working locally end2end

app.py

@@ -23,6 +23,8 @@ OUTPUT_DIR = "/tmp/output/"
 MODEL_DIR = os.path.join(BASE_DIR, "rcnn_model", "scripts")
 logo_path = os.path.join(BASE_DIR, "public", "logo.png")
 model_path = os.path.join(OUTPUT_DIR, "model_final.pth")
+#  changine the model directory to the tmp directory
+# model_path = os.path.join(OUTPUT_DIR, "model_final.pth")
 
 # Google Drive model
 GOOGLE_DRIVE_FILE_ID = "1yr64AOgaYZPTcQzG6cxG6lWBENHR9qjW"
@@ -53,61 +55,14 @@ cfg = write_config()
 # MAIN PREDICTION FUNCTION
 # ==================================
 
-# def predict(uploaded_file):
-#     if uploaded_file is None:
-#         return None, None, "No file uploaded."
-#     # uploaded_path = uploaded_file
-#     # input_filename = os.path.basename(uploaded_path)
-#     # print(f"✅ Image received at {uploaded_path}")
-
-#     uploaded_path = os.path.join(UPLOAD_DIR, uploaded_file.name)
-#     with open(uploaded_path, "wb") as f:
-#         f.write(uploaded_file.read())
-#     print(f"✅ Image saved to {uploaded_path}")
-
-
-#     # Save uploaded image
-#     # input_bytes = uploaded_file.read()
-#     # img = Image.open(BytesIO(input_bytes)).convert("RGB")
-#     # input_filename = uploaded_file.name
-#     # uploaded_path = os.path.join(UPLOAD_DIR, input_filename)
-#     # img.save(uploaded_path)
-#     # print(f"✅ Image saved to {uploaded_path}")
-
-#     # Prepare output paths
-#     input_filename = uploaded_file.name
-#     output_json_name = input_filename.replace(".png", "_result.json").replace(".jpg", "_result.json").replace(".jpeg", "_result.json")
-#     output_image_name = input_filename.replace(".png", "_result.png").replace(".jpg", "_result.png").replace(".jpeg", "_result.png")
-
-#     output_json_path = os.path.join(JSON_DIR, output_json_name)
-#     output_image_path = os.path.join(JSON_DIR, output_image_name)
-
-#     # Run model
-#     main(cfg, uploaded_path, output_json_path, output_image_path)
-
-#     # Read outputs
-#     result_img = Image.open(output_image_path) if os.path.exists(output_image_path) else None
-#     result_json = {}
-#     if os.path.exists(output_json_path):
-#         with open(output_json_path, "r") as jf:
-#             result_json = json.load(jf)
-
-#     return result_img, json.dumps(result_json, indent=2), None
-
 def predict(uploaded_file_path):
+    print("Inside Predict:" + uploaded_file_path)
     if uploaded_file_path is None:
         return None, None, "No file uploaded."
 
-    # uploaded_path = uploaded_file.name
-    # input_filename = os.path.basename(uploaded_path)
-    # print(f"✅ Image received at {uploaded_path}")
-
-    uploaded_path = os.path.join(UPLOAD_DIR, uploaded_file.name)
-    with open(uploaded_path, "wb") as f:
-        f.write(uploaded_file.read())
-    print(f"✅ Image saved to {uploaded_path}")
-
-    input_filename = uploaded_file.name
+    uploaded_path = os.path.join(UPLOAD_DIR, "input_image.png")
+    print("Saved uploaded image to:", uploaded_path)
+    input_filename = "input_image.png"
 
     # Prepare output paths
     output_json_name = input_filename.replace(".png", "_result.json").replace(".jpg", "_result.json").replace(".jpeg", "_result.json")
@@ -116,8 +71,9 @@ def predict(uploaded_file_path):
     output_json_path = os.path.join(JSON_DIR, output_json_name)
     output_image_path = os.path.join(JSON_DIR, output_image_name)
 
+    # print(f"Before calling main in app.py: {uploaded_file.name}")
     # Run model
-    main(cfg, uploaded_path, output_json_path, output_image_path)
+    main(cfg, uploaded_file_path, output_json_name, output_image_name)
 
     # Read outputs
     result_img = Image.open(output_image_path) if os.path.exists(output_image_path) else None
@@ -137,8 +93,8 @@ with gr.Blocks() as demo:
 
     with gr.Row():
         with gr.Column():
-            # uploaded_file = gr.File(label="Upload your Floorplan Image", type="filepath")
-            uploaded_file = gr.File(label="Upload your Floorplan Image", type="file")
+            uploaded_file = gr.File(label="Upload your Floorplan Image", type="filepath")
+            # uploaded_file = gr.File(label="Upload your Floorplan Image", type="file")
             run_button = gr.Button("Run Vectorizer 🔥")
 
         with gr.Column():

rcnn_model/scripts/rcnn_run.py

@@ -29,7 +29,8 @@ sample_data_directory = "rcnn_model/sample/"
 
 def main(cfg,img_source_path, coco_dest_filename, val_img_dest_filename):
     os.chdir(str(from_root()))
-
+    
+    print("Inside main: " + img_source_path)
     #configure model
     cfg.DATALOADER.NUM_WORKERS = 1
     cfg.SOLVER.IMS_PER_BATCH = 1                                           
@@ -49,6 +50,7 @@ def prediction_runner(filename, coco_dest_path, val_img_dest_path, predictor, se
     #set up annotation builder
     ann_builder = instantiate_ann_build()
 
+    print("Filename inside prediction_runner: ", filename)
     #set up image
     initImg = preprocess_image(filename)
     init_width = initImg.shape[1]

requirements.txt

@@ -8,4 +8,4 @@ matplotlib
 labelme2coco
 numpy
 from_root
-gradio==3.50.2
+pip install gradio==4.44.1