app.py

import gradio as gr
import torch
from torch import nn
from pathlib import Path

LABELS = Path('class_names.txt').read_text().splitlines()
title = "Sketch Classifier"
description = "This machine has vision. To test the machine, if you sketch something below, the machine will attempt to classify your sketch.  It can guess from 100 objects and it will present to you its top three guesses. Beside each guess, the length of the bar indicates the confidence with which the machine has identified your sketch. The longer the bar, the more confident the machine is."

model = nn.Sequential(
    nn.Conv2d(1, 32, 3, padding='same'),
    nn.ReLU(),
    nn.MaxPool2d(2),
    nn.Conv2d(32, 64, 3, padding='same'),
    nn.ReLU(),
    nn.MaxPool2d(2),
    nn.Conv2d(64, 128, 3, padding='same'),
    nn.ReLU(),
    nn.MaxPool2d(2),
    nn.Flatten(),
    nn.Linear(1152, 256),
    nn.ReLU(),
    nn.Linear(256, len(LABELS)),
)
state_dict = torch.load('pytorch_model.bin',    map_location='cpu')
model.load_state_dict(state_dict, strict=False)
model.eval()

def predict(img):
    x = torch.tensor(img, dtype=torch.float32).unsqueeze(0).unsqueeze(0) / 255.
    with torch.no_grad():
        out = model(x)
    probabilities = torch.nn.functional.softmax(out[0], dim=0)
    values, indices = torch.topk(probabilities, 3)
    confidences = {LABELS[i]: v.item() for i, v in zip(indices, values)}
    return confidences

gr.Interface(fn=predict, 
             inputs="sketchpad",
             outputs="label",
             title=title,
             description=description,
             live=True).launch(debug=True)