Update app.py

app.py

@@ -225,20 +225,19 @@ def inference_pipeline(img, thresh_b=0.5, thresh_v=0.5, seg_thresh=0.5):
 
 
 title = "Chest X-ray: UNet segmentation + 2 binary classifiers"
-desc = "Pipeline: UNet -> mask lungs -> two binary classifiers (Normal vs Bacterial, Normal vs Viral). " \
-       "If both classifiers fire, the stronger probability is chosen (fallback). Thresholds adjustable."
+desc = (
+    "Pipeline: UNet -> mask lungs -> two binary classifiers (Normal vs Bacterial, Normal vs Viral). "
+    "If both classifiers fire, the stronger probability is chosen (fallback). Thresholds adjustable."
+)
 
 
-def encode_image(img_path):
-    with open(img_path, "rb") as img:
-        return "data:image/png;base64," + base64.b64encode(img.read()).decode()
-
 example_samples = [
-    [f"<img src='{encode_image('images/NORMAL.jpeg')}' width='120'>", "NORMAL"],
-    [f"<img src='{encode_image('images/VIRAL.jpeg')}' width='120'>", "VIRAL"],
-    [f"<img src='{encode_image('images/BACT.jpeg')}' width='120'>", "BACTERIAL"],
+    ["images/NORMAL.jpeg", "NORMAL"],
+    ["images/VIRAL.jpeg", "VIRAL"],
+    ["images/BACT.jpeg", "BACTERIAL"],
 ]
 
+
 with gr.Blocks(title=title) as demo:
     gr.Markdown(f"### {title}")
     gr.Markdown(desc)
@@ -247,9 +246,12 @@ with gr.Blocks(title=title) as demo:
         fn=inference_pipeline,
         inputs=[
             gr.Image(type="numpy", label="Upload chest X-ray (RGB or grayscale)"),
-            gr.Slider(minimum=0.1, maximum=0.9, step=0.01, value=0.5, label="Bacterial threshold (thresh_b)"),
-            gr.Slider(minimum=0.1, maximum=0.9, step=0.01, value=0.5, label="Viral threshold (thresh_v)"),
-            gr.Slider(minimum=0.1, maximum=0.9, step=0.01, value=0.5, label="Segmentation mask threshold (seg_thresh)")
+            gr.Slider(minimum=0.1, maximum=0.9, step=0.01, value=0.5,
+                      label="Bacterial threshold (thresh_b)"),
+            gr.Slider(minimum=0.1, maximum=0.9, step=0.01, value=0.5,
+                      label="Viral threshold (thresh_v)"),
+            gr.Slider(minimum=0.1, maximum=0.9, step=0.01, value=0.5,
+                      label="Segmentation mask threshold (seg_thresh)")
         ],
         outputs=[
             gr.Label(num_top_classes=1, label="Prediction"),
@@ -266,8 +268,8 @@ with gr.Blocks(title=title) as demo:
     gr.Dataframe(
         headers=["Image", "Label"],
         value=example_samples,
+        datatype=["image", "str"], 
         interactive=False,
-        wrap=True,
         row_count=(len(example_samples), "fixed"),
         col_count=(2, "fixed")
     )