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 = "igmarco/grapes-semanticsegmentation"

# learner = from_pretrained_fastai(repo_id)

import torchvision.transforms as transforms
import PIL

device = torch.device("cuda" if torch.cuda.is_available() else "cpu") 
model = torch.jit.load("Pr1.pth")
model = model.cpu()

def transform_image(image):
    my_transforms = transforms.Compose([transforms.ToTensor(),
                                        transforms.Normalize(
                                            [0.485, 0.456, 0.406],
                                            [0.229, 0.224, 0.225])])
    image_aux = image
    return my_transforms(image_aux).unsqueeze(0).to(device)

# Definimos una función que se encarga de llevar a cabo las predicciones
def predict(img):
    img_pil = PIL.Image.fromarray(img, 'RGB')
    
    image = transforms.Resize((480,640))(img_pil)
    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]=0
    mask[mask==1]=150
    mask[mask==2]=25
    mask[mask==3]=74
    mask[mask==4]=255

    mask=np.reshape(mask,(480,640))

    return(Image.fromarray(mask.astype('uint8')))
    
# Creamos la interfaz y la lanzamos. 
gr.Interface(fn=predict, inputs=gr.inputs.Image(shape=(480, 640)), outputs=gr.outputs.Image(type="pil")).launch(share=False)
# gr.Interface(fn=predict, inputs=gr.inputs.Image(shape=(480, 640)), outputs=gr.outputs.Image(type="pil"),examples=['grapes1.jpg','grapes2.jpg']).launch(share=False)