nan commited on
Commit
8ec7973
β€’
1 Parent(s): 5808d8f

feat: update the intro text

Browse files
Files changed (2) hide show
  1. app.py +14 -13
  2. src/about.py +2 -2
app.py CHANGED
@@ -21,8 +21,6 @@ from utils import update_table, update_metric
21
  from src.benchmarks import DOMAIN_COLS_QA, LANG_COLS_QA, metric_list
22
 
23
 
24
- from functools import partial
25
-
26
  def restart_space():
27
  API.restart_space(repo_id=REPO_ID)
28
 
@@ -71,15 +69,25 @@ with demo:
71
  gr.Markdown(INTRODUCTION_TEXT, elem_classes="markdown-text")
72
 
73
  with gr.Tabs(elem_classes="tab-buttons") as tabs:
74
- with gr.TabItem("QA", elem_id="llm-benchmark-tab-table", id=0):
75
  with gr.Row():
76
  with gr.Column():
 
77
  with gr.Row():
78
  search_bar = gr.Textbox(
79
  placeholder=" πŸ” Search for your model (separate multiple queries with `;`) and press ENTER...",
80
  show_label=False,
81
  elem_id="search-bar",
82
  )
 
 
 
 
 
 
 
 
 
83
  # select domain
84
  with gr.Row():
85
  selected_domains = gr.CheckboxGroup(
@@ -98,7 +106,7 @@ with demo:
98
  elem_id="language-column-select",
99
  interactive=True
100
  )
101
- # select reranking models
102
  reranking_models = list(frozenset([eval_result.reranking_model for eval_result in raw_data_qa]))
103
  with gr.Row():
104
  selected_rerankings = gr.CheckboxGroup(
@@ -108,16 +116,7 @@ with demo:
108
  elem_id="reranking-select",
109
  interactive=True
110
  )
111
- with gr.Column(min_width=320):
112
- selected_metric = gr.Dropdown(
113
- choices=metric_list,
114
- value=metric_list[1],
115
- label="Select the metric",
116
- interactive=True,
117
- elem_id="metric-select",
118
- )
119
 
120
- # reload the leaderboard_df and raw_data when selected_metric is changed
121
  leaderboard_table = gr.components.Dataframe(
122
  value=leaderboard_df,
123
  # headers=shown_columns,
@@ -179,6 +178,8 @@ with demo:
179
  queue=True
180
  )
181
 
 
 
182
  with gr.TabItem("πŸ“ About", elem_id="llm-benchmark-tab-table", id=2):
183
  gr.Markdown(LLM_BENCHMARKS_TEXT, elem_classes="markdown-text")
184
 
 
21
  from src.benchmarks import DOMAIN_COLS_QA, LANG_COLS_QA, metric_list
22
 
23
 
 
 
24
  def restart_space():
25
  API.restart_space(repo_id=REPO_ID)
26
 
 
69
  gr.Markdown(INTRODUCTION_TEXT, elem_classes="markdown-text")
70
 
71
  with gr.Tabs(elem_classes="tab-buttons") as tabs:
72
+ with gr.TabItem("QA", elem_id="qa-benchmark-tab-table", id=0):
73
  with gr.Row():
74
  with gr.Column():
75
+ # search bar for model name
76
  with gr.Row():
77
  search_bar = gr.Textbox(
78
  placeholder=" πŸ” Search for your model (separate multiple queries with `;`) and press ENTER...",
79
  show_label=False,
80
  elem_id="search-bar",
81
  )
82
+ # select the metric
83
+ selected_metric = gr.Dropdown(
84
+ choices=metric_list,
85
+ value=metric_list[1],
86
+ label="Select the metric",
87
+ interactive=True,
88
+ elem_id="metric-select",
89
+ )
90
+ with gr.Column(min_width=320):
91
  # select domain
92
  with gr.Row():
93
  selected_domains = gr.CheckboxGroup(
 
106
  elem_id="language-column-select",
107
  interactive=True
108
  )
109
+ # select reranking model
110
  reranking_models = list(frozenset([eval_result.reranking_model for eval_result in raw_data_qa]))
111
  with gr.Row():
112
  selected_rerankings = gr.CheckboxGroup(
 
116
  elem_id="reranking-select",
117
  interactive=True
118
  )
 
 
 
 
 
 
 
 
119
 
 
120
  leaderboard_table = gr.components.Dataframe(
121
  value=leaderboard_df,
122
  # headers=shown_columns,
 
178
  queue=True
179
  )
180
 
181
+ # with gr.TabItem("Long Doc", elem_id="long-doc-benchmark-tab-table", id=1):
182
+
183
  with gr.TabItem("πŸ“ About", elem_id="llm-benchmark-tab-table", id=2):
184
  gr.Markdown(LLM_BENCHMARKS_TEXT, elem_classes="markdown-text")
185
 
src/about.py CHANGED
@@ -38,11 +38,11 @@ NUM_FEWSHOT = 0 # Change with your few shot
38
 
39
 
40
  # Your leaderboard name
41
- TITLE = """<h1 align="center" id="space-title">Demo leaderboard</h1>"""
42
 
43
  # What does your leaderboard evaluate?
44
  INTRODUCTION_TEXT = """
45
- Intro text
46
  """
47
 
48
  # Which evaluations are you running? how can people reproduce what you have?
 
38
 
39
 
40
  # Your leaderboard name
41
+ TITLE = """<h1 align="center" id="space-title">AIR-Bench</h1>"""
42
 
43
  # What does your leaderboard evaluate?
44
  INTRODUCTION_TEXT = """
45
+ AIR-Bench: Automated Heterogeneous Information Retrieval Benchmark
46
  """
47
 
48
  # Which evaluations are you running? how can people reproduce what you have?