Update app.py

app.py

@@ -1,21 +1,19 @@
-from huggingface_hub import from_pretrained_fastai
 import gradio as gr
-from fastai.vision.all import *
-
-
-
-# repo_id = "YOUR_USERNAME/YOUR_LEARNER_NAME"
-repo_id = "LuisCe/Practica2"
+from icevision.all import *
+import PIL
+class_map = ClassMap(['raccoon'])
+model = models.torchvision.faster_rcnn.model(backbone=models.torchvision.faster_rcnn.backbones.resnet50_fpn(pretrained=True), num_classes=len(class_map))
+state_dict = torch.load('fasterRCNNRaccoonRESNET50.pth')
+model.load_state_dict(state_dict)
+size = 384
+infer_tfms = tfms.A.Adapter([*tfms.A.resize_and_pad(size),tfms.A.Normalize()])
+def predict(img):
+  # img = PIL.Image.open(img)
+  np.int = int
+  img = PIL.Image.fromarray(img)
 
-learner = from_pretrained_fastai(repo_id)
-labels = learner.dls.vocab
+  pred_dict = models.torchvision.faster_rcnn.end2end_detect(img, infer_tfms, model.to("cpu"), class_map=class_map, detection_threshold=0.5)
+  return pred_dict['img']
 
-# 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=['raccoon-177.jpg','raccoon-197.jpg']).launch(share=False)
- 
+# Creamos la interfaz y la lanzamos.
+gr.Interface(fn=predict, inputs=["image"], outputs=["image"], examples=['raccoon/train/images/raccoon-197.jpg','raccoon/train/images/raccoon-177.jpg']).launch(share=True,debug=True)