Spaces:
AIR-Bench
/
Running on CPU Upgrade

nan commited on
Commit
e2d3123
1 Parent(s): 2508d96

refactor: move the column names to a seperated file

Browse files
app.py CHANGED
@@ -18,13 +18,8 @@ from src.benchmarks import (
18
  DEFAULT_METRIC_LONG_DOC
19
  )
20
  from src.display.css_html_js import custom_css
21
- from src.display.utils import (
22
- COL_NAME_IS_ANONYMOUS,
23
- COL_NAME_REVISION,
24
- COL_NAME_TIMESTAMP,
25
- COL_NAME_RERANKING_MODEL,
26
- COL_NAME_RETRIEVAL_MODEL
27
- )
28
  from src.envs import (
29
  API,
30
  EVAL_RESULTS_PATH,
 
18
  DEFAULT_METRIC_LONG_DOC
19
  )
20
  from src.display.css_html_js import custom_css
21
+ from src.display.column_names import COL_NAME_RETRIEVAL_MODEL, COL_NAME_RERANKING_MODEL, COL_NAME_REVISION, \
22
+ COL_NAME_TIMESTAMP, COL_NAME_IS_ANONYMOUS
 
 
 
 
 
23
  from src.envs import (
24
  API,
25
  EVAL_RESULTS_PATH,
src/display/column_names.py ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ COL_NAME_AVG = "Average ⬆️"
2
+ COL_NAME_RETRIEVAL_MODEL = "Retrieval Method"
3
+ COL_NAME_RERANKING_MODEL = "Reranking Model"
4
+ COL_NAME_RETRIEVAL_MODEL_LINK = "Retrieval Model LINK"
5
+ COL_NAME_RERANKING_MODEL_LINK = "Reranking Model LINK"
6
+ COL_NAME_RANK = "Rank 🏆"
7
+ COL_NAME_REVISION = "Revision"
8
+ COL_NAME_TIMESTAMP = "Submission Date"
9
+ COL_NAME_IS_ANONYMOUS = "Anonymous Submission"
src/display/utils.py CHANGED
@@ -1,6 +1,9 @@
1
  from dataclasses import dataclass, make_dataclass
2
 
3
  from src.benchmarks import BenchmarksQA, BenchmarksLongDoc
 
 
 
4
 
5
 
6
  def fields(raw_class):
@@ -19,17 +22,6 @@ class ColumnContent:
19
  never_hidden: bool = False
20
 
21
 
22
- COL_NAME_AVG = "Average ⬆️"
23
- COL_NAME_RETRIEVAL_MODEL = "Retrieval Method"
24
- COL_NAME_RERANKING_MODEL = "Reranking Model"
25
- COL_NAME_RETRIEVAL_MODEL_LINK = "Retrieval Model LINK"
26
- COL_NAME_RERANKING_MODEL_LINK = "Reranking Model LINK"
27
- COL_NAME_RANK = "Rank 🏆"
28
- COL_NAME_REVISION = "Revision"
29
- COL_NAME_TIMESTAMP = "Submission Date"
30
- COL_NAME_IS_ANONYMOUS = "Anonymous Submission"
31
-
32
-
33
  def get_default_auto_eval_column_dict():
34
  auto_eval_column_dict = []
35
  # Init
 
1
  from dataclasses import dataclass, make_dataclass
2
 
3
  from src.benchmarks import BenchmarksQA, BenchmarksLongDoc
4
+ from src.display.column_names import COL_NAME_AVG, COL_NAME_RETRIEVAL_MODEL, COL_NAME_RERANKING_MODEL, \
5
+ COL_NAME_RETRIEVAL_MODEL_LINK, COL_NAME_RERANKING_MODEL_LINK, COL_NAME_RANK, COL_NAME_REVISION, COL_NAME_TIMESTAMP, \
6
+ COL_NAME_IS_ANONYMOUS
7
 
8
 
9
  def fields(raw_class):
 
22
  never_hidden: bool = False
23
 
24
 
 
 
 
 
 
 
 
 
 
 
 
25
  def get_default_auto_eval_column_dict():
26
  auto_eval_column_dict = []
27
  # Init
src/read_evals.py CHANGED
@@ -8,20 +8,14 @@ import pandas as pd
8
 
9
  from src.benchmarks import get_safe_name
10
  from src.display.utils import (
11
- COL_NAME_RERANKING_MODEL,
12
- COL_NAME_RETRIEVAL_MODEL,
13
- COL_NAME_RERANKING_MODEL_LINK,
14
- COL_NAME_RETRIEVAL_MODEL_LINK,
15
- COL_NAME_REVISION,
16
- COL_NAME_TIMESTAMP,
17
- COL_NAME_IS_ANONYMOUS,
18
  COLS_QA,
19
  QA_BENCHMARK_COLS,
20
  COLS_LONG_DOC,
21
- LONG_DOC_BENCHMARK_COLS,
22
- COL_NAME_AVG,
23
- COL_NAME_RANK
24
  )
 
 
 
25
 
26
  from src.display.formatting import make_clickable_model
27
 
 
8
 
9
  from src.benchmarks import get_safe_name
10
  from src.display.utils import (
 
 
 
 
 
 
 
11
  COLS_QA,
12
  QA_BENCHMARK_COLS,
13
  COLS_LONG_DOC,
14
+ LONG_DOC_BENCHMARK_COLS
 
 
15
  )
16
+ from src.display.column_names import COL_NAME_AVG, COL_NAME_RETRIEVAL_MODEL, COL_NAME_RERANKING_MODEL, \
17
+ COL_NAME_RETRIEVAL_MODEL_LINK, COL_NAME_RERANKING_MODEL_LINK, COL_NAME_RANK, COL_NAME_REVISION, COL_NAME_TIMESTAMP, \
18
+ COL_NAME_IS_ANONYMOUS
19
 
20
  from src.display.formatting import make_clickable_model
21
 
src/utils.py CHANGED
@@ -8,8 +8,9 @@ import pandas as pd
8
 
9
  from src.benchmarks import BENCHMARK_COLS_QA, BENCHMARK_COLS_LONG_DOC, BenchmarksQA, BenchmarksLongDoc
10
  from src.display.formatting import styled_message, styled_error
11
- from src.display.utils import COLS_QA, TYPES_QA, COLS_LONG_DOC, TYPES_LONG_DOC, COL_NAME_RANK, COL_NAME_AVG, \
12
- COL_NAME_RERANKING_MODEL, COL_NAME_RETRIEVAL_MODEL, COL_NAME_IS_ANONYMOUS, COL_NAME_TIMESTAMP, COL_NAME_REVISION, get_default_auto_eval_column_dict
 
13
  from src.envs import API, SEARCH_RESULTS_REPO, LATEST_BENCHMARK_VERSION
14
  from src.read_evals import FullEvalResult, get_leaderboard_df, calculate_mean
15
 
 
8
 
9
  from src.benchmarks import BENCHMARK_COLS_QA, BENCHMARK_COLS_LONG_DOC, BenchmarksQA, BenchmarksLongDoc
10
  from src.display.formatting import styled_message, styled_error
11
+ from src.display.utils import COLS_QA, TYPES_QA, COLS_LONG_DOC, TYPES_LONG_DOC, get_default_auto_eval_column_dict
12
+ from src.display.column_names import COL_NAME_AVG, COL_NAME_RETRIEVAL_MODEL, COL_NAME_RERANKING_MODEL, COL_NAME_RANK, \
13
+ COL_NAME_REVISION, COL_NAME_TIMESTAMP, COL_NAME_IS_ANONYMOUS
14
  from src.envs import API, SEARCH_RESULTS_REPO, LATEST_BENCHMARK_VERSION
15
  from src.read_evals import FullEvalResult, get_leaderboard_df, calculate_mean
16
 
tests/test_utils.py CHANGED
@@ -2,7 +2,8 @@ import pandas as pd
2
  import pytest
3
 
4
  from src.utils import filter_models, search_table, filter_queries, select_columns, update_table_long_doc, get_iso_format_timestamp, get_default_cols, update_table
5
- from src.display.utils import COL_NAME_IS_ANONYMOUS, COL_NAME_REVISION, COL_NAME_TIMESTAMP, COL_NAME_RERANKING_MODEL, COL_NAME_RETRIEVAL_MODEL, COL_NAME_RANK, COL_NAME_AVG
 
6
 
7
 
8
  @pytest.fixture
 
2
  import pytest
3
 
4
  from src.utils import filter_models, search_table, filter_queries, select_columns, update_table_long_doc, get_iso_format_timestamp, get_default_cols, update_table
5
+ from src.display.column_names 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
 
8
 
9
  @pytest.fixture