get predict meta

app.py

@@ -2,15 +2,15 @@ import gradio as gr
 from backend import visualize_image
 
 # gradio inputs
-image_input = gr.inputs.Image(type="pil", label="Input Image")
-color_mode_select = gr.inputs.Radio(["Black/white", "Random", "Segmentation"], label="Color Mode", default="Segmentation")
-mode_dropdown = gr.inputs.Dropdown(["Trees", "Buildings", "Both"], label="Detection Mode", default="Both")
+image_input = gr.components.Image(type="pil", label="Input Image")
+color_mode_select = gr.components.Radio(["Black/white", "Random", "Segmentation"], label="Color Mode", default="Segmentation")
+mode_dropdown = gr.components.Dropdown(["Trees", "Buildings", "Both"], label="Detection Mode", default="Both")
 
-tree_threshold_slider = gr.inputs.Slider(0, 1, 0.1, 0.7, label='Set confidence threshold "%" for trees')
-building_threshold_slider = gr.inputs.Slider(0, 1, 0.1, 0.7, label='Set confidence threshold "%" for buildings')
+tree_threshold_slider = gr.components.Slider(0, 1, 0.1, 0.7, label='Set confidence threshold "%" for trees')
+building_threshold_slider = gr.components.Slider(0, 1, 0.1, 0.7, label='Set confidence threshold "%" for buildings')
 
 # gradio outputs
-output_image = gr.outputs.Image(type="pil", label="Output Image")
+output_image = gr.components.Image(type="pil", label="Output Image")
 title = "Aerial Image Segmentation"
 description = "An instance segmentation demo for identifying boundaries of buildings and trees in aerial images using DETR (End-to-End Object Detection) model with MaskRCNN-101 backbone"
 

backend.py

@@ -81,18 +81,18 @@ def visualize_image(im, mode="BOTH", tree_threshold=0.7, building_threshold=0.7,
         building_instances = segment_building(im, building_threshold)
         instances = Instances.cat([tree_instances, building_instances])
         
-    meta = MetadataCatalog.get("predict")
+    metadata = MetadataCatalog.get("predict")
+    print("metadata", type(metadata), metadata)
+    print('metadata.get("thing_classes")', type(metadata.get("thing_classes")), metadata.get("thing_classes"))
 
     visualizer = Visualizer(im[:, :, ::-1],
-                            metadata=meta,
+                            metadata=metadata,
                             scale=0.5,
                             instance_mode=color_mode)
 
-    dataset_names = MetadataCatalog.list()
-    print(dataset_names)
-    metadata = MetadataCatalog.get(dataset_names[0])
-    print("metadata", type(metadata), metadata)
-    print('metadata.get("thing_classes")', type(metadata.get("thing_classes")), metadata.get("thing_classes"))
+    # dataset_names = MetadataCatalog.list()
+    # print(dataset_names)
+
     # category_names = metadata.get("thing_classes")
     # visualizer = Visualizer(im[:, :, ::-1],
     #                         metadata=metadata,