Update src/streamlit_app.py

src/streamlit_app.py

@@ -48,6 +48,7 @@ st.markdown(
 # Load WHOOP logo
 WHOOP_LOGO = "https://www.whoop.com/wp-content/themes/whoop/library/images/whoop-logo-dark.svg"
 
+# Main application
 def main():
     # Display WHOOP logo at top
     st.image(WHOOP_LOGO, width=200)
@@ -65,11 +66,13 @@ def main():
         """
     )
 
+    # Environment setup
     hf_token = os.getenv("HF_TOKEN", None)
     cache_dir = "/tmp/cache"
     os.makedirs(cache_dir, exist_ok=True)
     os.environ["HUGGINGFACE_HUB_CACHE"] = cache_dir
 
+    # Nutrition data
     nutritional_info = {
         "pizza": {"serving": "100 g (1 slice)", "calories": "270 kcal", "protein": "12 g", "carbs": "34 g", "fat": "10 g", "ingredients": "dough, tomato sauce, mozzarella cheese", "method": "baked", "substitute": "cauliflower crust"},
         "hamburger": {"serving": "150 g", "calories": "300 kcal", "protein": "20 g", "carbs": "30 g", "fat": "12 g", "ingredients": "ground beef patty, bun, lettuce, tomato", "method": "grilled or pan-fried", "substitute": "chicken patty"},
@@ -82,19 +85,19 @@ def main():
         "steak": {"serving": "113 g (4 oz)", "calories": "250 kcal", "protein": "25 g", "carbs": "0 g", "fat": "15 g", "ingredients": "beef sirloin, salt, pepper", "method": "grilled or pan-seared", "substitute": "leaner cut (filet mignon)"},
         "chocolate_cake": {"serving": "100 g (1 slice)", "calories": "350 kcal", "protein": "5 g", "carbs": "50 g", "fat": "15 g", "ingredients": "flour, sugar, cocoa, butter, eggs", "method": "baked", "substitute": "gluten-free flour"}
     }
+    label_mapping = {"caesar_salad": "salad", "spaghetti_bolognese": "pasta"}
 
-    label_mapping = {
-        "caesar_salad": "salad",
-        "spaghetti_bolognese": "pasta"
-    }
-
+    # Sidebar
     st.sidebar.image(WHOOP_LOGO, width=150)
     st.sidebar.header("WHOOP Model Suite")
     st.sidebar.markdown(
-        "- 🖼️ **Image Classifier**: `shingguy1/fine_tuned_vit`
-        - 💬 **Nutrition Paraphraser**: `google/flan-t5-small`"
+        """
+        - 🖼️ **Image Classifier**: `shingguy1/fine_tuned_vit`
+        - 💬 **Nutrition Paraphraser**: `google/flan-t5-small`
+        """
     )
 
+    # Image transforms
     transform = transforms.Compose([
         transforms.Resize(256),
         transforms.CenterCrop(224),
@@ -107,38 +110,30 @@ def main():
     def load_models():
         device = torch.device("cpu")
         vit = ViTForImageClassification.from_pretrained(
-            "shingguy1/fine_tuned_vit",
-            cache_dir=cache_dir,
-            use_auth_token=hf_token
+            "shingguy1/fine_tuned_vit", cache_dir=cache_dir, use_auth_token=hf_token
         ).to(device)
         tok = AutoTokenizer.from_pretrained(
-            "google/flan-t5-small",
-            cache_dir=cache_dir,
-            use_auth_token=hf_token
+            "google/flan-t5-small", cache_dir=cache_dir, use_auth_token=hf_token
         )
         t5 = T5ForConditionalGeneration.from_pretrained(
-            "google/flan-t5-small",
-            cache_dir=cache_dir,
-            use_auth_token=hf_token
+            "google/flan-t5-small", cache_dir=cache_dir, use_auth_token=hf_token
         ).to(device)
         return vit, tok, t5, device
 
     model_vit, tokenizer_t5, model_t5, device = load_models()
 
-    uploaded = st.file_uploader("📷 Upload a food image...", type=["jpg", "png", "jpeg"])
+    # File uploader and inference loop
+    uploaded = st.file_uploader("📷 Upload a food image...", type=["jpg","png","jpeg"])
     if uploaded:
         img = Image.open(uploaded)
         st.image(img, caption="Your Food", use_column_width=True)
-
         inp = transform(img).unsqueeze(0).to(device)
-        with torch.no_grad():
-            out = model_vit(pixel_values=inp)
+        with torch.no_grad(): out = model_vit(pixel_values=inp)
         label = model_vit.config.id2label[out.logits.argmax(-1).item()]
         st.success(f"🍽️ Detected: **{label}**")
 
         true_label = label_mapping.get(label.lower(), label.lower())
         data = nutritional_info.get(true_label)
-
         if data:
             base_description = (
                 f"A typical {true_label} serving ({data['serving']}) contains about {data['calories']}, "
@@ -149,32 +144,17 @@ def main():
             prompt = (
                 f"Paraphrase the following nutritional facts in a friendly, conversational tone. "
                 f"Use varied sentence structures and synonyms, and feel free to generalize numeric details "
-                f"(e.g., ‘around 250 kcal’). Don’t add any new facts.
-
-" + base_description
+                f"(e.g., ‘around 250 kcal’). Don’t add any new facts.\n\n" + base_description
             )
         else:
-            prompt = (
-                f"Provide an approximate nutrition summary for {label}, including calories, "
-                f"macronutrients, and a brief description."
-            )
+            prompt = f"Provide an approximate nutrition summary for {label}, including calories, macronutrients, and a brief description."
 
         inputs = tokenizer_t5(prompt, return_tensors="pt", truncation=True).to(device)
-        output_ids = model_t5.generate(
-            inputs["input_ids"],
-            max_new_tokens=100,
-            do_sample=True,
-            top_p=0.9,
-            temperature=0.7,
-            early_stopping=True
-        )
+        output_ids = model_t5.generate(inputs["input_ids"], max_new_tokens=100, do_sample=True, top_p=0.9, temperature=0.7, early_stopping=True)
         response = tokenizer_t5.decode(output_ids[0], skip_special_tokens=True)
-
-        if "calories" not in response.lower() or len(response.split()) < 10:
-            response = base_description
+        if "calories" not in response.lower() or len(response.split()) < 10: response = base_description
 
         st.subheader("🧾 Nutrition Overview")
         st.info(response)
 
-if __name__ == "__main__":
-    main()
+if __name__ == "__main__": main()