app.py

# AUTOGENERATED! DO NOT EDIT! File to edit: app.ipynb.

# %% auto 0
__all__ = ['learn', 'categories', 'image', 'label', 'examples', 'title', 'description', 'article', 'interpretation', 'interface',
           'classify_image']

# %% app.ipynb 2
from fastai.vision.all import *
import gradio

# %% app.ipynb 3
learn = load_learner('model.pkl')

# %% app.ipynb 7
categories = ('ash', 'chestnut', 'ginkgo biloba', 'silver maple', 'willow oak')

def classify_image(img):
  pred, idx, probs = learn.predict(img)
  # Change each probability to a float, since Gradio doesn't support Tensors or NumPy
  return dict(zip(categories, map(float, probs)))

# %% app.ipynb 10
image = gradio.Image(shape=(192, 192))
label = gradio.Label()
examples = ['images/ash.jpg', 'images/chestnut.jpg', 'images/ginkgo_biloba.jpg',
            'images/silver_maple.jpg', 'images/willow_oak.jpg']
# More useful args
title = "Tree leaf classifier demo"
description = "A tree leaf classifier demo, trained on images downloaded from DuckDuckGo. Created as a demo of HuggingFace Spaces and Gradio."
article = "<p>From this blog post: <a href='https://briansigafoos.com/ml-quick-start' target='_blank'>Machine Learning quick start by Brian Sigafoos</a></p>"
interpretation = 'default'

interface = gradio.Interface(fn=classify_image, inputs=image, outputs=label,
                             examples=examples, title=title, description=description,
                             article=article, interpretation=interpretation)
interface.launch(inline=False)