Update app.py

app.py

@@ -1,24 +1,47 @@
-from huggingface_hub import from_pretrained_fastai
+from gradio.outputs import Label
+from icevision.all import *
+from icevision.models.checkpoint import *
+import PIL
 import gradio as gr
-from fastai.vision.all import *
 import os
-try:
-    import toml
-except ImportError:
-    os.system('pip install toml')
-    import toml
 
-repo_id = "edgilr/ChestXRays"
+# Load model
+checkpoint_path = "edgilr/model_checkpoint.pth"
+checkpoint_and_model = model_from_checkpoint(checkpoint_path)
+model = checkpoint_and_model["model"]
+model_type = checkpoint_and_model["model_type"]
+class_map = checkpoint_and_model["class_map"]
 
-learner = from_pretrained_fastai(repo_id)
-labels = learner.dls.vocab
+# Transforms
+img_size = checkpoint_and_model["img_size"]
+valid_tfms = tfms.A.Adapter([*tfms.A.resize_and_pad(img_size), tfms.A.Normalize()])
 
-# 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=['normal.jpeg','pneumonia.jpeg']).launch(share=False)
+# Populate examples in Gradio interface
+examples = [
+    ['1.jpg'],
+    ['2.jpg'],
+    ['3.jpg']
+]
+
+def show_preds(input_image):
+    img = PIL.Image.fromarray(input_image, "RGB")
+    pred_dict = model_type.end2end_detect(img, valid_tfms, model, 
+                                          class_map=class_map, 
+                                          detection_threshold=0.5,
+                                          display_label=False, 
+                                          display_bbox=True, 
+                                          return_img=True, 
+                                          font_size=16, 
+                                          label_color="#FF59D6")
+    return pred_dict["img"]
+
+gr_interface = gr.Interface(
+    fn=show_preds,
+    inputs=["image"],
+    outputs=[gr.outputs.Image(type="pil", label="RetinaNet Inference")],
+    title="Fridge Object Detector",
+    description="A VFNet model that detects common objects found in fridge. Upload an image or click an example image below to use.",
+    examples=examples,
+)
+gr_interface.launch(inline=False, share=False, debug=True)