Update src/streamlit_app.py

src/streamlit_app.py

@@ -25,6 +25,22 @@ CLASS_NAMES_PATH = "src/class_names4.json"
 
 IMG_SIZE = (224, 224)
 
+# --------------------------------------------------
+# ✅ FIXED CLASS NAMES (YOUR TABLE) — fallback if JSON is wrong
+# --------------------------------------------------
+CLASS_NAMES_TABLE = [
+    "amanita",       # 0
+    "boletus",       # 1
+    "chantelle",     # 2
+    "deterrimus",    # 3
+    "rufus",         # 4
+    "torminosus",    # 5
+    "aurantiacum",   # 6
+    "procera",       # 7
+    "involutus",     # 8
+    "russula",       # 9
+]
+
 # --------------------------------------------------
 # Helpers
 # --------------------------------------------------
@@ -35,12 +51,39 @@ def _safe_listdir(path: str):
         return f"Could not list dir '{path}': {e}"
 
 def load_class_names(path: str) -> list:
+    """
+    Loads class names from JSON.
+    If JSON is missing or invalid or looks like ["0","1","2"...],
+    we fall back to CLASS_NAMES_TABLE.
+    """
+    # If file not found -> fallback
     if not os.path.exists(path):
-        raise FileNotFoundError(f"Missing class names file: {path}")
-    with open(path, "r", encoding="utf-8") as f:
-        names = json.load(f)
+        return CLASS_NAMES_TABLE
+
+    # Try load JSON
+    try:
+        with open(path, "r", encoding="utf-8") as f:
+            names = json.load(f)
+    except Exception:
+        return CLASS_NAMES_TABLE
+
+    # Must be list and non-empty
     if not isinstance(names, list) or len(names) == 0:
-        raise ValueError("class_names JSON must be a non-empty list.")
+        return CLASS_NAMES_TABLE
+
+    # If JSON contains only numbers as strings -> it's wrong -> fallback
+    # Example: ["0","1","2","3"...]
+    if all(isinstance(x, str) and x.strip().isdigit() for x in names):
+        return CLASS_NAMES_TABLE
+
+    # If JSON contains dict like {"amanita":0,...} -> convert to correct order
+    if isinstance(names, dict):
+        # Expect name->idx mapping
+        idx_to_name = {int(v): k for k, v in names.items()}
+        ordered = [idx_to_name[i] for i in range(len(idx_to_name))]
+        return ordered
+
+    # Otherwise: assume it's already correct list of names
     return names
 
 def build_resnet50_classifier(num_classes: int) -> tf.keras.Model:
@@ -59,7 +102,7 @@ def build_resnet50_classifier(num_classes: int) -> tf.keras.Model:
         tf.keras.layers.Dense(num_classes, activation="softmax"),
     ])
 
-    # Compile is not required for inference, but harmless:
+    # Compile not required for inference, but ok
     model.compile(optimizer="adam", loss="sparse_categorical_crossentropy", metrics=["accuracy"])
     return model
 
@@ -71,9 +114,7 @@ def load_model_and_assets(weights_path: str, class_names_path: str):
     if not os.path.exists(weights_path):
         raise FileNotFoundError(f"Missing weights file: {weights_path}")
 
-    # Load weights into the exact same architecture
     model.load_weights(weights_path)
-
     return model, class_names
 
 def preprocess_image(pil_img: Image.Image) -> np.ndarray:
@@ -105,12 +146,12 @@ except Exception as e:
     st.stop()
 
 # --------------------------------------------------
-# Show available classes
+# Show available classes (INDEX + NAME)
 # --------------------------------------------------
 with st.expander("🧪 Available mushroom classes (you can test these)"):
     st.write(f"Total classes: **{len(class_names)}**")
     for i, name in enumerate(class_names):
-        st.write(f"**{i}** — {name}")
+        st.write(f"**{i}** — **{name}**")
 
 # --------------------------------------------------
 # Image upload + prediction
@@ -128,10 +169,11 @@ else:
 
     x = preprocess_image(img)
 
-    preds = model.predict(x, verbose=0)[0]  # shape: (num_classes,)
+    preds = model.predict(x, verbose=0)[0]
     pred_idx = int(np.argmax(preds))
     pred_conf = float(preds[pred_idx])
-    pred_name = class_names[pred_idx] if pred_idx < len(class_names) else f"Class {pred_idx}"
+
+    pred_name = class_names[pred_idx] if 0 <= pred_idx < len(class_names) else f"Class {pred_idx}"
 
     st.subheader("✅ Prediction")
     st.write(f"**Predicted class index:** {pred_idx}")
@@ -141,6 +183,7 @@ else:
     st.subheader("🏁 Top-3 predictions")
     top3_idx = np.argsort(preds)[::-1][:3]
     for rank, idx in enumerate(top3_idx, start=1):
-        name = class_names[int(idx)]
-        prob = float(preds[int(idx)])
-        st.write(f"{rank}. **{name}** (class {int(idx)}) — **{prob*100:.2f}%**")
+        idx = int(idx)
+        name = class_names[idx] if 0 <= idx < len(class_names) else f"Class {idx}"
+        prob = float(preds[idx])
+        st.write(f"{rank}. **{name}** (class {idx}) — **{prob*100:.2f}%**")