Spaces:
Runtime error
Runtime error
aldan.creo
commited on
Commit
·
4db55cd
1
Parent(s):
b448895
WIP
Browse files- app.py +46 -22
- requirements.txt +0 -1
app.py
CHANGED
@@ -1,6 +1,7 @@
|
|
1 |
import json
|
2 |
import logging
|
3 |
import os
|
|
|
4 |
from functools import partial
|
5 |
|
6 |
import gradio as gr
|
@@ -14,9 +15,11 @@ logger.setLevel(logging.INFO)
|
|
14 |
load_dotenv()
|
15 |
|
16 |
# dataset = load_dataset("detection-datasets/coco")
|
17 |
-
it_dataset =
|
18 |
-
"imagenet-1k", split="train", streaming=True, trust_remote_code=True
|
19 |
-
|
|
|
|
|
20 |
|
21 |
|
22 |
def gen_from_iterable_dataset(iterable_ds):
|
@@ -26,6 +29,10 @@ def gen_from_iterable_dataset(iterable_ds):
|
|
26 |
yield from iterable_ds
|
27 |
|
28 |
|
|
|
|
|
|
|
|
|
29 |
# imagenet_categories_data.json is a JSON file containing a hierarchy of ImageNet categories.
|
30 |
# We want to take all categories under "artifact, artefact".
|
31 |
# Each node has this structure:
|
@@ -58,14 +65,17 @@ def filter_imgs_by_label(x):
|
|
58 |
"""
|
59 |
Filter out the images that have label -1
|
60 |
"""
|
|
|
61 |
return x["label"] in artifact_categories
|
62 |
|
63 |
|
64 |
-
|
65 |
-
|
66 |
-
|
67 |
-
)
|
68 |
-
|
|
|
|
|
69 |
|
70 |
|
71 |
def get_user_prompt():
|
@@ -74,15 +84,11 @@ def get_user_prompt():
|
|
74 |
machine_labels = []
|
75 |
human_labels = []
|
76 |
for i in range(3):
|
77 |
-
data =
|
78 |
-
logger.info(f"Data: {data}")
|
79 |
images.append(data["image"])
|
80 |
# Get the label as a human readable string
|
81 |
machine_labels.append(data["label"])
|
82 |
-
|
83 |
-
human_label = dataset.features["label"].int2str(data["label"]) + str(
|
84 |
-
data["label"]
|
85 |
-
)
|
86 |
human_labels.append(human_label)
|
87 |
return {
|
88 |
"images": images,
|
@@ -94,7 +100,7 @@ def get_user_prompt():
|
|
94 |
hf_writer = gr.HuggingFaceDatasetSaver(
|
95 |
hf_token=os.environ["HF_TOKEN"], dataset_name="acmc/maker-faire-bot", private=True
|
96 |
)
|
97 |
-
csv_writer = gr.CSVLogger(
|
98 |
|
99 |
theme = gr.themes.Default(primary_hue="cyan", secondary_hue="fuchsia")
|
100 |
|
@@ -137,15 +143,14 @@ with gr.Blocks(theme=theme) as demo:
|
|
137 |
btn = gr.Button("Change", variant="secondary")
|
138 |
|
139 |
def change_image(user_prompt):
|
140 |
-
|
141 |
-
|
142 |
user_prompt = user_prompt.copy()
|
143 |
user_prompt["images"][i] = data["image"]
|
144 |
user_prompt["machine_labels"][i] = data["label"]
|
145 |
user_prompt["human_labels"][i] = dataset.features["label"].int2str(
|
146 |
data["label"]
|
147 |
)
|
148 |
-
logger.info(user_prompt)
|
149 |
return (
|
150 |
user_prompt,
|
151 |
user_prompt["images"][i],
|
@@ -192,20 +197,39 @@ with gr.Blocks(theme=theme) as demo:
|
|
192 |
submit_btn = gr.Button("Submit", variant="primary")
|
193 |
|
194 |
def log_results(prompt, object, explanation):
|
195 |
-
|
196 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
197 |
|
198 |
submit_btn.click(
|
199 |
log_results,
|
200 |
inputs=[user_prompt, user_answer_object, user_answer_explanation],
|
201 |
-
preprocess=
|
202 |
)
|
203 |
|
204 |
new_prompt_btn = gr.Button("New Prompt", variant="secondary")
|
205 |
new_prompt_btn.click(
|
206 |
get_user_prompt,
|
207 |
outputs=[user_prompt],
|
208 |
-
preprocess=
|
209 |
)
|
210 |
|
211 |
gr.Markdown(
|
|
|
1 |
import json
|
2 |
import logging
|
3 |
import os
|
4 |
+
import random
|
5 |
from functools import partial
|
6 |
|
7 |
import gradio as gr
|
|
|
15 |
load_dotenv()
|
16 |
|
17 |
# dataset = load_dataset("detection-datasets/coco")
|
18 |
+
it_dataset = (
|
19 |
+
load_dataset("imagenet-1k", split="train", streaming=True, trust_remote_code=True)
|
20 |
+
.shuffle(42)
|
21 |
+
.take(1000)
|
22 |
+
)
|
23 |
|
24 |
|
25 |
def gen_from_iterable_dataset(iterable_ds):
|
|
|
29 |
yield from iterable_ds
|
30 |
|
31 |
|
32 |
+
dataset = Dataset.from_generator(
|
33 |
+
partial(gen_from_iterable_dataset, it_dataset), features=it_dataset.features
|
34 |
+
)
|
35 |
+
|
36 |
# imagenet_categories_data.json is a JSON file containing a hierarchy of ImageNet categories.
|
37 |
# We want to take all categories under "artifact, artefact".
|
38 |
# Each node has this structure:
|
|
|
65 |
"""
|
66 |
Filter out the images that have label -1
|
67 |
"""
|
68 |
+
logger.info(f'label: {x["label"]} (present: {x["label"] in artifact_categories})')
|
69 |
return x["label"] in artifact_categories
|
70 |
|
71 |
|
72 |
+
dataset = dataset.filter(filter_imgs_by_label)
|
73 |
+
|
74 |
+
logging.basicConfig(level=logging.INFO)
|
75 |
+
logger = logging.getLogger(__name__)
|
76 |
+
logger.setLevel(logging.INFO)
|
77 |
+
|
78 |
+
load_dotenv()
|
79 |
|
80 |
|
81 |
def get_user_prompt():
|
|
|
84 |
machine_labels = []
|
85 |
human_labels = []
|
86 |
for i in range(3):
|
87 |
+
data = dataset[random.randint(0, len(dataset) - 1)]
|
|
|
88 |
images.append(data["image"])
|
89 |
# Get the label as a human readable string
|
90 |
machine_labels.append(data["label"])
|
91 |
+
human_label = dataset.features["label"].int2str(data["label"])
|
|
|
|
|
|
|
92 |
human_labels.append(human_label)
|
93 |
return {
|
94 |
"images": images,
|
|
|
100 |
hf_writer = gr.HuggingFaceDatasetSaver(
|
101 |
hf_token=os.environ["HF_TOKEN"], dataset_name="acmc/maker-faire-bot", private=True
|
102 |
)
|
103 |
+
csv_writer = gr.CSVLogger()
|
104 |
|
105 |
theme = gr.themes.Default(primary_hue="cyan", secondary_hue="fuchsia")
|
106 |
|
|
|
143 |
btn = gr.Button("Change", variant="secondary")
|
144 |
|
145 |
def change_image(user_prompt):
|
146 |
+
logger.info(f"Current user prompt: {user_prompt}")
|
147 |
+
data = dataset[random.randint(0, len(dataset) - 1)]
|
148 |
user_prompt = user_prompt.copy()
|
149 |
user_prompt["images"][i] = data["image"]
|
150 |
user_prompt["machine_labels"][i] = data["label"]
|
151 |
user_prompt["human_labels"][i] = dataset.features["label"].int2str(
|
152 |
data["label"]
|
153 |
)
|
|
|
154 |
return (
|
155 |
user_prompt,
|
156 |
user_prompt["images"][i],
|
|
|
197 |
submit_btn = gr.Button("Submit", variant="primary")
|
198 |
|
199 |
def log_results(prompt, object, explanation):
|
200 |
+
logger.info(f"logging - Prompt: {prompt}")
|
201 |
+
csv_writer.flag(
|
202 |
+
[
|
203 |
+
{
|
204 |
+
"machine_labels": prompt["machine_labels"],
|
205 |
+
"human_labels": prompt["human_labels"],
|
206 |
+
},
|
207 |
+
object,
|
208 |
+
explanation,
|
209 |
+
]
|
210 |
+
)
|
211 |
+
hf_writer.flag(
|
212 |
+
[
|
213 |
+
{
|
214 |
+
"machine_labels": prompt["machine_labels"],
|
215 |
+
"human_labels": prompt["human_labels"],
|
216 |
+
},
|
217 |
+
object,
|
218 |
+
explanation,
|
219 |
+
]
|
220 |
+
)
|
221 |
|
222 |
submit_btn.click(
|
223 |
log_results,
|
224 |
inputs=[user_prompt, user_answer_object, user_answer_explanation],
|
225 |
+
preprocess=True,
|
226 |
)
|
227 |
|
228 |
new_prompt_btn = gr.Button("New Prompt", variant="secondary")
|
229 |
new_prompt_btn.click(
|
230 |
get_user_prompt,
|
231 |
outputs=[user_prompt],
|
232 |
+
# preprocess=True,
|
233 |
)
|
234 |
|
235 |
gr.Markdown(
|
requirements.txt
CHANGED
@@ -1,3 +1,2 @@
|
|
1 |
datasets==2.19.0
|
2 |
-
gradio==4.28.0
|
3 |
python-dotenv==1.0.1
|
|
|
1 |
datasets==2.19.0
|
|
|
2 |
python-dotenv==1.0.1
|