Spaces:
Sleeping
Sleeping
File size: 1,220 Bytes
be017b2 630ff4d be017b2 9e4b544 fb587ae be017b2 9e4b544 630ff4d 9e4b544 630ff4d 9e4b544 630ff4d 9e4b544 fb587ae 9e4b544 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 |
import gradio as gr
from fastai.vision.all import *
import skimage
# Load the trained model
learn = load_learner('ripeorrotten_apple.pkl')
# Define the labels for your model
labels = ['ripe', 'rotten']
# Define the prediction function
def predict(img):
img = PILImage.create(img)
pred, pred_idx, probs = learn.predict(img)
return {labels[i]: float(probs[i]) for i in range(len(labels))}
# Customize the Gradio interface
title = "Apple Ripeness Classifier"
description = "Is your apple ripe or rotten? Use this AQCC (apple quality control classifier) trained on web images with fastai."
article = "<p style='text-align: center'><a href='https://tmabraham.github.io/blog/gradio_hf_spaces_tutorial' target='_blank'>Blog post</a></p>"
# Example images (ensure these are valid paths or URLs)
examples = ['apple-fruit-ripe.jpg']
interpretation = 'default'
enable_queue = True
# Create and launch the Gradio interface
gr.Interface(
fn=predict,
inputs=gr.inputs.Image(shape=(512, 512)),
outputs=gr.outputs.Label(num_top_classes=2),
title=title,
description=description,
article=article,
examples=examples,
interpretation=interpretation,
enable_queue=enable_queue
).launch()
|