updated image link

app.py

@@ -7,7 +7,26 @@ import json
 from collections import Counter
 import base64
 
+# -----------------------------------------------------------------
+# DEBUG: Check if the image file exists and encode it as base64
+# -----------------------------------------------------------------
+image_path = Path("hcw_classification.png")
+print(f"DEBUG: Looking for image at {image_path.absolute()}")
 
+if image_path.exists():
+    print(f"DEBUG: Image file found, size = {image_path.stat().st_size} bytes")
+    with open(image_path, "rb") as f:
+        img_b64 = base64.b64encode(f.read()).decode("utf-8")
+        image_data_uri = f"data:image/png;base64,{img_b64}"
+    print(f"DEBUG: Base64 length = {len(image_data_uri)} characters")
+else:
+    print("DEBUG: WARNING – image file NOT found, using fallback transparent pixel")
+    # fallback 1x1 transparent pixel
+    image_data_uri = "data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNk+M9QDwADhgGAWjR9awAAAABJRU5ErkJggg=="
+
+# -----------------------------------------------------------------
+# Load models and labels
+# -----------------------------------------------------------------
 model_paths = [Path(f"model/fold_{i}_best.pt") for i in range(5)]
 models = [YOLO(str(path)) for path in model_paths]
 
@@ -27,7 +46,6 @@ def get_label_info(name):
     print(f"Label '{name}' not found. Available labels are: {list(labels_info.keys())}")
     return ["Unknown", "Unknown"]
 
-
 def classify_image(img, conf_threshold=0.25, iou_threshold=0.45):
     predicted_classes = []
     results_for_plot = None
@@ -57,19 +75,9 @@ def classify_image(img, conf_threshold=0.25, iou_threshold=0.45):
 
     return Image.fromarray(annotated_img), waste_details
 
-# ------------------------------------------------------------
-# EMBED THE IMAGE AS BASE64 DATA URI (no external requests)
-# ------------------------------------------------------------
-image_path = Path("hcw_classification.png")
-if image_path.exists():
-    with open(image_path, "rb") as f:
-        img_b64 = base64.b64encode(f.read()).decode("utf-8")
-        image_data_uri = f"data:image/png;base64,{img_b64}"
-else:
-    # Fallback: a tiny 1x1 transparent pixel (in case the file is missing)
-    image_data_uri = "data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNk+M9QDwADhgGAWjR9awAAAABJRU5ErkJggg=="
-# ------------------------------------------------------------
-
+# -----------------------------------------------------------------
+# Build the article HTML – using + concatenation to avoid f‑string issues
+# -----------------------------------------------------------------
 article_html = (
     "<div style='display: flex; flex-wrap: wrap; gap: 30px; align-items: stretch;'>"
     "  <div style='flex: 2; min-width: 300px;'>"
@@ -107,12 +115,13 @@ article_html = (
     "  </div>"
     "  <div style='flex: 1; min-width: 250px; display: flex; flex-direction: column; align-items: center;'>"
     "    <figure style='margin:0; text-align:center;'>"
-    f"      <img src='{image_data_uri}' alt='HCW Classification Table Visualization' "
-    "           style='height: 500px; width: auto; max-width: 100%; object-fit: contain; "
-    "                  border: 1px solid #ddd; border-radius: 8px; box-shadow: 0 2px 8px rgba(0,0,0,0.1);' />"
-    "      <figcaption style='margin-top: 8px; font-style: italic; color: #555;'>"
-    "        Color codes for waste segregation (based on Nepal's HCW standards)"
-    "      </figcaption>"
+    # Insert the image using simple string concatenation – no f‑string
+    + "      <img src='" + image_data_uri + "' alt='HCW Classification Table Visualization' "
+    + "           style='height: 500px; width: auto; max-width: 100%; object-fit: contain; "
+    + "                  border: 1px solid #ddd; border-radius: 8px; box-shadow: 0 2px 8px rgba(0,0,0,0.1);' />"
+    + "      <figcaption style='margin-top: 8px; font-style: italic; color: #555;'>"
+    + "        Color codes for waste segregation (based on Nepal's HCW standards)"
+    + "      </figcaption>"
     "    </figure>"
     "  </div>"
     "</div>"
@@ -135,17 +144,12 @@ article_html = (
     "<a href='http://dwaste.live' target='_blank'>dwaste.live</a></p>"
 )
 
-iface = gr.Interface(
-    fn=classify_image,
-    inputs=[
-        gr.Image(type="pil", label="Upload Image"),
-    ],
-    outputs=[
-        gr.Image(label="Result"),
-        gr.Dataframe(headers=["Name", "Type", "Color Code"], label="Details")
-    ],
-    title="Health Care Waste Classification",
-    description=(
+# -----------------------------------------------------------------
+# Gradio interface using Blocks for reliable HTML insertion
+# -----------------------------------------------------------------
+with gr.Blocks() as iface:
+    gr.Markdown("# Health Care Waste Classification")
+    gr.Markdown(
         "DWaste uses lightweight AI to classify and sort health care waste (HCW) into various categories "
         "according to the <strong>National Health Care Waste Management Standards and Operating Procedures of Nepal</strong> "
         "color coding system to streamline waste management processes for healthcare facilities. "
@@ -153,8 +157,15 @@ iface = gr.Interface(
         '<p><strong>Related Research:</strong> See our supporting study on <a href="https://www.researchgate.net/publication/399424815_Health_care_waste_classification_using_deep_learning_aligned_with_Nepal\'s_bin_color_guidelines" target="_blank">Kathmandu University Journal of Science, Engineering and Technology</a>.</p>'
         "<p><strong>Disclaimer:</strong> This tool is for informational purposes only. Predictions made by the AI model "
         "may not always be accurate. Please use the results cautiously and verify if necessary.</p>"
-    ),
-    article=article_html
-)
+    )
+    with gr.Row():
+        with gr.Column():
+            input_img = gr.Image(type="pil", label="Upload Image")
+        with gr.Column():
+            output_img = gr.Image(label="Result")
+            output_df = gr.Dataframe(headers=["Name", "Type", "Color Code"], label="Details")
+    btn = gr.Button("Classify")
+    btn.click(fn=classify_image, inputs=[input_img], outputs=[output_img, output_df])
+    gr.HTML(article_html)   # The article + image appears below the interactive area
 
 iface.launch()