Update app.py

app.py

@@ -5,21 +5,14 @@ from PIL import Image
 import os
 import time
 import joblib
-import pickle
-import torch
-import torch.nn as nn
 
-# ===============================
-# Step 1: Load Test Dataset
-# ===============================
-
-label_df = pd.read_csv("labels.csv")  # columns: filename,label
+# Load test data
+label_df = pd.read_csv("labels.csv")
 test_dir = "test_images"
 
 def image_to_features(image: Image.Image) -> np.ndarray:
     image = image.resize((64, 64))
-    array = np.asarray(image.convert("L"))
-    return array.flatten()
+    return np.array(image.convert("L")).flatten()
 
 X_test = []
 y_test = []
@@ -27,101 +20,40 @@ y_test = []
 for _, row in label_df.iterrows():
     img_path = os.path.join(test_dir, row["filename"])
     label = row["label"]
-    img = Image.open(img_path)
-    features = image_to_features(img)
-    X_test.append(features)
-    y_test.append(label)
+    try:
+        img = Image.open(img_path)
+        X_test.append(image_to_features(img))
+        y_test.append(label)
+    except Exception as e:
+        print(f"Failed to load {img_path}: {e}")
 
 X_test = np.array(X_test)
 y_test = np.array(y_test)
 
-# ===============================
-# Step 2: Define Leaderboard File
-# ===============================
-
+# Initialize leaderboard
 LEADERBOARD_PATH = "leaderboard.csv"
 if not os.path.exists(LEADERBOARD_PATH):
     pd.DataFrame(columns=["Name", "Accuracy", "Avg Time (ms)"]).to_csv(LEADERBOARD_PATH, index=False)
 
-# ===============================
-# Step 3: Define Model Wrappers
-# ===============================
-
-class SimpleTorchModel(nn.Module):
-    def __init__(self):
-        super().__init__()
-        self.fc = nn.Sequential(
-            nn.Linear(64 * 64, 128),
-            nn.ReLU(),
-            nn.Linear(128, 2)
-        )
-
-    def forward(self, x):
-        return self.fc(x)
-
-class SKLearnWrapper:
-    def __init__(self, model):
-        self.model = model
-
-    def predict(self, X: np.ndarray) -> np.ndarray:
-        return self.model.predict(X)
-
-class TorchPredictWrapper:
-    def __init__(self, model_path: str):
-        self.model = SimpleTorchModel()
-        self.model.load_state_dict(torch.load(model_path, map_location="cpu"))
-        self.model.eval()
-
-    def predict(self, X: np.ndarray) -> np.ndarray:
-        with torch.no_grad():
-            tensor = torch.tensor(X, dtype=torch.float32)
-            logits = self.model(tensor)
-            return torch.argmax(logits, dim=1).numpy()
-
-# ===============================
-# Step 4: Load Model From File
-# ===============================
-
-def load_model(file_path: str):
-    try:
-        model = joblib.load(file_path)
-        return SKLearnWrapper(model)  # now wrapping just clf
-    except Exception as e1:
-        try:
-            with open(file_path, "rb") as f:
-                model = pickle.load(f)
-            return SKLearnWrapper(model)
-        except Exception as e2:
-            try:
-                return TorchPredictWrapper(file_path)
-            except Exception as e3:
-                raise RuntimeError(f"Unrecognized model format or failed to load:\n"
-                                   f"joblib error: {e1}\n"
-                                   f"pickle error: {e2}\n"
-                                   f"torch error: {e3}")
-
-
-# ===============================
-# Step 5: Evaluation Logic
-# ===============================
-
+# Evaluation function
 def evaluate_model(file, name):
     try:
-        model = load_model(file.name)
+        model = joblib.load(file.name)
 
         start = time.time()
         y_pred = model.predict(X_test)
-        elapsed = (time.time() - start) * 1000  # in ms
+        elapsed = (time.time() - start) * 1000
 
         accuracy = 100 * (y_pred == y_test).mean()
         avg_time = elapsed / len(X_test)
 
         leaderboard = pd.read_csv(LEADERBOARD_PATH)
-        leaderboard = pd.concat([leaderboard, pd.DataFrame([{
+        new_entry = pd.DataFrame([{
             "Name": name,
             "Accuracy": round(accuracy, 2),
             "Avg Time (ms)": round(avg_time, 2)
-        }])])
+        }])
+        leaderboard = pd.concat([leaderboard, new_entry], ignore_index=True)
         leaderboard.to_csv(LEADERBOARD_PATH, index=False)
 
         return leaderboard
@@ -129,19 +61,14 @@ def evaluate_model(file, name):
     except Exception as e:
         return f"❌ Error during evaluation: {e}"
 
-# ===============================
-# Step 6: Gradio Interface
-# ===============================
-
-interface = gr.Interface(
+# Gradio interface
+gr.Interface(
     fn=evaluate_model,
     inputs=[
-        gr.File(label="Upload your `.joblib`, `.pkl`, or `.pth` model"),
-        gr.Text(label="Your (group) name(s)")
+        gr.File(label="Upload your `.joblib` model"),
+        gr.Text(label="Your name or team")
     ],
     outputs=gr.Dataframe(label="Leaderboard"),
-    title="Insect Classifier Leaderboard",
-    description="Upload a `.joblib`, `.pkl`, or `.pth` model that implements `.predict(X)` on 64×64 grayscale image vectors. We will evaluate your model and update the leaderboard."
-)
-
-interface.launch()
+    title="🧠 Joblib Classifier Leaderboard",
+    description="Upload a `.joblib` file for 64×64 grayscale image classification. Accuracy and speed will be measured."
+).launch()