Update app.py to allow api calls to the site

app.py

@@ -11,17 +11,12 @@ from torch.nn.functional import cosine_similarity
 from sentence_transformers import SentenceTransformer
 import gradio as gr
 
-# ---------------------
-# 1. Download NLTK Data
-# ---------------------
+# Download necessary NLTK data
 nltk.download('punkt')
-nltk.download('punkt_tab')
 nltk.download('stopwords')
 nltk.download('wordnet')
 
-# ---------------------
-# 2. Load Dataset
-# ---------------------
+# Load dataset
 dataset_path = "dataset.json"
 if not os.path.exists(dataset_path):
     raise FileNotFoundError(f"{dataset_path} not found. Please ensure the file is in the current directory.")
@@ -29,42 +24,28 @@ if not os.path.exists(dataset_path):
 with open(dataset_path, "r") as f:
     data = json.load(f)
 
-# Convert dataset into a dictionary: disease name -> details
 disease_data = {d["name"]: d for d in data["diseases"]}
-print("Sample Disease (Acne):")
-print(json.dumps(disease_data.get("Acne", {}), indent=2))
 
-# ---------------------
-# 3. Advanced Preprocessing Functions
-# ---------------------
+# Preprocessing
 lemmatizer = WordNetLemmatizer()
 stop_words = set(stopwords.words('english'))
 
-# Mapping for common medical synonyms to standardize terminology
 medical_synonyms = {
-    "itchy": "itch",
-    "pruritic": "itch",
-    "rash": "eruption",
-    "redness": "red",
-    "swollen": "swelling",
-    "bumps": "lesion"
+    "itchy": "itch", "pruritic": "itch", "rash": "eruption",
+    "redness": "red", "swollen": "swelling", "bumps": "lesion"
 }
 
+
 def advanced_preprocess_text(text):
-    # Lowercase and remove unwanted characters (retain only letters and spaces)
     text = text.lower()
     text = re.sub(r'[^a-zA-Z\s]', ' ', text)
     tokens = word_tokenize(text)
-    # Replace tokens using the medical synonym mapping
     tokens = [medical_synonyms.get(token, token) for token in tokens]
-    # Remove stopwords and lemmatize
     tokens = [lemmatizer.lemmatize(token) for token in tokens if token not in stop_words]
     return " ".join(tokens)
 
-# ---------------------
-# 4. Prepare Disease Descriptions & Embeddings Using BERT
-# ---------------------
-# For each disease, combine symptoms, causes, cure, and home remedies into a detailed description.
+
+# Prepare disease descriptions
 disease_names = []
 disease_descriptions = []
 
@@ -82,51 +63,33 @@ for disease_name, details in disease_data.items():
     processed_desc = advanced_preprocess_text(description)
     disease_descriptions.append(processed_desc)
 
-# Initialize SentenceTransformer (BERT-based model)
+# Load BERT model
 bert_model = SentenceTransformer('all-MiniLM-L6-v2')
-# Compute embeddings for each disease description (as torch tensors)
 disease_embeddings = bert_model.encode(disease_descriptions, convert_to_tensor=True)
-print("Computed disease embeddings for", len(disease_embeddings), "diseases.")
-
-# ---------------------
-# 5. Detailed User Input Collection
-# ---------------------
-def get_detailed_user_input(primary_symptom, location, associated_symptoms, duration, severity, additional_info):
-    # Combine all responses into one detailed description.
-    combined_input = " ".join([primary_symptom, location, associated_symptoms, duration, severity, additional_info])
-    return combined_input
-
-# ---------------------
-# 6. Chatbot Prediction Function Using BERT Similarity
-# ---------------------
-def chatbot_response(user_input):
+
+
+def gradio_chatbot(primary_symptom, location, associated_symptoms, duration, severity, additional_info):
+    user_input = " ".join([
+        primary_symptom, location, associated_symptoms, duration, severity, additional_info
+    ])
     processed_input = advanced_preprocess_text(user_input)
     user_embedding = bert_model.encode(processed_input, convert_to_tensor=True)
     similarities = cosine_similarity(user_embedding, disease_embeddings)
     best_match_idx = torch.argmax(similarities).item()
     predicted_disease = disease_names[best_match_idx]
-    return predicted_disease
+    details = disease_data.get(predicted_disease, {})
 
-# ---------------------
-# 7. Gradio Chatbot Function
-# ---------------------
-def gradio_chatbot(primary_symptom, location, associated_symptoms, duration, severity, additional_info):
-    detailed_input = get_detailed_user_input(primary_symptom, location, associated_symptoms, duration, severity, additional_info)
-    predicted = chatbot_response(detailed_input)
-    details = disease_data.get(predicted, {})
     causes = ", ".join(details.get("causes", ["Not available"]))
     cure = ", ".join(details.get("cure", ["Not available"]))
     home_remedies = ", ".join(details.get("home_remedies", ["Not available"]))
-    
-    response = (f"🩺 **Predicted Disease:** {predicted}\n\n"
+
+    response = (f"🩺 **Predicted Disease:** {predicted_disease}\n\n"
                 f"**Causes:** {causes}\n\n"
                 f"**Cure:** {cure}\n\n"
                 f"**Home Remedies:** {home_remedies}")
     return response
 
-# ---------------------
-# 8. Gradio Interface Setup
-# ---------------------
+
 iface = gr.Interface(
     fn=gradio_chatbot,
     inputs=[
@@ -142,7 +105,4 @@ iface = gr.Interface(
     description="Answer a few questions about your symptoms and get a predicted skin disease along with recommended solutions."
 )
 
-# ---------------------
-# 9. Launch the Gradio App on Hugging Face Spaces
-# ---------------------
 iface.launch()