Spaces:
Running
on
CPU Upgrade
Running
on
CPU Upgrade
refactor: restructure the files
Browse files- app.py +1 -2
- src/columns.py +11 -12
- src/envs.py +0 -9
- src/loaders.py +1 -3
- src/models.py +3 -9
- src/utils.py +3 -8
- tests/test_utils.py +2 -9
app.py
CHANGED
@@ -22,8 +22,6 @@ from src.css_html_js import custom_css
|
|
22 |
from src.envs import (
|
23 |
API,
|
24 |
BENCHMARK_VERSION_LIST,
|
25 |
-
COL_NAME_RERANKING_MODEL,
|
26 |
-
COL_NAME_RETRIEVAL_MODEL,
|
27 |
DEFAULT_METRIC_LONG_DOC,
|
28 |
DEFAULT_METRIC_QA,
|
29 |
EVAL_RESULTS_PATH,
|
@@ -33,6 +31,7 @@ from src.envs import (
|
|
33 |
RESULTS_REPO,
|
34 |
TOKEN,
|
35 |
)
|
|
|
36 |
from src.loaders import load_eval_results
|
37 |
from src.models import TaskType, model_hyperlink
|
38 |
from src.utils import remove_html, reset_rank, set_listeners, submit_results, update_metric, upload_file
|
|
|
22 |
from src.envs import (
|
23 |
API,
|
24 |
BENCHMARK_VERSION_LIST,
|
|
|
|
|
25 |
DEFAULT_METRIC_LONG_DOC,
|
26 |
DEFAULT_METRIC_QA,
|
27 |
EVAL_RESULTS_PATH,
|
|
|
31 |
RESULTS_REPO,
|
32 |
TOKEN,
|
33 |
)
|
34 |
+
from src.columns import COL_NAME_RETRIEVAL_MODEL, COL_NAME_RERANKING_MODEL
|
35 |
from src.loaders import load_eval_results
|
36 |
from src.models import TaskType, model_hyperlink
|
37 |
from src.utils import remove_html, reset_rank, set_listeners, submit_results, update_metric, upload_file
|
src/columns.py
CHANGED
@@ -1,17 +1,5 @@
|
|
1 |
from dataclasses import dataclass, make_dataclass
|
2 |
|
3 |
-
from src.envs import (
|
4 |
-
COL_NAME_AVG,
|
5 |
-
COL_NAME_IS_ANONYMOUS,
|
6 |
-
COL_NAME_RANK,
|
7 |
-
COL_NAME_RERANKING_MODEL,
|
8 |
-
COL_NAME_RERANKING_MODEL_LINK,
|
9 |
-
COL_NAME_RETRIEVAL_MODEL,
|
10 |
-
COL_NAME_RETRIEVAL_MODEL_LINK,
|
11 |
-
COL_NAME_REVISION,
|
12 |
-
COL_NAME_TIMESTAMP,
|
13 |
-
)
|
14 |
-
|
15 |
|
16 |
def fields(raw_class):
|
17 |
return [v for k, v in raw_class.__dict__.items() if k[:2] != "__" and k[-2:] != "__"]
|
@@ -96,3 +84,14 @@ def get_default_col_names_and_types(benchmarks):
|
|
96 |
def get_fixed_col_names_and_types():
|
97 |
fixed_cols = get_default_auto_eval_column_dict()[:-3]
|
98 |
return [c.name for _, _, c in fixed_cols], [c.type for _, _, c in fixed_cols]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
from dataclasses import dataclass, make_dataclass
|
2 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
|
4 |
def fields(raw_class):
|
5 |
return [v for k, v in raw_class.__dict__.items() if k[:2] != "__" and k[-2:] != "__"]
|
|
|
84 |
def get_fixed_col_names_and_types():
|
85 |
fixed_cols = get_default_auto_eval_column_dict()[:-3]
|
86 |
return [c.name for _, _, c in fixed_cols], [c.type for _, _, c in fixed_cols]
|
87 |
+
|
88 |
+
|
89 |
+
COL_NAME_AVG = "Average ⬆️"
|
90 |
+
COL_NAME_RETRIEVAL_MODEL = "Retrieval Method"
|
91 |
+
COL_NAME_RERANKING_MODEL = "Reranking Model"
|
92 |
+
COL_NAME_RETRIEVAL_MODEL_LINK = "Retrieval Model LINK"
|
93 |
+
COL_NAME_RERANKING_MODEL_LINK = "Reranking Model LINK"
|
94 |
+
COL_NAME_RANK = "Rank 🏆"
|
95 |
+
COL_NAME_REVISION = "Revision"
|
96 |
+
COL_NAME_TIMESTAMP = "Submission Date"
|
97 |
+
COL_NAME_IS_ANONYMOUS = "Anonymous Submission"
|
src/envs.py
CHANGED
@@ -63,12 +63,3 @@ METRIC_LIST = [
|
|
63 |
"mrr_at_100",
|
64 |
"mrr_at_1000",
|
65 |
]
|
66 |
-
COL_NAME_AVG = "Average ⬆️"
|
67 |
-
COL_NAME_RETRIEVAL_MODEL = "Retrieval Method"
|
68 |
-
COL_NAME_RERANKING_MODEL = "Reranking Model"
|
69 |
-
COL_NAME_RETRIEVAL_MODEL_LINK = "Retrieval Model LINK"
|
70 |
-
COL_NAME_RERANKING_MODEL_LINK = "Reranking Model LINK"
|
71 |
-
COL_NAME_RANK = "Rank 🏆"
|
72 |
-
COL_NAME_REVISION = "Revision"
|
73 |
-
COL_NAME_TIMESTAMP = "Submission Date"
|
74 |
-
COL_NAME_IS_ANONYMOUS = "Anonymous Submission"
|
|
|
63 |
"mrr_at_100",
|
64 |
"mrr_at_1000",
|
65 |
]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
src/loaders.py
CHANGED
@@ -5,12 +5,10 @@ import pandas as pd
|
|
5 |
|
6 |
from src.envs import (
|
7 |
BENCHMARK_VERSION_LIST,
|
8 |
-
COL_NAME_IS_ANONYMOUS,
|
9 |
-
COL_NAME_REVISION,
|
10 |
-
COL_NAME_TIMESTAMP,
|
11 |
DEFAULT_METRIC_LONG_DOC,
|
12 |
DEFAULT_METRIC_QA,
|
13 |
)
|
|
|
14 |
from src.models import FullEvalResult, LeaderboardDataStore, TaskType
|
15 |
from src.utils import get_default_cols, get_leaderboard_df
|
16 |
|
|
|
5 |
|
6 |
from src.envs import (
|
7 |
BENCHMARK_VERSION_LIST,
|
|
|
|
|
|
|
8 |
DEFAULT_METRIC_LONG_DOC,
|
9 |
DEFAULT_METRIC_QA,
|
10 |
)
|
11 |
+
from src.columns import COL_NAME_REVISION, COL_NAME_TIMESTAMP, COL_NAME_IS_ANONYMOUS
|
12 |
from src.models import FullEvalResult, LeaderboardDataStore, TaskType
|
13 |
from src.utils import get_default_cols, get_leaderboard_df
|
14 |
|
src/models.py
CHANGED
@@ -7,15 +7,9 @@ from typing import List, Optional
|
|
7 |
|
8 |
import pandas as pd
|
9 |
|
10 |
-
from src.
|
11 |
-
COL_NAME_IS_ANONYMOUS
|
12 |
-
|
13 |
-
COL_NAME_RERANKING_MODEL_LINK,
|
14 |
-
COL_NAME_RETRIEVAL_MODEL,
|
15 |
-
COL_NAME_RETRIEVAL_MODEL_LINK,
|
16 |
-
COL_NAME_REVISION,
|
17 |
-
COL_NAME_TIMESTAMP,
|
18 |
-
)
|
19 |
|
20 |
def get_safe_name(name: str):
|
21 |
"""Get RFC 1123 compatible safe name"""
|
|
|
7 |
|
8 |
import pandas as pd
|
9 |
|
10 |
+
from src.columns import COL_NAME_RETRIEVAL_MODEL, COL_NAME_RERANKING_MODEL, COL_NAME_RETRIEVAL_MODEL_LINK, \
|
11 |
+
COL_NAME_RERANKING_MODEL_LINK, COL_NAME_REVISION, COL_NAME_TIMESTAMP, COL_NAME_IS_ANONYMOUS
|
12 |
+
|
|
|
|
|
|
|
|
|
|
|
|
|
13 |
|
14 |
def get_safe_name(name: str):
|
15 |
"""Get RFC 1123 compatible safe name"""
|
src/utils.py
CHANGED
@@ -8,16 +8,11 @@ import pandas as pd
|
|
8 |
|
9 |
from src.models import TaskType
|
10 |
from src.benchmarks import LongDocBenchmarks, QABenchmarks
|
11 |
-
from src.columns import get_default_col_names_and_types, get_fixed_col_names_and_types
|
|
|
|
|
12 |
from src.envs import (
|
13 |
API,
|
14 |
-
COL_NAME_AVG,
|
15 |
-
COL_NAME_IS_ANONYMOUS,
|
16 |
-
COL_NAME_RANK,
|
17 |
-
COL_NAME_RERANKING_MODEL,
|
18 |
-
COL_NAME_RETRIEVAL_MODEL,
|
19 |
-
COL_NAME_REVISION,
|
20 |
-
COL_NAME_TIMESTAMP,
|
21 |
LATEST_BENCHMARK_VERSION,
|
22 |
SEARCH_RESULTS_REPO,
|
23 |
)
|
|
|
8 |
|
9 |
from src.models import TaskType
|
10 |
from src.benchmarks import LongDocBenchmarks, QABenchmarks
|
11 |
+
from src.columns import get_default_col_names_and_types, get_fixed_col_names_and_types, COL_NAME_AVG, \
|
12 |
+
COL_NAME_RETRIEVAL_MODEL, COL_NAME_RERANKING_MODEL, COL_NAME_RANK, COL_NAME_REVISION, COL_NAME_TIMESTAMP, \
|
13 |
+
COL_NAME_IS_ANONYMOUS
|
14 |
from src.envs import (
|
15 |
API,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
16 |
LATEST_BENCHMARK_VERSION,
|
17 |
SEARCH_RESULTS_REPO,
|
18 |
)
|
tests/test_utils.py
CHANGED
@@ -2,15 +2,8 @@ import pandas as pd
|
|
2 |
import pytest
|
3 |
|
4 |
from app import update_table
|
5 |
-
from src.
|
6 |
-
|
7 |
-
COL_NAME_IS_ANONYMOUS,
|
8 |
-
COL_NAME_RANK,
|
9 |
-
COL_NAME_RERANKING_MODEL,
|
10 |
-
COL_NAME_RETRIEVAL_MODEL,
|
11 |
-
COL_NAME_REVISION,
|
12 |
-
COL_NAME_TIMESTAMP,
|
13 |
-
)
|
14 |
from src.utils import (
|
15 |
filter_models,
|
16 |
filter_queries,
|
|
|
2 |
import pytest
|
3 |
|
4 |
from app import update_table
|
5 |
+
from src.columns import COL_NAME_AVG, COL_NAME_RETRIEVAL_MODEL, COL_NAME_RERANKING_MODEL, COL_NAME_RANK, \
|
6 |
+
COL_NAME_REVISION, COL_NAME_TIMESTAMP, COL_NAME_IS_ANONYMOUS
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
7 |
from src.utils import (
|
8 |
filter_models,
|
9 |
filter_queries,
|