Update app.py

app.py

@@ -32,13 +32,11 @@ class EmbeddingNN(nn.Module):
         x = self.fc(x)
         return x
 
+@st.cache
 def process_csv(csv_data):
     df = pd.read_csv(StringIO(csv_data))
-    
-    # Replace LabelEncoder with custom encoding using pandas factorize
     df['library_encoded'], library_classes = df['library_name'].factorize()
     df['description_encoded'], description_classes = df['description'].factorize()
-
     return df, library_classes, description_classes
 
 def train_and_export(df):
@@ -63,13 +61,9 @@ def infer_from_onnx(model_buffer, library_name, library_classes, description_cla
     byte_stream = BytesIO(model_buffer.getvalue())
     onnx_model = onnx.load(byte_stream)
     sess = onnxruntime.InferenceSession(byte_stream.getvalue())
-
-    # Replace transform with custom encoding
     encoded_library = torch.tensor([list(library_classes).index(library_name)], dtype=torch.long)
-    
     outputs = sess.run(None, {sess.get_inputs()[0].name: encoded_library.numpy()})
     predicted_description = description_classes[outputs[0].argmax()]
-
     return predicted_description
 
 # Streamlit UI
@@ -103,8 +97,11 @@ numpy,Library for numerical computations in Python
 """
 
 csv_data = st.text_area("Paste your CSV data here:", value=DEFAULT_CSV)
+
+# Process CSV and cache the results
+df, library_classes, description_classes = process_csv(csv_data)
+
 if st.button('Convert CSV to ONNX Neural Net'):
-    df, library_classes, description_classes = process_csv(csv_data)
     model_buffer = train_and_export(df)
     st.download_button(
         label="Download ONNX model",