Spaces:
Runtime error
Runtime error
File size: 19,462 Bytes
9346f1c 4596a70 0227006 5f65cec 4596a70 9346f1c 4596a70 2a5f9fb 8c49cb6 2246286 8c49cb6 976f398 df66f6e 9d22eee df66f6e 976f398 df66f6e 2a5f9fb f2bc0a5 df66f6e f2bc0a5 8c49cb6 2a73469 10f9b3c 2a5f9fb 26286b2 d084b26 2a5f9fb 26286b2 d084b26 2a5f9fb 26286b2 a885f09 3dfaf22 ae85651 adb0416 2a73469 b1a1395 551debe ffefe11 adb0416 614ee1f 1f60a20 8c49cb6 72a0f0f e3a8804 ef5b51c 512b095 a2790cb 72a0f0f 512b095 aa7c3f4 adb0416 8c49cb6 ecef2dc 7644705 72a0f0f ef5b51c adb0416 ef5b51c adb0416 8c49cb6 e3a8804 8c49cb6 a2790cb 8c49cb6 2a5f9fb 8c49cb6 3ae1b8c ab6f548 3ae1b8c dc0413f 3ae1b8c dc0413f d2179b0 8c49cb6 d2179b0 7644705 5f65cec 22edb72 5f65cec 460d762 5f65cec f2bc0a5 5f65cec 8cb7546 5f65cec d16cee2 5f65cec 940ccca 10f9b3c 5f65cec |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 |
import gradio as gr
import json
import os
from datetime import datetime, timezone
import pandas as pd
from apscheduler.schedulers.background import BackgroundScheduler
from huggingface_hub import snapshot_download
from src.display.about import (
CITATION_BUTTON_LABEL,
CITATION_BUTTON_TEXT,
EVALUATION_QUEUE_TEXT,
INTRODUCTION_TEXT,
LLM_BENCHMARKS_TEXT,
FAQ_TEXT,
TITLE,
)
from src.display.css_html_js import custom_css
from src.display.utils import (
BENCHMARK_COLS,
COLS,
EVAL_COLS,
EVAL_TYPES,
NUMERIC_INTERVALS,
TYPES,
AutoEvalColumn,
ModelType,
fields,
WeightType,
Precision
)
from src.envs import API, EVAL_REQUESTS_PATH, EVAL_RESULTS_PATH, H4_TOKEN, IS_PUBLIC, QUEUE_REPO, REPO_ID, RESULTS_REPO
from src.populate import get_evaluation_queue_df, get_leaderboard_df
from src.submission.submit import add_new_eval
from src.submission.check_validity import already_submitted_models
from src.tools.collections import update_collections
from src.tools.plots import (
create_metric_plot_obj,
create_plot_df,
create_scores_df,
)
def restart_space():
API.restart_space(repo_id=REPO_ID, token=H4_TOKEN)
try:
print(EVAL_REQUESTS_PATH)
snapshot_download(
repo_id=QUEUE_REPO, local_dir=EVAL_REQUESTS_PATH, repo_type="dataset", tqdm_class=None, etag_timeout=30
)
except Exception:
restart_space()
try:
print(EVAL_RESULTS_PATH)
snapshot_download(
repo_id=RESULTS_REPO, local_dir=EVAL_RESULTS_PATH, repo_type="dataset", tqdm_class=None, etag_timeout=30
)
except Exception:
restart_space()
raw_data, original_df = get_leaderboard_df(EVAL_RESULTS_PATH, EVAL_REQUESTS_PATH, COLS, BENCHMARK_COLS)
update_collections(original_df.copy())
leaderboard_df = original_df.copy()
plot_df = create_plot_df(create_scores_df(raw_data))
(
finished_eval_queue_df,
running_eval_queue_df,
pending_eval_queue_df,
) = get_evaluation_queue_df(EVAL_REQUESTS_PATH, EVAL_COLS)
# Searching and filtering
def update_table(
hidden_df: pd.DataFrame,
columns: list,
type_query: list,
precision_query: str,
size_query: list,
show_deleted: bool,
query: str,
):
filtered_df = filter_models(hidden_df, type_query, size_query, precision_query, show_deleted)
filtered_df = filter_queries(query, filtered_df)
df = select_columns(filtered_df, columns)
return df
def search_table(df: pd.DataFrame, query: str) -> pd.DataFrame:
return df[(df[AutoEvalColumn.dummy.name].str.contains(query, case=False))]
def select_columns(df: pd.DataFrame, columns: list) -> pd.DataFrame:
always_here_cols = [
AutoEvalColumn.model_type_symbol.name,
AutoEvalColumn.model.name,
]
# We use COLS to maintain sorting
filtered_df = df[
always_here_cols + [c for c in COLS if c in df.columns and c in columns] + [AutoEvalColumn.dummy.name]
]
return filtered_df
def filter_queries(query: str, filtered_df: pd.DataFrame):
"""Added by Abishek"""
final_df = []
if query != "":
queries = [q.strip() for q in query.split(";")]
for _q in queries:
_q = _q.strip()
if _q != "":
temp_filtered_df = search_table(filtered_df, _q)
if len(temp_filtered_df) > 0:
final_df.append(temp_filtered_df)
if len(final_df) > 0:
filtered_df = pd.concat(final_df)
filtered_df = filtered_df.drop_duplicates(
subset=[AutoEvalColumn.model.name, AutoEvalColumn.precision.name, AutoEvalColumn.revision.name]
)
return filtered_df
def filter_models(
df: pd.DataFrame, type_query: list, size_query: list, precision_query: list, show_deleted: bool
) -> pd.DataFrame:
# Show all models
if show_deleted:
filtered_df = df
else: # Show only still on the hub models
filtered_df = df[df[AutoEvalColumn.still_on_hub.name] == True]
type_emoji = [t[0] for t in type_query]
filtered_df = filtered_df.loc[df[AutoEvalColumn.model_type_symbol.name].isin(type_emoji)]
filtered_df = filtered_df.loc[df[AutoEvalColumn.precision.name].isin(precision_query + ["None"])]
numeric_interval = pd.IntervalIndex(sorted([NUMERIC_INTERVALS[s] for s in size_query]))
params_column = pd.to_numeric(df[AutoEvalColumn.params.name], errors="coerce")
mask = params_column.apply(lambda x: any(numeric_interval.contains(x)))
filtered_df = filtered_df.loc[mask]
return filtered_df
# demo = gr.Blocks(css=custom_css)
# with demo:
# gr.HTML(TITLE)
# gr.Markdown(INTRODUCTION_TEXT, elem_classes="markdown-text")
#
# with gr.Tabs(elem_classes="tab-buttons") as tabs:
# with gr.TabItem("🏅 LLM Benchmark", elem_id="llm-benchmark-tab-table", id=0):
# with gr.Row():
# with gr.Column():
# with gr.Row():
# search_bar = gr.Textbox(
# placeholder=" 🔍 Search for your model and press ENTER...",
# show_label=False,
# elem_id="search-bar",
# )
# with gr.Row():
# shown_columns = gr.CheckboxGroup(
# choices=[
# c
# for c in COLS
# if c
# not in [
# AutoEvalColumn.dummy.name,
# AutoEvalColumn.model.name,
# AutoEvalColumn.model_type_symbol.name,
# AutoEvalColumn.still_on_hub.name,
# ]
# ],
# value=[
# c
# for c in COLS_LITE
# if c
# not in [
# AutoEvalColumn.dummy.name,
# AutoEvalColumn.model.name,
# AutoEvalColumn.model_type_symbol.name,
# AutoEvalColumn.still_on_hub.name,
# ]
# ],
# label="Select columns to show",
# elem_id="column-select",
# interactive=True,
# )
# with gr.Row():
# deleted_models_visibility = gr.Checkbox(
# value=True, label="Show gated/private/deleted models", interactive=True
# )
# with gr.Column(min_width=320):
# with gr.Box(elem_id="box-filter"):
# filter_columns_type = gr.CheckboxGroup(
# label="Model types",
# choices=[
# ModelType.PT.to_str(),
# ModelType.FT.to_str(),
# ModelType.IFT.to_str(),
# ModelType.RL.to_str(),
# ],
# value=[
# ModelType.PT.to_str(),
# ModelType.FT.to_str(),
# ModelType.IFT.to_str(),
# ModelType.RL.to_str(),
# ],
# interactive=True,
# elem_id="filter-columns-type",
# )
# filter_columns_precision = gr.CheckboxGroup(
# label="Precision",
# choices=["torch.float16", "torch.bfloat16", "torch.float32", "8bit", "4bit", "GPTQ"],
# value=["torch.float16", "torch.bfloat16", "torch.float32", "8bit", "4bit", "GPTQ"],
# interactive=True,
# elem_id="filter-columns-precision",
# )
# filter_columns_size = gr.CheckboxGroup(
# label="Model sizes",
# choices=list(NUMERIC_INTERVALS.keys()),
# value=list(NUMERIC_INTERVALS.keys()),
# interactive=True,
# elem_id="filter-columns-size",
# )
#
# leaderboard_table = gr.components.Dataframe(
# value=leaderboard_df[
# [AutoEvalColumn.model_type_symbol.name, AutoEvalColumn.model.name]
# + shown_columns.value
# + [AutoEvalColumn.dummy.name]
# ],
# headers=[
# AutoEvalColumn.model_type_symbol.name,
# AutoEvalColumn.model.name,
# ]
# + shown_columns.value
# + [AutoEvalColumn.dummy.name],
# datatype=TYPES,
# max_rows=None,
# elem_id="leaderboard-table",
# interactive=False,
# visible=True,
# )
#
# # Dummy leaderboard for handling the case when the user uses backspace key
# hidden_leaderboard_table_for_search = gr.components.Dataframe(
# value=original_df,
# headers=COLS,
# datatype=TYPES,
# max_rows=None,
# visible=False,
# )
# search_bar.submit(
# update_table,
# [
# hidden_leaderboard_table_for_search,
# leaderboard_table,
# shown_columns,
# filter_columns_type,
# filter_columns_precision,
# filter_columns_size,
# deleted_models_visibility,
# search_bar,
# ],
# leaderboard_table,
# )
# shown_columns.change(
# update_table,
# [
# hidden_leaderboard_table_for_search,
# leaderboard_table,
# shown_columns,
# filter_columns_type,
# filter_columns_precision,
# filter_columns_size,
# deleted_models_visibility,
# search_bar,
# ],
# leaderboard_table,
# queue=True,
# )
# filter_columns_type.change(
# update_table,
# [
# hidden_leaderboard_table_for_search,
# leaderboard_table,
# shown_columns,
# filter_columns_type,
# filter_columns_precision,
# filter_columns_size,
# deleted_models_visibility,
# search_bar,
# ],
# leaderboard_table,
# queue=True,
# )
# filter_columns_precision.change(
# update_table,
# [
# hidden_leaderboard_table_for_search,
# leaderboard_table,
# shown_columns,
# filter_columns_type,
# filter_columns_precision,
# filter_columns_size,
# deleted_models_visibility,
# search_bar,
# ],
# leaderboard_table,
# queue=True,
# )
# filter_columns_size.change(
# update_table,
# [
# hidden_leaderboard_table_for_search,
# leaderboard_table,
# shown_columns,
# filter_columns_type,
# filter_columns_precision,
# filter_columns_size,
# deleted_models_visibility,
# search_bar,
# ],
# leaderboard_table,
# queue=True,
# )
# deleted_models_visibility.change(
# update_table,
# [
# hidden_leaderboard_table_for_search,
# leaderboard_table,
# shown_columns,
# filter_columns_type,
# filter_columns_precision,
# filter_columns_size,
# deleted_models_visibility,
# search_bar,
# ],
# leaderboard_table,
# queue=True,
# )
# with gr.TabItem("📝 About", elem_id="llm-benchmark-tab-table", id=2):
# gr.Markdown(LLM_BENCHMARKS_TEXT, elem_classes="markdown-text")
#
# with gr.TabItem("🚀 Submit here! ", elem_id="llm-benchmark-tab-table", id=3):
# with gr.Column():
# with gr.Row():
# gr.Markdown(EVALUATION_QUEUE_TEXT, elem_classes="markdown-text")
#
# with gr.Column():
# with gr.Accordion(
# f"✅ Finished Evaluations ({len(finished_eval_queue_df)})",
# open=False,
# ):
# with gr.Row():
# finished_eval_table = gr.components.Dataframe(
# value=finished_eval_queue_df,
# headers=EVAL_COLS,
# datatype=EVAL_TYPES,
# max_rows=5,
# )
# with gr.Accordion(
# f"🔄 Running Evaluation Queue ({len(running_eval_queue_df)})",
# open=False,
# ):
# with gr.Row():
# running_eval_table = gr.components.Dataframe(
# value=running_eval_queue_df,
# headers=EVAL_COLS,
# datatype=EVAL_TYPES,
# max_rows=5,
# )
#
# with gr.Accordion(
# f"⏳ Pending Evaluation Queue ({len(pending_eval_queue_df)})",
# open=False,
# ):
# with gr.Row():
# pending_eval_table = gr.components.Dataframe(
# value=pending_eval_queue_df,
# headers=EVAL_COLS,
# datatype=EVAL_TYPES,
# max_rows=5,
# )
# with gr.Row():
# gr.Markdown("# ✉️✨ Submit your model here!", elem_classes="markdown-text")
#
# with gr.Row():
# with gr.Column():
# model_name_textbox = gr.Textbox(label="Model name")
# revision_name_textbox = gr.Textbox(label="revision", placeholder="main")
# private = gr.Checkbox(False, label="Private", visible=not IS_PUBLIC)
# model_type = gr.Dropdown(
# choices=[
# ModelType.PT.to_str(" : "),
# ModelType.FT.to_str(" : "),
# ModelType.IFT.to_str(" : "),
# ModelType.RL.to_str(" : "),
# ],
# label="Model type",
# multiselect=False,
# value=None,
# interactive=True,
# )
#
# with gr.Column():
# precision = gr.Dropdown(
# choices=[
# "float16",
# "bfloat16",
# "8bit (LLM.int8)",
# "4bit (QLoRA / FP4)",
# "GPTQ"
# ],
# label="Precision",
# multiselect=False,
# value="float16",
# interactive=True,
# )
# weight_type = gr.Dropdown(
# choices=["Original", "Delta", "Adapter"],
# label="Weights type",
# multiselect=False,
# value="Original",
# interactive=True,
# )
# base_model_name_textbox = gr.Textbox(label="Base model (for delta or adapter weights)")
#
# submit_button = gr.Button("Submit Eval")
# submission_result = gr.Markdown()
# submit_button.click(
# add_new_eval,
# [
# model_name_textbox,
# base_model_name_textbox,
# revision_name_textbox,
# precision,
# private,
# weight_type,
# model_type,
# ],
# submission_result,
# )
#
# with gr.Row():
# with gr.Accordion("📙 Citation", open=False):
# citation_button = gr.Textbox(
# value=CITATION_BUTTON_TEXT,
# label=CITATION_BUTTON_LABEL,
# elem_id="citation-button",
# ).style(show_copy_button=True)
#
# dummy = gr.Textbox(visible=False)
# demo.load(
# change_tab,
# dummy,
# tabs,
# _js=get_window_url_params,
# )
dummy1 = gr.Textbox(visible=False)
hidden_leaderboard_table_for_search = gr.components.Dataframe(
headers=COLS,
datatype=TYPES,
visible=False
)
def display(x, y):
return original_df
INTRODUCTION_TEXT = """
This is a copied space from Open Source LLM leaderboard. Instead of displaying
the results as table the space simply provides a gradio API interface to access
the full leaderboard data easily.
Example python on how to access the data:
```python
from gradio_client import Client
import json
client = Client("https://felixz-open-llm-leaderboard.hf.space/")
json_data = client.predict("","", api_name='/predict')
with open(json_data, 'r') as file:
file_data = file.read()
# Load the JSON data
data = json.loads(file_data)
# Get the headers and the data
headers = data['headers']
data = data['data']
```
"""
interface = gr.Interface(
fn=display,
inputs=[ gr.Markdown(INTRODUCTION_TEXT, elem_classes="markdown-text"),
dummy1
],
outputs=[hidden_leaderboard_table_for_search]
)
# Client auth error.. need to see how this works.
#scheduler = BackgroundScheduler()
#scheduler.add_job(restart_space, "interval", seconds=21600)
#scheduler.start()
interface.launch()
#demo.queue(concurrency_count=40).launch()
|