update app.py

app.py

@@ -5,17 +5,21 @@ import gradio as gr
 import matplotlib
 import matplotlib.pyplot as plt
 import numpy as np
+import torch
 
 from PIL import Image
 
-from segment_anything import SamAutomaticMaskGenerator, sam_model_registry
+from segment_anything import SamAutomaticMaskGenerator, SamPredictor, sam_model_registry
 
 # suppress server-side GUI windows
 matplotlib.pyplot.switch_backend('Agg') 
 
-# setup model
+# setup models
+device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
 sam = sam_model_registry["vit_b"](checkpoint="./sam_vit_b_01ec64.pth")
+sam.to(device=device)
 mask_generator = SamAutomaticMaskGenerator(sam)
+predictor = SamPredictor(sam)
 
 
 # copied from: https://github.com/facebookresearch/segment-anything
@@ -60,15 +64,11 @@ with gr.Blocks() as demo:
 
     gr.Markdown("The Segment Anything Model (SAM) produces high quality object masks from input prompts such as points or boxes, and it can be used to generate masks for all objects in an image. It has been trained on a dataset of 11 million images and 1.1 billion masks, and has strong zero-shot performance on a variety of segmentation tasks.")
 
-    with gr.Tabs():
+    with gr.Row():
+        image_input = gr.Image()
+        image_output = gr.Image()
 
-        with gr.TabItem("Mask Generator"):
-
-            with gr.Row():
-                image_input = gr.Image()
-                image_output = gr.Image()
-            segment_image_button = gr.Button('Generate Mask')
-    
-        segment_image_button.click(segment_image, inputs=[image_input], outputs=image_output)
+    segment_image_button = gr.Button('Generate Mask')
+    segment_image_button.click(segment_image, inputs=[image_input], outputs=image_output)
 
 demo.launch()