Update app.py

app.py

@@ -27,9 +27,9 @@ def parse_roboflow_url(url: str):
 
 def convert_seg_to_bbox(api_key: str, dataset_url: str):
     """
-    1) Download segmentation dataset from Roboflow
-    2) Convert masks → YOLOv8 bboxes
-    Returns before_gallery, after_gallery, local_dataset_path, project_slug
+    Download segmentation dataset from Roboflow,
+    convert masks → YOLOv8 bboxes,
+    and return before/after galleries plus local path & auto-generated slug.
     """
     rf = Roboflow(api_key=api_key)
     ws, proj, ver = parse_roboflow_url(dataset_url)
@@ -85,7 +85,7 @@ def convert_seg_to_bbox(api_key: str, dataset_url: str):
     # locate images and write labels
     img_dir = None
     for dp, _, files in os.walk(root):
-        if any(f.lower().endswith(('.jpg','.png','jpeg')) for f in files):
+        if any(f.lower().endswith(('.jpg', '.png', 'jpeg')) for f in files):
             img_dir = dp
             break
     if not img_dir:
@@ -97,7 +97,7 @@ def convert_seg_to_bbox(api_key: str, dataset_url: str):
         if not os.path.isfile(src):
             continue
         shutil.copy(src, os.path.join(img_out, fname))
-        with open(os.path.join(lbl_out, fname.rsplit('.',1)[0] + ".txt"), 'w') as lf:
+        with open(os.path.join(lbl_out, fname.rsplit('.', 1)[0] + ".txt"), 'w') as lf:
             lf.write("\n".join(annos.get(pid, [])))
 
     # build preview galleries
@@ -110,54 +110,61 @@ def convert_seg_to_bbox(api_key: str, dataset_url: str):
         for a in coco['annotations']:
             if a['image_id'] != fname2id[fn]:
                 continue
-            pts = np.array(a['segmentation'][0], np.int32).reshape(-1,2)
-            cv2.polylines(seg_vis, [pts], True, (255,0,0), 2)
+            pts = np.array(a['segmentation'][0], np.int32).reshape(-1, 2)
+            cv2.polylines(seg_vis, [pts], True, (255, 0, 0), 2)
 
         box_vis = img.copy()
         for line in annos.get(fname2id[fn], []):
             _, cxn, cyn, wnorm, hnorm = map(float, line.split())
             iw, ih = images_info[fname2id[fn]]['width'], images_info[fname2id[fn]]['height']
-            w0, h0 = int(wnorm*iw), int(hnorm*ih)
-            x0, y0 = int(cxn*iw - w0/2), int(cyn*ih - h0/2)
-            cv2.rectangle(box_vis, (x0,y0), (x0+w0,y0+h0), (0,255,0), 2)
+            w0, h0 = int(wnorm * iw), int(hnorm * ih)
+            x0, y0 = int(cxn * iw - w0/2), int(cyn * ih - h0/2)
+            cv2.rectangle(box_vis, (x0, y0), (x0 + w0, y0 + h0), (0, 255, 0), 2)
 
         before.append(Image.fromarray(seg_vis))
         after.append(Image.fromarray(box_vis))
 
-    return before, after, out_root, proj  # proj is our slug
+    detection_slug = f"{proj}-detection"
+    return before, after, out_root, detection_slug
 
 
 def upload_and_train_detection(api_key: str, project_slug: str, dataset_path: str):
     """
-    1) Upload local YOLOv8 dataset to Roboflow
-    2) Generate & train a new detection version
-    Returns the hosted inference endpoint URL.
+    Upload a local YOLOv8 dataset to Roboflow, generate & train a new detection version,
+    and return the hosted inference endpoint URL.
     """
     rf = Roboflow(api_key=api_key)
     ws = rf.workspace()
 
-    # upload dataset
+    # 1) upload
     ws.upload_dataset(
         dataset_path,
         project_slug,
+        num_workers=10,
         project_license="MIT",
-        project_type="object-detection"
+        project_type="object-detection",
+        batch_name=None,
+        num_retries=0
     )
 
-    # generate a new version
+    # 2) generate a new version
     proj = ws.project(project_slug)
-    version_number = proj.generate_version(preprocessing={}, augmentation={})
+    new_version = proj.generate_version(settings={
+        "preprocessing": {},
+        "augmentation": {}
+    })
 
-    # train model (fast)
-    proj.version(version_number).train(speed="fast")
+    # 3) train (fast)
+    version = proj.version(new_version)
+    version.train(speed="fast")
 
-    # fetch hosted endpoint
-    m = proj.version(str(version_number)).model
+    # 4) get endpoint
+    m = version.model
     return f"{m['base_url']}{m['id']}?api_key={api_key}"
 
 
 with gr.Blocks() as app:
-    gr.Markdown("## 🔄 Segmentation → YOLOv8 Converter + Auto Trainer")
+    gr.Markdown("## 🔄 Segmentation → YOLOv8 Converter + Auto‑Trainer")
 
     # Converter UI
     api_input    = gr.Textbox(label="Roboflow API Key", type="password")
@@ -174,7 +181,7 @@ with gr.Blocks() as app:
         outputs=[before_gal, after_gal, state_path, state_slug]
     )
 
-    # Train UI
+    # Trainer UI
     train_btn     = gr.Button("Upload & Train Detection Model")
     endpoint_text = gr.Textbox(label="Hosted Detection Endpoint URL")