111111112,
|
| 207 |
+
"grad_norm": 51.06685256958008,
|
| 208 |
+
"learning_rate": 2.0791666666666666e-05,
|
| 209 |
+
"loss": 8.6714,
|
| 210 |
+
"step": 1500
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 8.004388888888888,
|
| 214 |
+
"grad_norm": 50.67875289916992,
|
| 215 |
+
"learning_rate": 2.2180555555555557e-05,
|
| 216 |
+
"loss": 7.8126,
|
| 217 |
+
"step": 1600
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 8.005,
|
| 221 |
+
"eval_accuracy": 0.41420118343195267,
|
| 222 |
+
"eval_f1": 0.3676157668761219,
|
| 223 |
+
"eval_loss": 2.7834277153015137,
|
| 224 |
+
"eval_precision": 0.42014839107599283,
|
| 225 |
+
"eval_recall": 0.41420118343195267,
|
| 226 |
+
"eval_runtime": 19.7317,
|
| 227 |
+
"eval_samples_per_second": 17.13,
|
| 228 |
+
"eval_steps_per_second": 8.565,
|
| 229 |
+
"step": 1622
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 9.002159722222222,
|
| 233 |
+
"grad_norm": 49.01296615600586,
|
| 234 |
+
"learning_rate": 2.3569444444444447e-05,
|
| 235 |
+
"loss": 6.4863,
|
| 236 |
+
"step": 1700
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 9.0049375,
|
| 240 |
+
"grad_norm": 47.182640075683594,
|
| 241 |
+
"learning_rate": 2.4958333333333335e-05,
|
| 242 |
+
"loss": 6.0952,
|
| 243 |
+
"step": 1800
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 9.004993055555556,
|
| 247 |
+
"eval_accuracy": 0.47633136094674555,
|
| 248 |
+
"eval_f1": 0.4368087369566659,
|
| 249 |
+
"eval_loss": 2.512887477874756,
|
| 250 |
+
"eval_precision": 0.4949564324564325,
|
| 251 |
+
"eval_recall": 0.47633136094674555,
|
| 252 |
+
"eval_runtime": 20.1068,
|
| 253 |
+
"eval_samples_per_second": 16.81,
|
| 254 |
+
"eval_steps_per_second": 8.405,
|
| 255 |
+
"step": 1802
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 10.002708333333333,
|
| 259 |
+
"grad_norm": 49.08199691772461,
|
| 260 |
+
"learning_rate": 2.6347222222222222e-05,
|
| 261 |
+
"loss": 4.3155,
|
| 262 |
+
"step": 1900
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 10.00498611111111,
|
| 266 |
+
"eval_accuracy": 0.5059171597633136,
|
| 267 |
+
"eval_f1": 0.48941789637055916,
|
| 268 |
+
"eval_loss": 2.2756905555725098,
|
| 269 |
+
"eval_precision": 0.5593944517021441,
|
| 270 |
+
"eval_recall": 0.5059171597633136,
|
| 271 |
+
"eval_runtime": 18.9167,
|
| 272 |
+
"eval_samples_per_second": 17.868,
|
| 273 |
+
"eval_steps_per_second": 8.934,
|
| 274 |
+
"step": 1982
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 11.000479166666667,
|
| 278 |
+
"grad_norm": 37.75128936767578,
|
| 279 |
+
"learning_rate": 2.7736111111111113e-05,
|
| 280 |
+
"loss": 4.3448,
|
| 281 |
+
"step": 2000
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 11.003256944444445,
|
| 285 |
+
"grad_norm": 36.987205505371094,
|
| 286 |
+
"learning_rate": 2.9111111111111112e-05,
|
| 287 |
+
"loss": 3.0214,
|
| 288 |
+
"step": 2100
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 11.005006944444444,
|
| 292 |
+
"eval_accuracy": 0.5473372781065089,
|
| 293 |
+
"eval_f1": 0.5177564874310436,
|
| 294 |
+
"eval_loss": 2.0459656715393066,
|
| 295 |
+
"eval_precision": 0.563966482235713,
|
| 296 |
+
"eval_recall": 0.5473372781065089,
|
| 297 |
+
"eval_runtime": 19.7866,
|
| 298 |
+
"eval_samples_per_second": 17.082,
|
| 299 |
+
"eval_steps_per_second": 8.541,
|
| 300 |
+
"step": 2163
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 12.001027777777777,
|
| 304 |
+
"grad_norm": 30.315874099731445,
|
| 305 |
+
"learning_rate": 3.05e-05,
|
| 306 |
+
"loss": 2.7849,
|
| 307 |
+
"step": 2200
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 12.003805555555555,
|
| 311 |
+
"grad_norm": 27.424291610717773,
|
| 312 |
+
"learning_rate": 3.188888888888889e-05,
|
| 313 |
+
"loss": 2.0687,
|
| 314 |
+
"step": 2300
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 12.005,
|
| 318 |
+
"eval_accuracy": 0.591715976331361,
|
| 319 |
+
"eval_f1": 0.5730650184792196,
|
| 320 |
+
"eval_loss": 1.8803057670593262,
|
| 321 |
+
"eval_precision": 0.6183537616229923,
|
| 322 |
+
"eval_recall": 0.591715976331361,
|
| 323 |
+
"eval_runtime": 19.466,
|
| 324 |
+
"eval_samples_per_second": 17.364,
|
| 325 |
+
"eval_steps_per_second": 8.682,
|
| 326 |
+
"step": 2343
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 13.00157638888889,
|
| 330 |
+
"grad_norm": 19.18665313720703,
|
| 331 |
+
"learning_rate": 3.327777777777778e-05,
|
| 332 |
+
"loss": 1.5252,
|
| 333 |
+
"step": 2400
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 13.004354166666667,
|
| 337 |
+
"grad_norm": 31.665971755981445,
|
| 338 |
+
"learning_rate": 3.466666666666667e-05,
|
| 339 |
+
"loss": 1.3523,
|
| 340 |
+
"step": 2500
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 13.004993055555556,
|
| 344 |
+
"eval_accuracy": 0.591715976331361,
|
| 345 |
+
"eval_f1": 0.5601332592456852,
|
| 346 |
+
"eval_loss": 1.7261461019515991,
|
| 347 |
+
"eval_precision": 0.6065053536207382,
|
| 348 |
+
"eval_recall": 0.591715976331361,
|
| 349 |
+
"eval_runtime": 19.273,
|
| 350 |
+
"eval_samples_per_second": 17.537,
|
| 351 |
+
"eval_steps_per_second": 8.769,
|
| 352 |
+
"step": 2523
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 14.002125,
|
| 356 |
+
"grad_norm": 10.98436450958252,
|
| 357 |
+
"learning_rate": 3.605555555555556e-05,
|
| 358 |
+
"loss": 0.9208,
|
| 359 |
+
"step": 2600
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 14.004902777777778,
|
| 363 |
+
"grad_norm": 45.65860366821289,
|
| 364 |
+
"learning_rate": 3.7444444444444446e-05,
|
| 365 |
+
"loss": 0.7828,
|
| 366 |
+
"step": 2700
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 14.00498611111111,
|
| 370 |
+
"eval_accuracy": 0.6035502958579881,
|
| 371 |
+
"eval_f1": 0.5923542762300159,
|
| 372 |
+
"eval_loss": 1.6275168657302856,
|
| 373 |
+
"eval_precision": 0.6644219028834413,
|
| 374 |
+
"eval_recall": 0.6035502958579881,
|
| 375 |
+
"eval_runtime": 19.479,
|
| 376 |
+
"eval_samples_per_second": 17.352,
|
| 377 |
+
"eval_steps_per_second": 8.676,
|
| 378 |
+
"step": 2703
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 15.002673611111112,
|
| 382 |
+
"grad_norm": 24.026081085205078,
|
| 383 |
+
"learning_rate": 3.883333333333333e-05,
|
| 384 |
+
"loss": 0.4222,
|
| 385 |
+
"step": 2800
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 15.005006944444444,
|
| 389 |
+
"eval_accuracy": 0.6420118343195266,
|
| 390 |
+
"eval_f1": 0.6244853437161129,
|
| 391 |
+
"eval_loss": 1.5283950567245483,
|
| 392 |
+
"eval_precision": 0.67558435154589,
|
| 393 |
+
"eval_recall": 0.6420118343195266,
|
| 394 |
+
"eval_runtime": 19.8172,
|
| 395 |
+
"eval_samples_per_second": 17.056,
|
| 396 |
+
"eval_steps_per_second": 8.528,
|
| 397 |
+
"step": 2884
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 16.000444444444444,
|
| 401 |
+
"grad_norm": 11.13332748413086,
|
| 402 |
+
"learning_rate": 4.022222222222222e-05,
|
| 403 |
+
"loss": 0.4752,
|
| 404 |
+
"step": 2900
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 16.003222222222224,
|
| 408 |
+
"grad_norm": 8.170760154724121,
|
| 409 |
+
"learning_rate": 4.1611111111111114e-05,
|
| 410 |
+
"loss": 0.3113,
|
| 411 |
+
"step": 3000
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 16.005,
|
| 415 |
+
"eval_accuracy": 0.6272189349112426,
|
| 416 |
+
"eval_f1": 0.6092443552502724,
|
| 417 |
+
"eval_loss": 1.5459295511245728,
|
| 418 |
+
"eval_precision": 0.6664412510566357,
|
| 419 |
+
"eval_recall": 0.6272189349112426,
|
| 420 |
+
"eval_runtime": 19.569,
|
| 421 |
+
"eval_samples_per_second": 17.272,
|
| 422 |
+
"eval_steps_per_second": 8.636,
|
| 423 |
+
"step": 3064
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 17.000993055555554,
|
| 427 |
+
"grad_norm": 1.5062611103057861,
|
| 428 |
+
"learning_rate": 4.3e-05,
|
| 429 |
+
"loss": 0.2905,
|
| 430 |
+
"step": 3100
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 17.003770833333334,
|
| 434 |
+
"grad_norm": 1.255529761314392,
|
| 435 |
+
"learning_rate": 4.438888888888889e-05,
|
| 436 |
+
"loss": 0.2021,
|
| 437 |
+
"step": 3200
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 17.004993055555556,
|
| 441 |
+
"eval_accuracy": 0.665680473372781,
|
| 442 |
+
"eval_f1": 0.6484432398633582,
|
| 443 |
+
"eval_loss": 1.4440850019454956,
|
| 444 |
+
"eval_precision": 0.6991018596787828,
|
| 445 |
+
"eval_recall": 0.665680473372781,
|
| 446 |
+
"eval_runtime": 20.6892,
|
| 447 |
+
"eval_samples_per_second": 16.337,
|
| 448 |
+
"eval_steps_per_second": 8.168,
|
| 449 |
+
"step": 3244
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 18.001541666666668,
|
| 453 |
+
"grad_norm": 2.2327980995178223,
|
| 454 |
+
"learning_rate": 4.577777777777778e-05,
|
| 455 |
+
"loss": 0.1544,
|
| 456 |
+
"step": 3300
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 18.004319444444445,
|
| 460 |
+
"grad_norm": 1.6391757726669312,
|
| 461 |
+
"learning_rate": 4.716666666666667e-05,
|
| 462 |
+
"loss": 0.1698,
|
| 463 |
+
"step": 3400
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 18.004986111111112,
|
| 467 |
+
"eval_accuracy": 0.6124260355029586,
|
| 468 |
+
"eval_f1": 0.5942196856398039,
|
| 469 |
+
"eval_loss": 1.5340176820755005,
|
| 470 |
+
"eval_precision": 0.6510575962499039,
|
| 471 |
+
"eval_recall": 0.6124260355029586,
|
| 472 |
+
"eval_runtime": 19.0949,
|
| 473 |
+
"eval_samples_per_second": 17.701,
|
| 474 |
+
"eval_steps_per_second": 8.851,
|
| 475 |
+
"step": 3424
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 19.00209027777778,
|
| 479 |
+
"grad_norm": 7.581464767456055,
|
| 480 |
+
"learning_rate": 4.855555555555556e-05,
|
| 481 |
+
"loss": 0.0791,
|
| 482 |
+
"step": 3500
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 19.004868055555555,
|
| 486 |
+
"grad_norm": 0.5042413473129272,
|
| 487 |
+
"learning_rate": 4.994444444444445e-05,
|
| 488 |
+
"loss": 0.1199,
|
| 489 |
+
"step": 3600
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 19.005006944444446,
|
| 493 |
+
"eval_accuracy": 0.650887573964497,
|
| 494 |
+
"eval_f1": 0.6288250211327134,
|
| 495 |
+
"eval_loss": 1.3934611082077026,
|
| 496 |
+
"eval_precision": 0.6745703014933784,
|
| 497 |
+
"eval_recall": 0.650887573964497,
|
| 498 |
+
"eval_runtime": 18.5254,
|
| 499 |
+
"eval_samples_per_second": 18.245,
|
| 500 |
+
"eval_steps_per_second": 9.123,
|
| 501 |
+
"step": 3605
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 20.00263888888889,
|
| 505 |
+
"grad_norm": 0.40117892622947693,
|
| 506 |
+
"learning_rate": 4.9851851851851855e-05,
|
| 507 |
+
"loss": 0.0244,
|
| 508 |
+
"step": 3700
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 20.005,
|
| 512 |
+
"eval_accuracy": 0.6686390532544378,
|
| 513 |
+
"eval_f1": 0.6573669919823766,
|
| 514 |
+
"eval_loss": 1.478171706199646,
|
| 515 |
+
"eval_precision": 0.7130459284305437,
|
| 516 |
+
"eval_recall": 0.6686390532544378,
|
| 517 |
+
"eval_runtime": 19.5864,
|
| 518 |
+
"eval_samples_per_second": 17.257,
|
| 519 |
+
"eval_steps_per_second": 8.628,
|
| 520 |
+
"step": 3785
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 21.000409722222223,
|
| 524 |
+
"grad_norm": 0.6385509967803955,
|
| 525 |
+
"learning_rate": 4.969753086419753e-05,
|
| 526 |
+
"loss": 0.1644,
|
| 527 |
+
"step": 3800
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 21.0031875,
|
| 531 |
+
"grad_norm": 0.36917442083358765,
|
| 532 |
+
"learning_rate": 4.954320987654321e-05,
|
| 533 |
+
"loss": 0.0407,
|
| 534 |
+
"step": 3900
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 21.004993055555556,
|
| 538 |
+
"eval_accuracy": 0.6686390532544378,
|
| 539 |
+
"eval_f1": 0.6557107422492038,
|
| 540 |
+
"eval_loss": 1.388999581336975,
|
| 541 |
+
"eval_precision": 0.7149478726401803,
|
| 542 |
+
"eval_recall": 0.6686390532544378,
|
| 543 |
+
"eval_runtime": 19.0631,
|
| 544 |
+
"eval_samples_per_second": 17.731,
|
| 545 |
+
"eval_steps_per_second": 8.865,
|
| 546 |
+
"step": 3965
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 22.000958333333333,
|
| 550 |
+
"grad_norm": 0.5856227874755859,
|
| 551 |
+
"learning_rate": 4.938888888888889e-05,
|
| 552 |
+
"loss": 0.1314,
|
| 553 |
+
"step": 4000
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 22.00373611111111,
|
| 557 |
+
"grad_norm": 0.7390286326408386,
|
| 558 |
+
"learning_rate": 4.923611111111112e-05,
|
| 559 |
+
"loss": 0.0719,
|
| 560 |
+
"step": 4100
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 22.004986111111112,
|
| 564 |
+
"eval_accuracy": 0.6597633136094675,
|
| 565 |
+
"eval_f1": 0.6477249246480016,
|
| 566 |
+
"eval_loss": 1.4896520376205444,
|
| 567 |
+
"eval_precision": 0.7189349112426036,
|
| 568 |
+
"eval_recall": 0.6597633136094675,
|
| 569 |
+
"eval_runtime": 19.4131,
|
| 570 |
+
"eval_samples_per_second": 17.411,
|
| 571 |
+
"eval_steps_per_second": 8.705,
|
| 572 |
+
"step": 4145
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 23.001506944444444,
|
| 576 |
+
"grad_norm": 0.09171107411384583,
|
| 577 |
+
"learning_rate": 4.9081790123456794e-05,
|
| 578 |
+
"loss": 0.1134,
|
| 579 |
+
"step": 4200
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 23.004284722222224,
|
| 583 |
+
"grad_norm": 27.191944122314453,
|
| 584 |
+
"learning_rate": 4.892746913580247e-05,
|
| 585 |
+
"loss": 0.1163,
|
| 586 |
+
"step": 4300
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 23.005006944444446,
|
| 590 |
+
"eval_accuracy": 0.6715976331360947,
|
| 591 |
+
"eval_f1": 0.6638741600280061,
|
| 592 |
+
"eval_loss": 1.3919016122817993,
|
| 593 |
+
"eval_precision": 0.7218300929839392,
|
| 594 |
+
"eval_recall": 0.6715976331360947,
|
| 595 |
+
"eval_runtime": 18.0114,
|
| 596 |
+
"eval_samples_per_second": 18.766,
|
| 597 |
+
"eval_steps_per_second": 9.383,
|
| 598 |
+
"step": 4326
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 24.002055555555554,
|
| 602 |
+
"grad_norm": 0.14431528747081757,
|
| 603 |
+
"learning_rate": 4.877314814814815e-05,
|
| 604 |
+
"loss": 0.061,
|
| 605 |
+
"step": 4400
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 24.004833333333334,
|
| 609 |
+
"grad_norm": 33.57377624511719,
|
| 610 |
+
"learning_rate": 4.861882716049383e-05,
|
| 611 |
+
"loss": 0.1167,
|
| 612 |
+
"step": 4500
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 24.005,
|
| 616 |
+
"eval_accuracy": 0.6538461538461539,
|
| 617 |
+
"eval_f1": 0.6379572761229566,
|
| 618 |
+
"eval_loss": 1.569033145904541,
|
| 619 |
+
"eval_precision": 0.7189384333615103,
|
| 620 |
+
"eval_recall": 0.6538461538461539,
|
| 621 |
+
"eval_runtime": 18.5212,
|
| 622 |
+
"eval_samples_per_second": 18.249,
|
| 623 |
+
"eval_steps_per_second": 9.125,
|
| 624 |
+
"step": 4506
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 25.002604166666668,
|
| 628 |
+
"grad_norm": 0.03929486870765686,
|
| 629 |
+
"learning_rate": 4.8464506172839505e-05,
|
| 630 |
+
"loss": 0.0366,
|
| 631 |
+
"step": 4600
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 25.004993055555556,
|
| 635 |
+
"eval_accuracy": 0.6745562130177515,
|
| 636 |
+
"eval_f1": 0.6541387606476363,
|
| 637 |
+
"eval_loss": 1.5032190084457397,
|
| 638 |
+
"eval_precision": 0.6978867286559595,
|
| 639 |
+
"eval_recall": 0.6745562130177515,
|
| 640 |
+
"eval_runtime": 18.3411,
|
| 641 |
+
"eval_samples_per_second": 18.429,
|
| 642 |
+
"eval_steps_per_second": 9.214,
|
| 643 |
+
"step": 4686
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 26.000375,
|
| 647 |
+
"grad_norm": 0.10611555725336075,
|
| 648 |
+
"learning_rate": 4.831018518518518e-05,
|
| 649 |
+
"loss": 0.1361,
|
| 650 |
+
"step": 4700
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 26.00315277777778,
|
| 654 |
+
"grad_norm": 1.335172176361084,
|
| 655 |
+
"learning_rate": 4.815586419753087e-05,
|
| 656 |
+
"loss": 0.1065,
|
| 657 |
+
"step": 4800
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 26.004986111111112,
|
| 661 |
+
"eval_accuracy": 0.6390532544378699,
|
| 662 |
+
"eval_f1": 0.6134555615324846,
|
| 663 |
+
"eval_loss": 1.4892995357513428,
|
| 664 |
+
"eval_precision": 0.6475309946463793,
|
| 665 |
+
"eval_recall": 0.6390532544378699,
|
| 666 |
+
"eval_runtime": 19.1889,
|
| 667 |
+
"eval_samples_per_second": 17.614,
|
| 668 |
+
"eval_steps_per_second": 8.807,
|
| 669 |
+
"step": 4866
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 27.000923611111112,
|
| 673 |
+
"grad_norm": 0.25120797753334045,
|
| 674 |
+
"learning_rate": 4.8001543209876545e-05,
|
| 675 |
+
"loss": 0.1148,
|
| 676 |
+
"step": 4900
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 27.00370138888889,
|
| 680 |
+
"grad_norm": 0.08281093835830688,
|
| 681 |
+
"learning_rate": 4.784722222222223e-05,
|
| 682 |
+
"loss": 0.0454,
|
| 683 |
+
"step": 5000
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 27.005006944444446,
|
| 687 |
+
"eval_accuracy": 0.6242603550295858,
|
| 688 |
+
"eval_f1": 0.6022126952600326,
|
| 689 |
+
"eval_loss": 1.5013433694839478,
|
| 690 |
+
"eval_precision": 0.6600591715976332,
|
| 691 |
+
"eval_recall": 0.6242603550295858,
|
| 692 |
+
"eval_runtime": 20.5193,
|
| 693 |
+
"eval_samples_per_second": 16.472,
|
| 694 |
+
"eval_steps_per_second": 8.236,
|
| 695 |
+
"step": 5047
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 28.001472222222223,
|
| 699 |
+
"grad_norm": 0.05560300499200821,
|
| 700 |
+
"learning_rate": 4.769290123456791e-05,
|
| 701 |
+
"loss": 0.0973,
|
| 702 |
+
"step": 5100
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 28.00425,
|
| 706 |
+
"grad_norm": 1.4039995670318604,
|
| 707 |
+
"learning_rate": 4.7538580246913585e-05,
|
| 708 |
+
"loss": 0.0844,
|
| 709 |
+
"step": 5200
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 28.005,
|
| 713 |
+
"eval_accuracy": 0.6597633136094675,
|
| 714 |
+
"eval_f1": 0.6388020099558561,
|
| 715 |
+
"eval_loss": 1.5608867406845093,
|
| 716 |
+
"eval_precision": 0.6974077204846435,
|
| 717 |
+
"eval_recall": 0.6597633136094675,
|
| 718 |
+
"eval_runtime": 20.1446,
|
| 719 |
+
"eval_samples_per_second": 16.779,
|
| 720 |
+
"eval_steps_per_second": 8.389,
|
| 721 |
+
"step": 5227
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 29.002020833333333,
|
| 725 |
+
"grad_norm": 0.056026242673397064,
|
| 726 |
+
"learning_rate": 4.738425925925926e-05,
|
| 727 |
+
"loss": 0.0921,
|
| 728 |
+
"step": 5300
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 29.00479861111111,
|
| 732 |
+
"grad_norm": 0.39889276027679443,
|
| 733 |
+
"learning_rate": 4.722993827160494e-05,
|
| 734 |
+
"loss": 0.109,
|
| 735 |
+
"step": 5400
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 29.004993055555556,
|
| 739 |
+
"eval_accuracy": 0.665680473372781,
|
| 740 |
+
"eval_f1": 0.6506595311329039,
|
| 741 |
+
"eval_loss": 1.4839813709259033,
|
| 742 |
+
"eval_precision": 0.7151002843310535,
|
| 743 |
+
"eval_recall": 0.665680473372781,
|
| 744 |
+
"eval_runtime": 18.7473,
|
| 745 |
+
"eval_samples_per_second": 18.029,
|
| 746 |
+
"eval_steps_per_second": 9.015,
|
| 747 |
+
"step": 5407
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 30.002569444444443,
|
| 751 |
+
"grad_norm": 0.05277135968208313,
|
| 752 |
+
"learning_rate": 4.707561728395062e-05,
|
| 753 |
+
"loss": 0.1508,
|
| 754 |
+
"step": 5500
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 30.004986111111112,
|
| 758 |
+
"eval_accuracy": 0.6035502958579881,
|
| 759 |
+
"eval_f1": 0.590259690999336,
|
| 760 |
+
"eval_loss": 1.801746129989624,
|
| 761 |
+
"eval_precision": 0.6784164553395322,
|
| 762 |
+
"eval_recall": 0.6035502958579881,
|
| 763 |
+
"eval_runtime": 17.6073,
|
| 764 |
+
"eval_samples_per_second": 19.197,
|
| 765 |
+
"eval_steps_per_second": 9.598,
|
| 766 |
+
"step": 5587
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 31.000340277777777,
|
| 770 |
+
"grad_norm": 0.6189529895782471,
|
| 771 |
+
"learning_rate": 4.6921296296296296e-05,
|
| 772 |
+
"loss": 0.0796,
|
| 773 |
+
"step": 5600
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 31.003118055555557,
|
| 777 |
+
"grad_norm": 0.06006496772170067,
|
| 778 |
+
"learning_rate": 4.676697530864197e-05,
|
| 779 |
+
"loss": 0.1114,
|
| 780 |
+
"step": 5700
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 31.005006944444446,
|
| 784 |
+
"eval_accuracy": 0.6390532544378699,
|
| 785 |
+
"eval_f1": 0.613423755731448,
|
| 786 |
+
"eval_loss": 1.6675891876220703,
|
| 787 |
+
"eval_precision": 0.6721329952099183,
|
| 788 |
+
"eval_recall": 0.6390532544378699,
|
| 789 |
+
"eval_runtime": 17.8113,
|
| 790 |
+
"eval_samples_per_second": 18.977,
|
| 791 |
+
"eval_steps_per_second": 9.488,
|
| 792 |
+
"step": 5768
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 32.00088888888889,
|
| 796 |
+
"grad_norm": 0.04244798794388771,
|
| 797 |
+
"learning_rate": 4.661265432098766e-05,
|
| 798 |
+
"loss": 0.1236,
|
| 799 |
+
"step": 5800
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 32.00366666666667,
|
| 803 |
+
"grad_norm": 0.24667857587337494,
|
| 804 |
+
"learning_rate": 4.6458333333333335e-05,
|
| 805 |
+
"loss": 0.0931,
|
| 806 |
+
"step": 5900
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 32.005,
|
| 810 |
+
"eval_accuracy": 0.6745562130177515,
|
| 811 |
+
"eval_f1": 0.6520330097253173,
|
| 812 |
+
"eval_loss": 1.534504771232605,
|
| 813 |
+
"eval_precision": 0.7082065370526909,
|
| 814 |
+
"eval_recall": 0.6745562130177515,
|
| 815 |
+
"eval_runtime": 16.9203,
|
| 816 |
+
"eval_samples_per_second": 19.976,
|
| 817 |
+
"eval_steps_per_second": 9.988,
|
| 818 |
+
"step": 5948
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 33.0014375,
|
| 822 |
+
"grad_norm": 0.04040142521262169,
|
| 823 |
+
"learning_rate": 4.630401234567901e-05,
|
| 824 |
+
"loss": 0.0767,
|
| 825 |
+
"step": 6000
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 33.004215277777774,
|
| 829 |
+
"grad_norm": 0.18193276226520538,
|
| 830 |
+
"learning_rate": 4.614969135802469e-05,
|
| 831 |
+
"loss": 0.0619,
|
| 832 |
+
"step": 6100
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 33.00499305555556,
|
| 836 |
+
"eval_accuracy": 0.6301775147928994,
|
| 837 |
+
"eval_f1": 0.6008137161983316,
|
| 838 |
+
"eval_loss": 1.746155858039856,
|
| 839 |
+
"eval_precision": 0.6424168779938011,
|
| 840 |
+
"eval_recall": 0.6301775147928994,
|
| 841 |
+
"eval_runtime": 18.9428,
|
| 842 |
+
"eval_samples_per_second": 17.843,
|
| 843 |
+
"eval_steps_per_second": 8.922,
|
| 844 |
+
"step": 6128
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 34.00198611111111,
|
| 848 |
+
"grad_norm": 0.009634270332753658,
|
| 849 |
+
"learning_rate": 4.599691358024691e-05,
|
| 850 |
+
"loss": 0.1314,
|
| 851 |
+
"step": 6200
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 34.00476388888889,
|
| 855 |
+
"grad_norm": 0.21367499232292175,
|
| 856 |
+
"learning_rate": 4.584259259259259e-05,
|
| 857 |
+
"loss": 0.2698,
|
| 858 |
+
"step": 6300
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 34.00498611111111,
|
| 862 |
+
"eval_accuracy": 0.6331360946745562,
|
| 863 |
+
"eval_f1": 0.6126230606999838,
|
| 864 |
+
"eval_loss": 1.7031519412994385,
|
| 865 |
+
"eval_precision": 0.6710693153000845,
|
| 866 |
+
"eval_recall": 0.6331360946745562,
|
| 867 |
+
"eval_runtime": 19.9792,
|
| 868 |
+
"eval_samples_per_second": 16.918,
|
| 869 |
+
"eval_steps_per_second": 8.459,
|
| 870 |
+
"step": 6308
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 35.00253472222222,
|
| 874 |
+
"grad_norm": 0.03326309099793434,
|
| 875 |
+
"learning_rate": 4.5688271604938275e-05,
|
| 876 |
+
"loss": 0.1108,
|
| 877 |
+
"step": 6400
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 35.005006944444446,
|
| 881 |
+
"eval_accuracy": 0.6538461538461539,
|
| 882 |
+
"eval_f1": 0.62648739111679,
|
| 883 |
+
"eval_loss": 1.7694625854492188,
|
| 884 |
+
"eval_precision": 0.6784205193228862,
|
| 885 |
+
"eval_recall": 0.6538461538461539,
|
| 886 |
+
"eval_runtime": 18.3538,
|
| 887 |
+
"eval_samples_per_second": 18.416,
|
| 888 |
+
"eval_steps_per_second": 9.208,
|
| 889 |
+
"step": 6489
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 36.000305555555556,
|
| 893 |
+
"grad_norm": 104.1079330444336,
|
| 894 |
+
"learning_rate": 4.553395061728395e-05,
|
| 895 |
+
"loss": 0.1261,
|
| 896 |
+
"step": 6500
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 36.003083333333336,
|
| 900 |
+
"grad_norm": 0.1133200004696846,
|
| 901 |
+
"learning_rate": 4.537962962962963e-05,
|
| 902 |
+
"loss": 0.1006,
|
| 903 |
+
"step": 6600
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 36.005,
|
| 907 |
+
"eval_accuracy": 0.5828402366863905,
|
| 908 |
+
"eval_f1": 0.5661115757104835,
|
| 909 |
+
"eval_loss": 2.018780469894409,
|
| 910 |
+
"eval_precision": 0.6289293300387975,
|
| 911 |
+
"eval_recall": 0.5828402366863905,
|
| 912 |
+
"eval_runtime": 17.6731,
|
| 913 |
+
"eval_samples_per_second": 19.125,
|
| 914 |
+
"eval_steps_per_second": 9.563,
|
| 915 |
+
"step": 6669
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 37.00085416666667,
|
| 919 |
+
"grad_norm": 0.024959033355116844,
|
| 920 |
+
"learning_rate": 4.5225308641975314e-05,
|
| 921 |
+
"loss": 0.1183,
|
| 922 |
+
"step": 6700
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 37.00363194444444,
|
| 926 |
+
"grad_norm": 0.09667906910181046,
|
| 927 |
+
"learning_rate": 4.507098765432099e-05,
|
| 928 |
+
"loss": 0.0823,
|
| 929 |
+
"step": 6800
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 37.00499305555556,
|
| 933 |
+
"eval_accuracy": 0.6568047337278107,
|
| 934 |
+
"eval_f1": 0.6425439452066671,
|
| 935 |
+
"eval_loss": 1.6487033367156982,
|
| 936 |
+
"eval_precision": 0.6874260355029587,
|
| 937 |
+
"eval_recall": 0.6568047337278107,
|
| 938 |
+
"eval_runtime": 20.3974,
|
| 939 |
+
"eval_samples_per_second": 16.571,
|
| 940 |
+
"eval_steps_per_second": 8.285,
|
| 941 |
+
"step": 6849
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 38.00140277777778,
|
| 945 |
+
"grad_norm": 0.025008585304021835,
|
| 946 |
+
"learning_rate": 4.491666666666667e-05,
|
| 947 |
+
"loss": 0.1114,
|
| 948 |
+
"step": 6900
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 38.00418055555556,
|
| 952 |
+
"grad_norm": 3.3624982833862305,
|
| 953 |
+
"learning_rate": 4.476234567901235e-05,
|
| 954 |
+
"loss": 0.0632,
|
| 955 |
+
"step": 7000
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 38.00498611111111,
|
| 959 |
+
"eval_accuracy": 0.636094674556213,
|
| 960 |
+
"eval_f1": 0.6253254280582216,
|
| 961 |
+
"eval_loss": 1.8013869524002075,
|
| 962 |
+
"eval_precision": 0.691735347985348,
|
| 963 |
+
"eval_recall": 0.636094674556213,
|
| 964 |
+
"eval_runtime": 18.4919,
|
| 965 |
+
"eval_samples_per_second": 18.278,
|
| 966 |
+
"eval_steps_per_second": 9.139,
|
| 967 |
+
"step": 7029
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 39.00195138888889,
|
| 971 |
+
"grad_norm": 0.028111210092902184,
|
| 972 |
+
"learning_rate": 4.4608024691358025e-05,
|
| 973 |
+
"loss": 0.0691,
|
| 974 |
+
"step": 7100
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 39.004729166666664,
|
| 978 |
+
"grad_norm": 0.022145342081785202,
|
| 979 |
+
"learning_rate": 4.445370370370371e-05,
|
| 980 |
+
"loss": 0.1162,
|
| 981 |
+
"step": 7200
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 39.005006944444446,
|
| 985 |
+
"eval_accuracy": 0.6449704142011834,
|
| 986 |
+
"eval_f1": 0.6195841797912803,
|
| 987 |
+
"eval_loss": 1.67414128780365,
|
| 988 |
+
"eval_precision": 0.6671691556306941,
|
| 989 |
+
"eval_recall": 0.6449704142011834,
|
| 990 |
+
"eval_runtime": 19.1991,
|
| 991 |
+
"eval_samples_per_second": 17.605,
|
| 992 |
+
"eval_steps_per_second": 8.802,
|
| 993 |
+
"step": 7210
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 40.0025,
|
| 997 |
+
"grad_norm": 0.012291846796870232,
|
| 998 |
+
"learning_rate": 4.429938271604939e-05,
|
| 999 |
+
"loss": 0.0846,
|
| 1000 |
+
"step": 7300
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 40.005,
|
| 1004 |
+
"eval_accuracy": 0.636094674556213,
|
| 1005 |
+
"eval_f1": 0.6205015070399685,
|
| 1006 |
+
"eval_loss": 1.8031796216964722,
|
| 1007 |
+
"eval_precision": 0.6948467692900892,
|
| 1008 |
+
"eval_recall": 0.636094674556213,
|
| 1009 |
+
"eval_runtime": 19.3607,
|
| 1010 |
+
"eval_samples_per_second": 17.458,
|
| 1011 |
+
"eval_steps_per_second": 8.729,
|
| 1012 |
+
"step": 7390
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 41.00027083333333,
|
| 1016 |
+
"grad_norm": 0.45003169775009155,
|
| 1017 |
+
"learning_rate": 4.4145061728395065e-05,
|
| 1018 |
+
"loss": 0.146,
|
| 1019 |
+
"step": 7400
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 41.00304861111111,
|
| 1023 |
+
"grad_norm": 4.705352306365967,
|
| 1024 |
+
"learning_rate": 4.399074074074074e-05,
|
| 1025 |
+
"loss": 0.1528,
|
| 1026 |
+
"step": 7500
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 41.00499305555556,
|
| 1030 |
+
"eval_accuracy": 0.6331360946745562,
|
| 1031 |
+
"eval_f1": 0.6101778563317025,
|
| 1032 |
+
"eval_loss": 1.8375493288040161,
|
| 1033 |
+
"eval_precision": 0.673196675119752,
|
| 1034 |
+
"eval_recall": 0.6331360946745562,
|
| 1035 |
+
"eval_runtime": 17.7659,
|
| 1036 |
+
"eval_samples_per_second": 19.025,
|
| 1037 |
+
"eval_steps_per_second": 9.513,
|
| 1038 |
+
"step": 7570
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 42.000819444444446,
|
| 1042 |
+
"grad_norm": 94.2012939453125,
|
| 1043 |
+
"learning_rate": 4.383641975308642e-05,
|
| 1044 |
+
"loss": 0.1548,
|
| 1045 |
+
"step": 7600
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 42.003597222222226,
|
| 1049 |
+
"grad_norm": 0.008578118868172169,
|
| 1050 |
+
"learning_rate": 4.36820987654321e-05,
|
| 1051 |
+
"loss": 0.0695,
|
| 1052 |
+
"step": 7700
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 42.00498611111111,
|
| 1056 |
+
"eval_accuracy": 0.6568047337278107,
|
| 1057 |
+
"eval_f1": 0.6417349744272821,
|
| 1058 |
+
"eval_loss": 1.672219157218933,
|
| 1059 |
+
"eval_precision": 0.7030008453085377,
|
| 1060 |
+
"eval_recall": 0.6568047337278107,
|
| 1061 |
+
"eval_runtime": 18.0993,
|
| 1062 |
+
"eval_samples_per_second": 18.675,
|
| 1063 |
+
"eval_steps_per_second": 9.337,
|
| 1064 |
+
"step": 7750
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 43.00136805555555,
|
| 1068 |
+
"grad_norm": 0.005858391989022493,
|
| 1069 |
+
"learning_rate": 4.3527777777777776e-05,
|
| 1070 |
+
"loss": 0.0513,
|
| 1071 |
+
"step": 7800
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 43.00414583333333,
|
| 1075 |
+
"grad_norm": 0.24502906203269958,
|
| 1076 |
+
"learning_rate": 4.3375000000000004e-05,
|
| 1077 |
+
"loss": 0.1516,
|
| 1078 |
+
"step": 7900
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 43.005006944444446,
|
| 1082 |
+
"eval_accuracy": 0.6715976331360947,
|
| 1083 |
+
"eval_f1": 0.6494567398413551,
|
| 1084 |
+
"eval_loss": 1.7810900211334229,
|
| 1085 |
+
"eval_precision": 0.7009087066779375,
|
| 1086 |
+
"eval_recall": 0.6715976331360947,
|
| 1087 |
+
"eval_runtime": 18.6526,
|
| 1088 |
+
"eval_samples_per_second": 18.121,
|
| 1089 |
+
"eval_steps_per_second": 9.06,
|
| 1090 |
+
"step": 7931
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 44.001916666666666,
|
| 1094 |
+
"grad_norm": 0.023847075179219246,
|
| 1095 |
+
"learning_rate": 4.322067901234568e-05,
|
| 1096 |
+
"loss": 0.1172,
|
| 1097 |
+
"step": 8000
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 44.004694444444446,
|
| 1101 |
+
"grad_norm": 0.1472051441669464,
|
| 1102 |
+
"learning_rate": 4.306635802469136e-05,
|
| 1103 |
+
"loss": 0.1565,
|
| 1104 |
+
"step": 8100
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 44.005,
|
| 1108 |
+
"eval_accuracy": 0.6538461538461539,
|
| 1109 |
+
"eval_f1": 0.6346678469459534,
|
| 1110 |
+
"eval_loss": 1.8077248334884644,
|
| 1111 |
+
"eval_precision": 0.6883840518455903,
|
| 1112 |
+
"eval_recall": 0.6538461538461539,
|
| 1113 |
+
"eval_runtime": 18.0276,
|
| 1114 |
+
"eval_samples_per_second": 18.749,
|
| 1115 |
+
"eval_steps_per_second": 9.375,
|
| 1116 |
+
"step": 8111
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 45.00246527777778,
|
| 1120 |
+
"grad_norm": 0.013580802828073502,
|
| 1121 |
+
"learning_rate": 4.291203703703704e-05,
|
| 1122 |
+
"loss": 0.0728,
|
| 1123 |
+
"step": 8200
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 45.00499305555556,
|
| 1127 |
+
"eval_accuracy": 0.6272189349112426,
|
| 1128 |
+
"eval_f1": 0.604730419153496,
|
| 1129 |
+
"eval_loss": 1.7950366735458374,
|
| 1130 |
+
"eval_precision": 0.6755196480048551,
|
| 1131 |
+
"eval_recall": 0.6272189349112426,
|
| 1132 |
+
"eval_runtime": 18.0454,
|
| 1133 |
+
"eval_samples_per_second": 18.731,
|
| 1134 |
+
"eval_steps_per_second": 9.365,
|
| 1135 |
+
"step": 8291
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 45.00499305555556,
|
| 1139 |
+
"step": 8291,
|
| 1140 |
+
"total_flos": 1.700902027656954e+20,
|
| 1141 |
+
"train_loss": 3.3378389804185824,
|
| 1142 |
+
"train_runtime": 8591.5525,
|
| 1143 |
+
"train_samples_per_second": 33.521,
|
| 1144 |
+
"train_steps_per_second": 4.19
|
| 1145 |
+
}
|
| 1146 |
+
],
|
| 1147 |
+
"logging_steps": 100,
|
| 1148 |
+
"max_steps": 36000,
|
| 1149 |
+
"num_input_tokens_seen": 0,
|
| 1150 |
+
"num_train_epochs": 9223372036854775807,
|
| 1151 |
+
"save_steps": 500,
|
| 1152 |
+
"stateful_callbacks": {
|
| 1153 |
+
"EarlyStoppingCallback": {
|
| 1154 |
+
"args": {
|
| 1155 |
+
"early_stopping_patience": 20,
|
| 1156 |
+
"early_stopping_threshold": 0.0
|
| 1157 |
+
},
|
| 1158 |
+
"attributes": {
|
| 1159 |
+
"early_stopping_patience_counter": 20
|
| 1160 |
+
}
|
| 1161 |
+
},
|
| 1162 |
+
"TrainerControl": {
|
| 1163 |
+
"args": {
|
| 1164 |
+
"should_epoch_stop": false,
|
| 1165 |
+
"should_evaluate": false,
|
| 1166 |
+
"should_log": false,
|
| 1167 |
+
"should_save": true,
|
| 1168 |
+
"should_training_stop": true
|
| 1169 |
+
},
|
| 1170 |
+
"attributes": {}
|
| 1171 |
+
}
|
| 1172 |
+
},
|
| 1173 |
+
"total_flos": 1.700902027656954e+20,
|
| 1174 |
+
"train_batch_size": 2,
|
| 1175 |
+
"trial_name": null,
|
| 1176 |
+
"trial_params": null
|
| 1177 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:850dd5b831c3ea1fe83ed0256b2e2be148b8a875bd7de7f8442cdb88594d573f
|
| 3 |
+
size 5304
|