Update app.py

app.py

@@ -1,17 +1,51 @@
-from huggingface_hub import from_pretrained_fastai
-import gradio as gr
 from fastai.vision.all import *
+import gradio as gr
+import torchvision.transforms as transforms
+from pathlib import Path
+import PIL
+from huggingface_hub import from_pretrained_fastai
+
+# Cargamos el learner
+
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu") 
+repo_id = "macapa/segmentation-mod"
+model = from_pretrained_fastai(repo_id)
+model = model.cpu()
+model.eval()
+
 
-repo_id = "macapa/blindness-image-classification"
+def transform_image(image):
 
-learner = from_pretrained_fastai(repo_id)
-labels = learner.dls.vocab
+  #mask = PILMask.create(Path(str(image).replace("Images","Labels").replace("color","gt").replace(".jpg",".png")))
+  #image = PIL.Image.open(image)
+  my_transforms = transforms.Compose([transforms.ToTensor(),
+                                      transforms.Normalize(
+                                          [0.485, 0.456, 0.406],
+                                          [0.229, 0.224, 0.225])])
+  image_aux = image
+  #my_transforms(image_aux).unsqueeze(0).to(device)
+  image = transforms.Resize((480,640))(Image.fromarray(image))
+  tensor = my_transforms(image_aux).unsqueeze(0).to(device)
+  #tensor = transform_image(image=image)
+     
+  
+  
+  model.to(device)
+  with torch.no_grad():
+    outputs = model(tensor)
+  
+  outputs = torch.argmax(outputs,1)
+  
+  mask = np.array(outputs.cpu())
+  mask[mask==0]=255
+  mask[mask==1]=150
+  mask[mask==2]=76
+  mask[mask==3]=25
+  mask[mask==4]=0
+  
+  mask=np.reshape(mask,(480,640))
+  return Image.fromarray(mask.astype('uint8'))
 
-# Definimos una función que se encarga de llevar a cabo las predicciones
-def predict(img):
-    #img = PILImage.create(img)
-    pred,pred_idx,probs = learner.predict(img)
-    return {labels[i]: float(probs[i]) for i in range(len(labels))}
     
 # Creamos la interfaz y la lanzamos. 
-gr.Interface(fn=predict, inputs=gr.inputs.Image(shape=(128, 128)), outputs=gr.outputs.Label(num_top_classes=3),examples=['026dcd9af143.png','042470a92154.png']).launch(share=False)
+gr.Interface(fn=transform_image, inputs=gr.inputs.Image(shape=(640, 480)), outputs=gr.outputs.Image(),examples=['color_156.jpg','color_179.jpg']).launch(share=False)