Update app.py

app.py

@@ -45,7 +45,7 @@ model.to(device)
 model.eval()
 
 # ---------------------------
-# Pre‑processing pipeline
+# Pre-processing pipeline
 # ---------------------------
 transform = transforms.Compose(
     [
@@ -56,35 +56,29 @@ transform = transforms.Compose(
 )
 
 # ---------------------------
-# Inference function
+# Inference function & temperature setting
 # ---------------------------
 
+TEMPERATURE = 3.5  # fixed temperature (between 3 and 4) for confidence calibration
 
-def predict(
-    img: Image.Image,
-    magnification: int,
-    ra_conc: float,
-    temperature: float = 1.0,
-):
-    """Return probabilities for High/Low CPM classes with optional temperature scaling.
+
+def predict(img: Image.Image, magnification: int, ra_conc: float):
+    """Return probabilities for High/Low CPM classes.
 
     Args:
         img: Microscopy image.
         magnification: Tag for objective magnification (×4/10/20).
         ra_conc: Tag for RA concentration (µM).
-        temperature: Temperature parameter for confidence calibration. T>1 lowers
-            confidence, T<1 increases confidence.
     """
     img_tensor = transform(img).unsqueeze(0).to(device)
     with torch.no_grad():
         logit = model(img_tensor)
-        # Temperature scaling for calibration
-        logit_scaled = logit / temperature
+        # Apply fixed temperature scaling to mitigate over‑confidence
+        logit_scaled = logit / TEMPERATURE
         prob_high = torch.sigmoid(logit_scaled).item()
 
     prob_low = 1.0 - prob_high
 
-    # gr.Label expects a mapping {class_name: probability}
     return {
         "High CPM Score": prob_high,
         "Low CPM Score": prob_low,
@@ -100,21 +94,13 @@ demo = gr.Interface(
         gr.Image(type="pil", label="Microscopy Image"),
         gr.Dropdown(choices=[4, 10, 20], value=10, label="Magnification (×)"),
         gr.Dropdown(choices=[0.1, 0.5, 1.0], value=0.1, label="RA Concentration (µM)"),
-        gr.Slider(
-            minimum=0.5,
-            maximum=5.0,
-            step=0.1,
-            value=1.0,
-            label="Temperature (confidence calibration)",
-            info="Increase temperature to reduce overconfidence",
-        ),
     ],
     outputs=gr.Label(num_top_classes=2, label="Predicted CPM Class & Probability"),
     title="iPS Cell Quality Classifier",
     description=(
         "Upload a microscopy image, choose magnification & RA concentration "
-        "(metadata only), then optionally adjust the *Temperature* slider to "
-        "calibrate confidence if predictions look over‑ or under‑confident."
+        "(metadata only). Probabilities have been temperature‑scaled for more "
+        "realistic confidence estimates."
     ),
 )