Commit
·
c1bc2bf
1
Parent(s):
d7041cd
test
Browse files
app.py
CHANGED
|
@@ -61,13 +61,7 @@ def get_dataset_csv(
|
|
| 61 |
):
|
| 62 |
df = ORIGINAL_DF[ORIGINAL_DF['Size'].isin(model_size)]
|
| 63 |
df = df.drop(columns="Size")
|
| 64 |
-
|
| 65 |
-
# if metric_choice != "None":
|
| 66 |
-
# metric_choice = metric_choice + "/std"
|
| 67 |
-
# sort_basis = df[metric_choice].apply(format_csv_numbers)
|
| 68 |
-
# sorted_indices = sort_basis.argsort()[::-1]
|
| 69 |
-
# df = df.iloc[sorted_indices]
|
| 70 |
-
|
| 71 |
leaderboard_table = gr.components.Dataframe(
|
| 72 |
value=df,
|
| 73 |
interactive=False,
|
|
@@ -81,12 +75,6 @@ def get_dataset_csv_per(
|
|
| 81 |
df = ORIGINAL_DF_PER[ORIGINAL_DF_PER['Size'].isin(model_size)]
|
| 82 |
df = df.drop(columns="Size")
|
| 83 |
|
| 84 |
-
# if metric_choice != "None":
|
| 85 |
-
# metric_choice = metric_choice + "/std"
|
| 86 |
-
# sort_basis = df[metric_choice].apply(format_csv_numbers)
|
| 87 |
-
# sorted_indices = sort_basis.argsort()[::-1]
|
| 88 |
-
# df = df.iloc[sorted_indices]
|
| 89 |
-
|
| 90 |
leaderboard_table = gr.components.Dataframe(
|
| 91 |
value=df,
|
| 92 |
interactive=False,
|
|
@@ -106,16 +94,6 @@ def get_dataset_csv_sub_gen(
|
|
| 106 |
subclass_choice_label = ["Model", subclass_choice+"_Accuracy", subclass_choice+"_Precision", subclass_choice+"_Recall"]
|
| 107 |
df = df[subclass_choice_label]
|
| 108 |
|
| 109 |
-
# if metric_choice != "None":
|
| 110 |
-
# # metric_choice = metric_choice + "/std"
|
| 111 |
-
# metric_choice = metric_choice.split("_")[0]
|
| 112 |
-
# metric_choice = subclass_choice + "_" + metric_choice
|
| 113 |
-
# # sort_basis = df[metric_choice].apply(format_csv_numbers)
|
| 114 |
-
# sort_basis = df[metric_choice]
|
| 115 |
-
|
| 116 |
-
# sorted_indices = sort_basis.argsort()[::-1]
|
| 117 |
-
# df = df.iloc[sorted_indices]
|
| 118 |
-
|
| 119 |
leaderboard_table = gr.components.Dataframe(
|
| 120 |
value=df,
|
| 121 |
interactive=False,
|
|
@@ -135,16 +113,6 @@ def get_dataset_csv_sub_per(
|
|
| 135 |
subclass_choice_label = ["Model", subclass_choice+"_Accuracy", subclass_choice+"_Precision", subclass_choice+"_Recall"]
|
| 136 |
df = df[subclass_choice_label]
|
| 137 |
|
| 138 |
-
# if metric_choice != "None":
|
| 139 |
-
# # metric_choice = metric_choice + "/std"
|
| 140 |
-
# metric_choice = metric_choice.split("_")[0]
|
| 141 |
-
# metric_choice = subclass_choice + "_" + metric_choice
|
| 142 |
-
# # sort_basis = df[metric_choice].apply(format_csv_numbers)
|
| 143 |
-
# sort_basis = df[metric_choice]
|
| 144 |
-
|
| 145 |
-
# sorted_indices = sort_basis.argsort()[::-1]
|
| 146 |
-
# df = df.iloc[sorted_indices]
|
| 147 |
-
|
| 148 |
leaderboard_table = gr.components.Dataframe(
|
| 149 |
value=df,
|
| 150 |
interactive=False,
|
|
@@ -197,14 +165,6 @@ with gr.Blocks() as demo:
|
|
| 197 |
info="Please choose the type to display.",
|
| 198 |
)
|
| 199 |
|
| 200 |
-
# with gr.Column(scale=0.8):
|
| 201 |
-
# metric_choice = gr.Dropdown(
|
| 202 |
-
# choices=METRICS,
|
| 203 |
-
# value="None",
|
| 204 |
-
# label="Metric",
|
| 205 |
-
# info="Please choose the metric to display.",
|
| 206 |
-
# )
|
| 207 |
-
|
| 208 |
with gr.Column(scale=10):
|
| 209 |
model_choice = gr.CheckboxGroup(
|
| 210 |
choices=CLASSIFICATION["model_size"],
|
|
@@ -213,40 +173,8 @@ with gr.Blocks() as demo:
|
|
| 213 |
info="Please choose the model size to display.",
|
| 214 |
)
|
| 215 |
|
| 216 |
-
|
| 217 |
-
# with gr.Column(scale=0.8):
|
| 218 |
-
# subclass_choice = gr.Dropdown(
|
| 219 |
-
# choices=SUBCLASS,
|
| 220 |
-
# value="Discrimination",
|
| 221 |
-
# label="Subclass",
|
| 222 |
-
# info="Please choose the subclass to display.",
|
| 223 |
-
# )
|
| 224 |
-
|
| 225 |
-
|
| 226 |
-
#👉 this part is for csv table generatived
|
| 227 |
with gr.Tabs(elem_classes="tab-buttons") as tabs:
|
| 228 |
-
|
| 229 |
-
# with gr.TabItem("🏅 Overall Generatived", elem_id="od-benchmark-tab-table", id=1):
|
| 230 |
-
# dataframe = gr.components.Dataframe(
|
| 231 |
-
# elem_id="leaderboard-table",
|
| 232 |
-
# )
|
| 233 |
-
# #👉 this part is for csv table perplexity
|
| 234 |
-
# with gr.TabItem("🏅 Overall Perplexity", elem_id="od-benchmark-tab-table", id=2):
|
| 235 |
-
# datafram_per = gr.components.Dataframe(
|
| 236 |
-
# elem_id="leaderboard-table",
|
| 237 |
-
# )
|
| 238 |
-
|
| 239 |
-
# #👉 this part is for csv subclass table generatived
|
| 240 |
-
# with gr.TabItem("🏅 Subclass Generatived", elem_id="od-benchmark-tab-table", id=3):
|
| 241 |
-
# dataframe_sub_gen = gr.components.Dataframe(
|
| 242 |
-
# elem_id="leaderboard-table",
|
| 243 |
-
# )
|
| 244 |
-
|
| 245 |
-
# #👉 this part is for csv subclass table perplexity
|
| 246 |
-
# with gr.TabItem("🏅 Subclass Perplexity", elem_id="od-benchmark-tab-table", id=4):
|
| 247 |
-
# dataframe_sub_per = gr.components.Dataframe(
|
| 248 |
-
# elem_id="leaderboard-table",
|
| 249 |
-
# )
|
| 250 |
# ----------------- modify text -----------------
|
| 251 |
|
| 252 |
with gr.TabItem("🏅 Generation", elem_id="od-benchmark-tab-table", id=6):
|
|
@@ -269,105 +197,6 @@ with gr.Blocks() as demo:
|
|
| 269 |
|
| 270 |
gr.Markdown(f"Last updated on **{_LAST_UPDATED}**", elem_classes="markdown-text")
|
| 271 |
|
| 272 |
-
# 👉 this part is for citation
|
| 273 |
-
# with gr.Row():
|
| 274 |
-
# with gr.Accordion("📙 Citation", open=False):
|
| 275 |
-
# gr.Textbox(
|
| 276 |
-
# value=_BIBTEX,
|
| 277 |
-
# lines=7,
|
| 278 |
-
# label="Copy the BibTeX snippet to cite this source",
|
| 279 |
-
# elem_id="citation-button",
|
| 280 |
-
# show_copy_button=True
|
| 281 |
-
# )
|
| 282 |
-
|
| 283 |
-
# this is result based on generative
|
| 284 |
-
# metric_choice.change(
|
| 285 |
-
# get_dataset_csv,
|
| 286 |
-
# inputs=[model_choice, metric_choice],
|
| 287 |
-
# outputs=dataframe,
|
| 288 |
-
# )
|
| 289 |
-
|
| 290 |
-
# model_choice.change(
|
| 291 |
-
# get_dataset_csv,
|
| 292 |
-
# inputs=[model_choice, metric_choice],
|
| 293 |
-
# outputs=dataframe,
|
| 294 |
-
# )
|
| 295 |
-
|
| 296 |
-
# demo.load(
|
| 297 |
-
# fn=get_dataset_csv,
|
| 298 |
-
# inputs=[model_choice, metric_choice],
|
| 299 |
-
# outputs=dataframe,
|
| 300 |
-
# )
|
| 301 |
-
|
| 302 |
-
# # this is result based on Perplexity
|
| 303 |
-
# metric_choice.change(
|
| 304 |
-
# get_dataset_csv_per,
|
| 305 |
-
# inputs=[model_choice, metric_choice],
|
| 306 |
-
# outputs=datafram_per,
|
| 307 |
-
# )
|
| 308 |
-
|
| 309 |
-
# model_choice.change(
|
| 310 |
-
# get_dataset_csv_per,
|
| 311 |
-
# inputs=[model_choice, metric_choice],
|
| 312 |
-
# outputs=datafram_per,
|
| 313 |
-
# )
|
| 314 |
-
|
| 315 |
-
# demo.load(
|
| 316 |
-
# fn=get_dataset_csv_per,
|
| 317 |
-
# inputs=[model_choice, metric_choice],
|
| 318 |
-
# outputs=datafram_per,
|
| 319 |
-
# )
|
| 320 |
-
|
| 321 |
-
# this is subclass result generatived
|
| 322 |
-
# metric_choice.change(
|
| 323 |
-
# get_dataset_csv_sub_gen,
|
| 324 |
-
# inputs=[model_choice, metric_choice, subclass_choice],
|
| 325 |
-
# outputs=dataframe_sub_gen,
|
| 326 |
-
# )
|
| 327 |
-
|
| 328 |
-
# model_choice.change(
|
| 329 |
-
# get_dataset_csv_sub_gen,
|
| 330 |
-
# inputs=[model_choice, metric_choice, subclass_choice],
|
| 331 |
-
# outputs=dataframe_sub_gen,
|
| 332 |
-
# )
|
| 333 |
-
|
| 334 |
-
# subclass_choice.change(
|
| 335 |
-
# get_dataset_csv_sub_gen,
|
| 336 |
-
# inputs=[model_choice, metric_choice, subclass_choice],
|
| 337 |
-
# outputs=dataframe_sub_gen,
|
| 338 |
-
# )
|
| 339 |
-
|
| 340 |
-
# demo.load(
|
| 341 |
-
# fn=get_dataset_csv_sub_gen,
|
| 342 |
-
# inputs=[model_choice, metric_choice, subclass_choice],
|
| 343 |
-
# outputs=dataframe_sub_gen,
|
| 344 |
-
# )
|
| 345 |
-
|
| 346 |
-
# # this is subclass result Perplexity
|
| 347 |
-
# # metric_choice.change(
|
| 348 |
-
# # get_dataset_csv_sub_per,
|
| 349 |
-
# # inputs=[model_choice, metric_choice, subclass_choice],
|
| 350 |
-
# # outputs=dataframe_sub_per,
|
| 351 |
-
# # )
|
| 352 |
-
|
| 353 |
-
# model_choice.change(
|
| 354 |
-
# get_dataset_csv_sub_per,
|
| 355 |
-
# inputs=[model_choice, metric_choice, subclass_choice],
|
| 356 |
-
# outputs=dataframe_sub_per,
|
| 357 |
-
# )
|
| 358 |
-
|
| 359 |
-
# subclass_choice.change(
|
| 360 |
-
# get_dataset_csv_sub_per,
|
| 361 |
-
# inputs=[model_choice, metric_choice, subclass_choice],
|
| 362 |
-
# outputs=dataframe_sub_per,
|
| 363 |
-
# )
|
| 364 |
-
|
| 365 |
-
# demo.load(
|
| 366 |
-
# fn=get_dataset_csv_sub_per,
|
| 367 |
-
# inputs=[model_choice, metric_choice, subclass_choice],
|
| 368 |
-
# outputs=dataframe_sub_per,
|
| 369 |
-
# )
|
| 370 |
-
|
| 371 |
# --------------------------- all --------------------------------
|
| 372 |
# this is all result Perplexity
|
| 373 |
|
|
@@ -382,18 +211,6 @@ with gr.Blocks() as demo:
|
|
| 382 |
inputs=[model_choice, main_choice],
|
| 383 |
outputs=dataframe_all_per,
|
| 384 |
)
|
| 385 |
-
|
| 386 |
-
# metric_choice.change(
|
| 387 |
-
# get_dataset_classfier_per,
|
| 388 |
-
# inputs=[model_choice, main_choice],
|
| 389 |
-
# outputs=dataframe_all_per,
|
| 390 |
-
# )
|
| 391 |
-
|
| 392 |
-
# subclass_choice.change(
|
| 393 |
-
# get_dataset_classfier_per,
|
| 394 |
-
# inputs=[model_choice, metric_choice, main_choice],
|
| 395 |
-
# outputs=dataframe_all_per,
|
| 396 |
-
# )
|
| 397 |
|
| 398 |
demo.load(
|
| 399 |
fn=get_dataset_classfier_per,
|
|
@@ -414,18 +231,6 @@ with gr.Blocks() as demo:
|
|
| 414 |
outputs=dataframe_all_gen,
|
| 415 |
)
|
| 416 |
|
| 417 |
-
# metric_choice.change(
|
| 418 |
-
# get_dataset_classfier_gen,
|
| 419 |
-
# inputs=[model_choice, metric_choice, main_choice],
|
| 420 |
-
# outputs=dataframe_all_gen,
|
| 421 |
-
# )
|
| 422 |
-
|
| 423 |
-
# subclass_choice.change(
|
| 424 |
-
# get_dataset_classfier_gen,
|
| 425 |
-
# inputs=[model_choice, metric_choice, main_choice],
|
| 426 |
-
# outputs=dataframe_all_gen,
|
| 427 |
-
# )
|
| 428 |
-
|
| 429 |
demo.load(
|
| 430 |
fn=get_dataset_classfier_gen,
|
| 431 |
inputs=[model_choice, main_choice],
|
|
|
|
| 61 |
):
|
| 62 |
df = ORIGINAL_DF[ORIGINAL_DF['Size'].isin(model_size)]
|
| 63 |
df = df.drop(columns="Size")
|
| 64 |
+
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 65 |
leaderboard_table = gr.components.Dataframe(
|
| 66 |
value=df,
|
| 67 |
interactive=False,
|
|
|
|
| 75 |
df = ORIGINAL_DF_PER[ORIGINAL_DF_PER['Size'].isin(model_size)]
|
| 76 |
df = df.drop(columns="Size")
|
| 77 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 78 |
leaderboard_table = gr.components.Dataframe(
|
| 79 |
value=df,
|
| 80 |
interactive=False,
|
|
|
|
| 94 |
subclass_choice_label = ["Model", subclass_choice+"_Accuracy", subclass_choice+"_Precision", subclass_choice+"_Recall"]
|
| 95 |
df = df[subclass_choice_label]
|
| 96 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 97 |
leaderboard_table = gr.components.Dataframe(
|
| 98 |
value=df,
|
| 99 |
interactive=False,
|
|
|
|
| 113 |
subclass_choice_label = ["Model", subclass_choice+"_Accuracy", subclass_choice+"_Precision", subclass_choice+"_Recall"]
|
| 114 |
df = df[subclass_choice_label]
|
| 115 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 116 |
leaderboard_table = gr.components.Dataframe(
|
| 117 |
value=df,
|
| 118 |
interactive=False,
|
|
|
|
| 165 |
info="Please choose the type to display.",
|
| 166 |
)
|
| 167 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 168 |
with gr.Column(scale=10):
|
| 169 |
model_choice = gr.CheckboxGroup(
|
| 170 |
choices=CLASSIFICATION["model_size"],
|
|
|
|
| 173 |
info="Please choose the model size to display.",
|
| 174 |
)
|
| 175 |
|
| 176 |
+
#👉 this part is for csv table generatived
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 177 |
with gr.Tabs(elem_classes="tab-buttons") as tabs:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 178 |
# ----------------- modify text -----------------
|
| 179 |
|
| 180 |
with gr.TabItem("🏅 Generation", elem_id="od-benchmark-tab-table", id=6):
|
|
|
|
| 197 |
|
| 198 |
gr.Markdown(f"Last updated on **{_LAST_UPDATED}**", elem_classes="markdown-text")
|
| 199 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 200 |
# --------------------------- all --------------------------------
|
| 201 |
# this is all result Perplexity
|
| 202 |
|
|
|
|
| 211 |
inputs=[model_choice, main_choice],
|
| 212 |
outputs=dataframe_all_per,
|
| 213 |
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 214 |
|
| 215 |
demo.load(
|
| 216 |
fn=get_dataset_classfier_per,
|
|
|
|
| 231 |
outputs=dataframe_all_gen,
|
| 232 |
)
|
| 233 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 234 |
demo.load(
|
| 235 |
fn=get_dataset_classfier_gen,
|
| 236 |
inputs=[model_choice, main_choice],
|