Hugging Face's logo

Spaces:
opll-org
/
Open-Persian-LLM-Leaderboard
Running

App Files Community
tohid.abedini commited on
Commit
e72de6c
·
1 Parent(s): 17c1bbe

[Add] 2nd phase of changes

Browse files
Files changed (3) hide show
  1. app.py +6 -9
  2. leaderboard_base.json +38 -1
  3. leaderboard_persian.json +38 -1
app.py CHANGED
@@ -108,30 +108,27 @@ with gr.Blocks(css=custom_css) as demo:
108
  # Part LLM Leaderboard
109
  """)
110
 
111
-
112
  with gr.Tab("🎖️ Persian Leaderboard"):
113
  gr.Markdown("""## Persian LLM Leaderboard
114
  Evaluating Persian Fine-Tuned models
115
  """)
116
  Leaderboard(
117
  value=persian_df,
118
- select_columns=["Arch", "Model", "L_rms", "I_rms",
119
- "F_nat", "DOCKQ", "CAPRI"],
120
  search_columns=["model_name_for_query"],
121
  hide_columns=["model_name_for_query",],
122
- filter_columns=["Arch"],
123
  )
124
  with gr.Tab("🥇 Base Leaderboard"):
125
- gr.Markdown("""## Base Leaderboard
126
- Evaluating Protein-Ligand prediction
127
  """)
128
  Leaderboard(
129
  value=base_df,
130
- select_columns=["Arch", "Model", "Mean lDDT-PLI", "Mean lDDT-LP",
131
- "Median RMSD", "% lDDT-PLI >= 0.5", "% pass PoseBusters"],
132
  search_columns=["model_name_for_query"],
133
  hide_columns=["model_name_for_query",],
134
- filter_columns=["Arch"],
135
  )
136
  with gr.TabItem("📝 About"):
137
  gr.Markdown(LLM_BENCHMARKS_ABOUT_TEXT)
 
108
  # Part LLM Leaderboard
109
  """)
110
 
 
111
  with gr.Tab("🎖️ Persian Leaderboard"):
112
  gr.Markdown("""## Persian LLM Leaderboard
113
  Evaluating Persian Fine-Tuned models
114
  """)
115
  Leaderboard(
116
  value=persian_df,
117
+ select_columns=["Model", "Precision", "#Params (B)", "Part Multiple Choice", "ARC Easy", "ARC Challenging", "MMLU Pro", "GSM8k Persian", "Multiple Choice Persian"],
 
118
  search_columns=["model_name_for_query"],
119
  hide_columns=["model_name_for_query",],
120
+ filter_columns=["Precision", "#Params (B)"],
121
  )
122
  with gr.Tab("🥇 Base Leaderboard"):
123
+ gr.Markdown("""## Base LLM Leaderboard
124
+ Evaluating Base Models
125
  """)
126
  Leaderboard(
127
  value=base_df,
128
+ select_columns=["Model", "Precision", "#Params (B)", "Part Multiple Choice", "ARC Easy", "ARC Challenging", "MMLU Pro", "GSM8k Persian", "Multiple Choice Persian"],
 
129
  search_columns=["model_name_for_query"],
130
  hide_columns=["model_name_for_query",],
131
+ filter_columns=["Precision", "#Params (B)"],
132
  )
133
  with gr.TabItem("📝 About"):
134
  gr.Markdown(LLM_BENCHMARKS_ABOUT_TEXT)
leaderboard_base.json CHANGED
@@ -1 +1,38 @@
1
- {"Arch":{"0":"GNN"},"Model":{"0":"example/example"},"Mean lDDT-PLI":{"0":0.2}, "Mean lDDT-LP":{"0":0.8}, "Median RMSD":{"0":5.78}, "% lDDT-PLI >= 0.5":{"0":20.78}, "% pass PoseBusters":{"0":58.47},"Runtime":{"0":"2 +-0.2"},"Hub License":{"0":"apache-2.0"},"#Params (B)":{"0":72.29},"Model sha":{"0":"fda5cf998a0f2d89b53b5fa490793e3e50bb8239"},"model_name_for_query":{"0":"example/example"}}
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "Arch": {
3
+ "0": "GNN"
4
+ },
5
+ "Model": {
6
+ "0": "example/example"
7
+ },
8
+ "Mean lDDT-PLI": {
9
+ "0": 0.2
10
+ },
11
+ "Mean lDDT-LP": {
12
+ "0": 0.8
13
+ },
14
+ "Median RMSD": {
15
+ "0": 5.78
16
+ },
17
+ "% lDDT-PLI >= 0.5": {
18
+ "0": 20.78
19
+ },
20
+ "% pass PoseBusters": {
21
+ "0": 58.47
22
+ },
23
+ "Runtime": {
24
+ "0": "2 +-0.2"
25
+ },
26
+ "Hub License": {
27
+ "0": "apache-2.0"
28
+ },
29
+ "#Params (B)": {
30
+ "0": 72.29
31
+ },
32
+ "Model sha": {
33
+ "0": "fda5cf998a0f2d89b53b5fa490793e3e50bb8239"
34
+ },
35
+ "model_name_for_query": {
36
+ "0": "example/example"
37
+ }
38
+ }
leaderboard_persian.json CHANGED
@@ -1 +1,38 @@
1
- {"Arch":{"0":"GNN"},"Model":{"0":"example/example"},"L_rms":{"0":81.22},"I_rms":{"0":79.78},"F_nat":{"0":91.15},"DOCKQ":{"0":77.95},"CAPRI":{"0":74.5},"Runtime":{"0":"2 +-0.2"},"Hub License":{"0":"apache-2.0"},"#Params (B)":{"0":72.29},"Model sha":{"0":"fda5cf998a0f2d89b53b5fa490793e3e50bb8239"},"model_name_for_query":{"0":"example/example"}}
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "Arch": {
3
+ "0": "GNN"
4
+ },
5
+ "Model": {
6
+ "0": "example/example"
7
+ },
8
+ "L_rms": {
9
+ "0": 81.22
10
+ },
11
+ "I_rms": {
12
+ "0": 79.78
13
+ },
14
+ "F_nat": {
15
+ "0": 91.15
16
+ },
17
+ "DOCKQ": {
18
+ "0": 77.95
19
+ },
20
+ "CAPRI": {
21
+ "0": 74.5
22
+ },
23
+ "Runtime": {
24
+ "0": "2 +-0.2"
25
+ },
26
+ "Hub License": {
27
+ "0": "apache-2.0"
28
+ },
29
+ "#Params (B)": {
30
+ "0": 72.29
31
+ },
32
+ "Model sha": {
33
+ "0": "fda5cf998a0f2d89b53b5fa490793e3e50bb8239"
34
+ },
35
+ "model_name_for_query": {
36
+ "0": "example/example"
37
+ }
38
+ }