app.py

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

# %% auto 0
__all__ = ['learn', 'categories', 'image', 'label', 'examples', 'intf', 'is_cat', 'classify_image']

# %% dog_v_cat.ipynb 1
from fastai.vision.all import *
import gradio as gr

def is_cat(x): 
    return x[0].isupper()

# %% dog_v_cat.ipynb 2
from fastai.vision.all import *
import gradio as gr

def is_cat(x): 
    return x[0].isupper()

# %% dog_v_cat.ipynb 4
from fastai.vision.all import *
import gradio as gr

def is_cat(x): 
    return x[0].isupper()

# %% dog_v_cat.ipynb 6
from fastai.vision.all import *
import gradio as gr

def is_cat(x): 
    return x[0].isupper()

# %% dog_v_cat.ipynb 7
from fastai.vision.all import *
import gradio as gr

def is_cat(x): 
    return x[0].isupper()

# %% dog_v_cat.ipynb 8
from fastai.vision.all import *
import gradio as gr

def is_cat(x): 
    return x[0].isupper()

# %% dog_v_cat.ipynb 9
from fastai.vision.all import *
import gradio as gr

def is_cat(x): 
    return x[0].isupper()

# %% dog_v_cat.ipynb 10
from fastai.vision.all import *
import gradio as gr

def is_cat(x): 
    return x[0].isupper()

# %% dog_v_cat.ipynb 12
learn = load_learner('/kaggle/input/models/model.pkl')

# %% dog_v_cat.ipynb 14
categories = ('Dog', 'Cat')
def classify_image(img):
    pred, idx, probs = learn.predict(img)
    return dict(zip(categories, map(float, probs)))

# %% dog_v_cat.ipynb 16
image = gr.inputs.Image(shape=(192,192))
label = gr.outputs.Label()
examples = ['/kaggle/input/dog-or-cat-test/dog.jpg','/kaggle/input/dog-or-cat-test/cat.jpg', '/kaggle/input/dog-or-cat-test/challenge.jpg']

intf = gr.Interface(fn=classify_image, inputs=image, outputs=label, examples=examples)

# %% dog_v_cat.ipynb 17
image = gr.inputs.Image(shape=(192,192))
label = gr.outputs.Label()
examples = ['/kaggle/input/dog-or-cat-test/dog.jpg','/kaggle/input/dog-or-cat-test/cat.jpg', '/kaggle/input/dog-or-cat-test/challenge.jpg']

intf = gr.Interface(fn=classify_image, inputs=image, outputs=label, examples=examples)
intf.launch(inline=False)