Updates

app.py

@@ -156,9 +156,10 @@ def perform_deduplication(
 with gr.Blocks(css="#status_output { height: 50px; overflow: auto; }") as demo:
     gr.Markdown("# Semantic Deduplication")
     gr.Markdown("""
-    This demo showcases a semantic deduplication process where we identify duplicate texts within a single dataset or across two datasets.
-    The deduplication is based on cosine similarity between the embeddings of the texts.
-    You can adjust the similarity threshold to control the strictness of the deduplication.
+    This demo showcases semantic deduplication using Model2Vec for HuggingFace datasets.
+    It can be used to identify duplicate texts within a single dataset or across two datasets.
+    You can adjust the similarity threshold to control the strictness of the deduplication.\n
+    NOTE: this demo runs on a free CPU backend, so it may be slow for large datasets. For faster results, please run the code locally.
     """)
 
     deduplication_type = gr.Radio(
@@ -181,8 +182,7 @@ with gr.Blocks(css="#status_output { height: 50px; overflow: auto; }") as demo:
             dataset2_text_column = gr.Textbox(value=default_text_column, label="Text Column Name")
 
     threshold = gr.Slider(0.0, 1.0, value=default_threshold, label="Similarity Threshold")
-    compute_button = gr.Button("Compute")
-    stop_button = gr.Button("Stop")
+    compute_button = gr.Button("Deduplicate")
     status_output = gr.Markdown(elem_id="status_output")
     result_output = gr.Markdown()
 
@@ -206,8 +206,6 @@ with gr.Blocks(css="#status_output { height: 50px; overflow: auto; }") as demo:
         outputs=[status_output, result_output],
     )
 
-    # Stop button functionality
-    stop_button.click(lambda: demo.stop(), None, None)
 
 demo.launch()