Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -103,9 +103,11 @@ model_id2 = st.text_input("Enter the second HuggingFace Model ID")
|
|
103 |
if st.button("Compare Models"):
|
104 |
with st.spinner('Comparing models and loading tokenizers...'):
|
105 |
if model_id1 and model_id2:
|
|
|
106 |
struct1 = get_model_structure(model_id1)
|
107 |
struct2 = get_model_structure(model_id2)
|
108 |
|
|
|
109 |
diff = compare_structures(struct1, struct2)
|
110 |
left_html, right_html, diff_found = display_diff(diff)
|
111 |
|
@@ -116,19 +118,26 @@ if st.button("Compare Models"):
|
|
116 |
col1, col2 = st.columns([1.5, 1.5]) # Adjust the ratio to make columns wider
|
117 |
|
118 |
with col1:
|
119 |
-
st.write("### Model 1")
|
120 |
st.markdown(left_html, unsafe_allow_html=True)
|
121 |
|
122 |
with col2:
|
123 |
-
st.write("### Model 2")
|
124 |
st.markdown(right_html, unsafe_allow_html=True)
|
125 |
|
126 |
# Tokenizer verification
|
127 |
try:
|
128 |
vocab_size1 = get_tokenizer_vocab_size(model_id1)
|
129 |
vocab_size2 = get_tokenizer_vocab_size(model_id2)
|
|
|
|
|
|
|
|
|
|
|
|
|
130 |
st.write(f"**{model_id1} Tokenizer Vocab Size**: {vocab_size1}")
|
131 |
st.write(f"**{model_id2} Tokenizer Vocab Size**: {vocab_size2}")
|
|
|
132 |
except Exception as e:
|
133 |
st.error(f"Error loading tokenizers: {e}")
|
134 |
else:
|
|
|
103 |
if st.button("Compare Models"):
|
104 |
with st.spinner('Comparing models and loading tokenizers...'):
|
105 |
if model_id1 and model_id2:
|
106 |
+
# Get model structures
|
107 |
struct1 = get_model_structure(model_id1)
|
108 |
struct2 = get_model_structure(model_id2)
|
109 |
|
110 |
+
# Compare model structures
|
111 |
diff = compare_structures(struct1, struct2)
|
112 |
left_html, right_html, diff_found = display_diff(diff)
|
113 |
|
|
|
118 |
col1, col2 = st.columns([1.5, 1.5]) # Adjust the ratio to make columns wider
|
119 |
|
120 |
with col1:
|
121 |
+
st.write(f"### Model 1: {model_id1}")
|
122 |
st.markdown(left_html, unsafe_allow_html=True)
|
123 |
|
124 |
with col2:
|
125 |
+
st.write(f"### Model 2: {model_id2}")
|
126 |
st.markdown(right_html, unsafe_allow_html=True)
|
127 |
|
128 |
# Tokenizer verification
|
129 |
try:
|
130 |
vocab_size1 = get_tokenizer_vocab_size(model_id1)
|
131 |
vocab_size2 = get_tokenizer_vocab_size(model_id2)
|
132 |
+
|
133 |
+
if vocab_size1 == vocab_size2:
|
134 |
+
st.success("The tokenizer vocab sizes are identical.")
|
135 |
+
else:
|
136 |
+
st.warning("The tokenizer vocab sizes are different.")
|
137 |
+
|
138 |
st.write(f"**{model_id1} Tokenizer Vocab Size**: {vocab_size1}")
|
139 |
st.write(f"**{model_id2} Tokenizer Vocab Size**: {vocab_size2}")
|
140 |
+
|
141 |
except Exception as e:
|
142 |
st.error(f"Error loading tokenizers: {e}")
|
143 |
else:
|