fixed cached examples

app.py

@@ -2,7 +2,6 @@ import glob
 import gradio as gr
 import numpy as np
 from os import environ
-from pathlib import Path
 from PIL import Image
 from torchvision import transforms as T
 from transformers import MaskFormerForInstanceSegmentation, MaskFormerImageProcessor
@@ -51,7 +50,7 @@ def visualize_instance_seg_mask(img_in, mask, id2label):
         img_out[i, j, :] = id2color[mask[i, j]]
         id2count[mask[i, j]] = id2count[mask[i, j]] + 1
 
-    image_res = (0.5 * img_in + 0.5 * img_out) / 255
+    image_res = (0.5 * img_in + 0.5 * img_out).astype(np.uint8)
 
     dataframe = [[
         f"{id2label[id]}",
@@ -70,19 +69,19 @@ def visualize_instance_seg_mask(img_in, mask, id2label):
 
 
 def query_image(image_path):
-    img = np.array(Image.open(Path(image_path)))
+    img = np.array(Image.open(image_path))
     img_size = (img.shape[0], img.shape[1])
     inputs = preprocessor(images=test_transform(img), return_tensors="pt")
 
     outputs = model(**inputs)
 
     results = preprocessor.post_process_semantic_segmentation(outputs=outputs, target_sizes=[img_size])[0]
-    results = visualize_instance_seg_mask(img, results.numpy(), model.config.id2label)
-    return results
+    mask_img, dataframe = visualize_instance_seg_mask(img, results.numpy(), model.config.id2label)
+    return mask_img, dataframe
 
 
 demo = gr.Interface(
-    query_image,
+    fn=query_image,
     inputs=[gr.Image(type="filepath", label="Input Image")],
     outputs=[
         gr.Image(label="Vegetation"),
@@ -92,7 +91,7 @@ demo = gr.Interface(
     allow_flagging="never",
     analytics_enabled=None,
     examples=example_images,
-    cache_examples=False
+    cache_examples=True
 )
 
 demo.launch(show_api=False)

requirements.txt

@@ -1,3 +1,4 @@
+gradio==3.14
 Pillow
 scipy
 torch