Update app.py

app.py

@@ -34,17 +34,19 @@ from huggingface_hub import login
 # Load model directly
 from transformers import AutoTokenizer, AutoModelForMaskedLM
 
-tokenizer = AutoTokenizer.from_pretrained("sentence-transformers/all-mpnet-base-v2")
+tokenizer = AutoTokenizer.from_pretrained("sentence-transformers/all-mpnet-base-v2", clean_up_tokenization_spaces=True)
 model = AutoModelForMaskedLM.from_pretrained("sentence-transformers/all-mpnet-base-v2")
 
-def parse_command(message):
-    candidate_labels = ["filter", "sort", "export", "log"]
-    result = classifier(message, candidate_labels)
-    return result['labels'][0] if result['scores'][0] > 0.5 else None
+# Define classifier for zero-shot classification
+classifier = pipeline("zero-shot-classification", model="facebook/bart-large-mnli")
+
+# Define nlp using a simple tokenizer
+from transformers import AutoTokenizer
+nlp = AutoTokenizer.from_pretrained("bert-base-uncased")
+
+
+
 
-# Usage
-command = parse_command("Filter apples, oranges in column Description")
-print(command)  # Output: 'filter'
 
 HUGGINGFACE_TOKEN = os.getenv("HUGGINGFACE_TOKEN")
 if not HUGGINGFACE_TOKEN:
@@ -1057,9 +1059,7 @@ def create_interface() -> gr.Blocks():
                 interactive=False
             )
 
-        # Progress Indicator
-        with gr.Row():
-            progress = gr.Progress()
+
 
         # Connect buttons to their respective functions
         start_button.click(
@@ -1070,7 +1070,7 @@ def create_interface() -> gr.Blocks():
                 scrape_interval,
                 content_type,
                 selector,
-                progress,
+
             ],
             outputs=status_output,
         )
@@ -1249,4 +1249,4 @@ if __name__ == "__main__":
     demo.launch()
 
     # Run automated tests
-    unittest.main(argv=[''], exit=False)
+    unittest.main(argv=[''], exit=False)