onnx/ShuffleNet-v2

1import onnx2import numpy as np3import onnxruntime as ort4from PIL import Image5import cv26import os7import gradio as gr8 9import mxnet10from torchvision import transforms11 12os.system("wget https://s3.amazonaws.com/onnx-model-zoo/synset.txt")13 14 15with open('synset.txt', 'r') as f:16    labels = [l.rstrip() for l in f]17    18os.system("wget https://github.com/AK391/models/raw/main/vision/classification/shufflenet/model/shufflenet-v2-10.onnx")19 20os.system("wget https://s3.amazonaws.com/model-server/inputs/kitten.jpg")21 22 23 24model_path = 'shufflenet-v2-10.onnx'25model = onnx.load(model_path)26session = ort.InferenceSession(model.SerializeToString())27 28 29    30preprocess = transforms.Compose([31        transforms.Resize(256),32        transforms.CenterCrop(224),33        transforms.ToTensor(),34        transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),35])36        37 38    39 40def predict(img):41    input_tensor = preprocess(img)42    img = input_tensor.unsqueeze(0)43    ort_inputs = {session.get_inputs()[0].name: img.cpu().detach().numpy()}44    preds = session.run(None, ort_inputs)[0]45    preds = np.squeeze(preds)46    a = np.argsort(preds)47    results = {}48    for i in a[0:5]:    49        results[labels[a[i]]] = float(preds[a[i]])50    return results51       52 53title="ShuffleNet-v2"54description="ShuffleNet is a deep convolutional network for image classification. ShuffleNetV2 is an improved architecture that is the state-of-the-art in terms of speed and accuracy tradeoff used for image classification."55 56examples=[['kitten.jpg']]57gr.Interface(predict,gr.inputs.Image(type='pil'),"label",title=title,description=description,examples=examples).launch(enable_queue=True,debug=True)