图像分类模型嵌入flask中开发PythonWeb项目

from exts import db
from datetime import datetime
# 用户表，记录用户账号信息
class UserModel(db.Model):
    __tablename__ = 'user'
    id = db.Column(db.Integer, primary_key=True, autoincrement=True)
    username = db.Column(db.String(100), nullable=False)
    password = db.Column(db.String(200), nullable=False)
    email = db.Column(db.String(100), nullable=False, unique=True)  # 邮箱唯一
    join_time = db.Column(db.DateTime, default=datetime.now())
# 用户注册表，记录用户注册时的信息
class EmailCaptchaModel(db.Model):
    __tablename__ = 'email_captcha'
    id = db.Column(db.Integer, primary_key=True, autoincrement=True)
    email = db.Column(db.String(100), nullable=False)
    captcha = db.Column(db.String(100), nullable=False)
# 用户上传表，记录用户在网页中上传的图片
class ImageUploadModel(db.Model):
    __tablename__ = 'image_upload'
    id = db.Column(db.Integer, primary_key=True, autoincrement=True)
    user_id = db.Column(db.Integer, db.ForeignKey('user.id'), nullable=False)
    img_name = db.Column(db.String(100), nullable=False)
    img_format = db.Column(db.String(100), nullable=False)
    upload_time = db.Column(db.DateTime, default=datetime.now())
# 结果记录表，记录用户在网页中上传图片的结果
class ImageRecordModel(db.Model):
    __tablename__ = 'image_record'
    id = db.Column(db.Integer, primary_key=True, autoincrement=True)
    user_id = db.Column(db.Integer, db.ForeignKey('user.id'), nullable=False)
    img_id = db.Column(db.Integer, db.ForeignKey('image_upload.id'), nullable=False)
    img_acc = db.Column(db.String(50), nullable=False)
    img_class = db.Column(db.String(50), nullable=False)
    ident_time = db.Column(db.DateTime, default=datetime.now())
    record=db.relationship('UserModel', backref='user_record', uselist=False)

import json
import os
import numpy as np
import torch
from matplotlib import pyplot as plt
from torchvision import transforms
from PIL import Image
from model.mobilenetv2 import MobileNetV2 as create_model
from model.utils import GradCAM, show_cam_on_image, center_crop_img
def pred(img_path):
    device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
    print(f"using {device} device.")
    num_classes = 4
    img_size = 224
    data_transform = transforms.Compose(
        [transforms.Resize(int(img_size * 1.14)),
         transforms.CenterCrop(img_size),
         transforms.ToTensor(),
         transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])])
    img = Image.open(img_path)
    # [N, C, H, W]
    img = data_transform(img)
    # expand batch dimension
    img = torch.unsqueeze(img, dim=0)
    # read class_indict
    json_path = 'model/class_indices.json'
    assert os.path.exists(json_path), "file: '{}' dose not exist.".format(json_path)
    json_file = open(json_path, "r", encoding='utf-8')
    class_indict = json.load(json_file)
    # create model
    model = create_model(num_classes=num_classes).to(device)
    # load model weights
    model_weight_path = "model/bestmodel.pth"
    model.load_state_dict(torch.load(model_weight_path, map_location=device))
    model.eval()
    with torch.no_grad():
        # predict class
        output = torch.squeeze(model(img.to(device))).cpu()
        predict = torch.softmax(output, dim=0)
        predict_cla = torch.argmax(predict).numpy()
    # print_res = "class: {}   prob: {:.3}".format(class_indict[str(predict_cla)],
    #                                              predict[predict_cla].numpy())
    class_prob=predict[predict_cla].numpy()
    class_name=class_indict[str(predict_cla)]
    return class_prob,class_name
if __name__=="__main__":
    img_path = "static/img/1.jpg"
    class_prob,class_name= pred(img_path)
    print(class_name,class_prob)