Spaces:
Sleeping
Sleeping
felix
commited on
Commit
•
b9fbf95
1
Parent(s):
67e2a45
add
Browse files
app.py
CHANGED
@@ -44,6 +44,7 @@ imgs = glob.glob(os.path.join(data_path, '*.png'))
|
|
44 |
|
45 |
# Extracting images that start with "hf_llm_diagram"
|
46 |
hf_llm_diagrams = [img for img in imgs if 'hf_llm_diagram' in os.path.basename(img)]
|
|
|
47 |
|
48 |
# Getting the remaining images
|
49 |
remaining_imgs = [img for img in imgs if 'hf_llm_diagram' not in os.path.basename(img)]
|
@@ -105,6 +106,13 @@ cols[1].image(hf_llm_diagrams[2],caption="Commercially permissive license only",
|
|
105 |
|
106 |
print_model_list(hf_llm_diagrams[2],cols[1])
|
107 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
108 |
st.divider()
|
109 |
st.write("HuggingFace and Other Leaderboards: A Comparative Model Evaluation")
|
110 |
st.caption("Only models evaluated on both leaderboards are included.")
|
@@ -130,6 +138,7 @@ st.write(
|
|
130 |
<li><a href="https://www.mosaicml.com/llm-evaluation">MosaicML</a> Balanced set of static benchmarks</li>
|
131 |
<li><a href="https://opencompass.org.cn/leaderboard-llm">OpenCompass</a> Balanced set of static benchmarks</li>
|
132 |
<li><a href="https://llmbench.ai/data">AgentBench</a> Benchmark evaluating Agent abilities</li>
|
|
|
133 |
</ul>
|
134 |
""", unsafe_allow_html=True
|
135 |
)
|
|
|
44 |
|
45 |
# Extracting images that start with "hf_llm_diagram"
|
46 |
hf_llm_diagrams = [img for img in imgs if 'hf_llm_diagram' in os.path.basename(img)]
|
47 |
+
bigcode_diagrams = [img for img in imgs if 'bigcode' in os.path.basename(img)]
|
48 |
|
49 |
# Getting the remaining images
|
50 |
remaining_imgs = [img for img in imgs if 'hf_llm_diagram' not in os.path.basename(img)]
|
|
|
106 |
|
107 |
print_model_list(hf_llm_diagrams[2],cols[1])
|
108 |
|
109 |
+
st.divider()
|
110 |
+
|
111 |
+
st.write("Big Code Models Leaderboard")
|
112 |
+
st.image(bigcode_diagrams[0],use_column_width="auto")
|
113 |
+
|
114 |
+
print_model_list(bigcode_diagrams[0],st,True)
|
115 |
+
|
116 |
st.divider()
|
117 |
st.write("HuggingFace and Other Leaderboards: A Comparative Model Evaluation")
|
118 |
st.caption("Only models evaluated on both leaderboards are included.")
|
|
|
138 |
<li><a href="https://www.mosaicml.com/llm-evaluation">MosaicML</a> Balanced set of static benchmarks</li>
|
139 |
<li><a href="https://opencompass.org.cn/leaderboard-llm">OpenCompass</a> Balanced set of static benchmarks</li>
|
140 |
<li><a href="https://llmbench.ai/data">AgentBench</a> Benchmark evaluating Agent abilities</li>
|
141 |
+
<li><a href="https://huggingface.co/spaces/bigcode/bigcode-models-leaderboard">BigCode</a> Compare performance of base multilingual code generation models</li>
|
142 |
</ul>
|
143 |
""", unsafe_allow_html=True
|
144 |
)
|
data/20230820_0354/bigcode_leaderboard.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
[]
|
data/20230823_0545/bigcode_leaderboard.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
[]
|
data/20230824_0626/bigcode_leaderboard.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
[]
|
data/20230825_0828/bigcode_leaderboard.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
[]
|
fix_old_data.py
CHANGED
@@ -6,6 +6,8 @@ data_path = './data/'
|
|
6 |
no_generated_image_path = './no_generated_image.png' # Update this path to the location of no_generated_image.png
|
7 |
|
8 |
expected_files = [
|
|
|
|
|
9 |
'hf_llm_diagramv2.json',
|
10 |
'hf_llm_diagramv2.png',
|
11 |
'hf_llm_diagramv2_other_permissive.json',
|
|
|
6 |
no_generated_image_path = './no_generated_image.png' # Update this path to the location of no_generated_image.png
|
7 |
|
8 |
expected_files = [
|
9 |
+
'bigcode_leaderboard.json',
|
10 |
+
'bigcode_leaderboard.png',
|
11 |
'hf_llm_diagramv2.json',
|
12 |
'hf_llm_diagramv2.png',
|
13 |
'hf_llm_diagramv2_other_permissive.json',
|