基于MobileNetv2的垃圾分类函数式自动微分-昇思25天打卡

基于MobileNetv2的垃圾分类

本文档主要介绍垃圾分类代码开发的方法。通过读取本地图像数据作为输入,对图像中的垃圾物体进行检测,并且将检测结果图片保存到文件中。

1、实验目的

  • 了解熟悉垃圾分类应用代码的编写(Python语言);
  • 了解Linux操作系统的基本使用;
  • 掌握atc命令进行模型转换的基本操作。

2、MobileNetv2模型原理介绍

MobileNet网络是由Google团队于2017年提出的专注于移动端、嵌入式或IoT设备的轻量级CNN网络,相比于传统的卷积神经网络,MobileNet网络使用深度可分离卷积(Depthwise Separable Convolution)的思想在准确率小幅度降低的前提下,大大减小了模型参数与运算量。并引入宽度系数 α和分辨率系数 β使模型满足不同应用场景的需求。

由于MobileNet网络中Relu激活函数处理低维特征信息时会存在大量的丢失,所以MobileNetV2网络提出使用倒残差结构(Inverted residual block)和Linear Bottlenecks来设计网络,以提高模型的准确率,且优化后的模型更小。

请添加图片描述

图中Inverted residual block结构是先使用1x1卷积进行升维,然后使用3x3的DepthWise卷积,最后使用1x1的卷积进行降维,与Residual block结构相反。Residual block是先使用1x1的卷积进行降维,然后使用3x3的卷积,最后使用1x1的卷积进行升维。

  • 说明:
    详细内容可参见MobileNetV2论文

3、实验环境

本案例支持win_x86和Linux系统,CPU/GPU/Ascend均可运行。

在动手进行实践之前,确保您已经正确安装了MindSpore。不同平台下的环境准备请参考《MindSpore环境搭建实验手册》。

4、数据处理

4.1数据准备

MobileNetV2的代码默认使用ImageFolder格式管理数据集,每一类图片整理成单独的一个文件夹, 数据集结构如下:

└─ImageFolder

├─train
│   class1Folder
│   ......
└─evalclass1Folder......
%%capture captured_output
# 实验环境已经预装了mindspore==2.2.14,如需更换mindspore版本,可更改下面mindspore的版本号
!pip uninstall mindspore -y
!pip install -i https://pypi.mirrors.ustc.edu.cn/simple mindspore==2.2.14
# 查看当前 mindspore 版本
!pip show mindspore
Name: mindspore
Version: 2.2.14
Summary: MindSpore is a new open source deep learning training/inference framework that could be used for mobile, edge and cloud scenarios.
Home-page: https://www.mindspore.cn
Author: The MindSpore Authors
Author-email: contact@mindspore.cn
License: Apache 2.0
Location: /home/nginx/miniconda/envs/jupyter/lib/python3.9/site-packages
Requires: asttokens, astunparse, numpy, packaging, pillow, protobuf, psutil, scipy
Required-by: mindnlp
from download import download# 下载data_en数据集
url = "https://ascend-professional-construction-dataset.obs.cn-north-4.myhuaweicloud.com:443/MindStudio-pc/data_en.zip" 
path = download(url, "./", kind="zip", replace=True)
Downloading data from https://ascend-professional-construction-dataset.obs.cn-north-4.myhuaweicloud.com:443/MindStudio-pc/data_en.zip (21.3 MB)file_sizes: 100%|███████████████████████████| 22.4M/22.4M [00:00<00:00, 106MB/s]
Extracting zip file...
Successfully downloaded / unzipped to ./
from download import download# 下载预训练权重文件
url = "https://ascend-professional-construction-dataset.obs.cn-north-4.myhuaweicloud.com:443/ComputerVision/mobilenetV2-200_1067.zip" 
path = download(url, "./", kind="zip", replace=True)
Downloading data from https://ascend-professional-construction-dataset.obs.cn-north-4.myhuaweicloud.com:443/ComputerVision/mobilenetV2-200_1067.zip (25.5 MB)file_sizes: 100%|███████████████████████████| 26.7M/26.7M [00:00<00:00, 111MB/s]
Extracting zip file...
Successfully downloaded / unzipped to ./

4.2数据加载

将模块导入,具体如下:
import math
import numpy as np
import os
import randomfrom matplotlib import pyplot as plt
from easydict import EasyDict
from PIL import Image
import numpy as np
import mindspore.nn as nn
from mindspore import ops as P
from mindspore.ops import add
from mindspore import Tensor
import mindspore.common.dtype as mstype
import mindspore.dataset as de
import mindspore.dataset.vision as C
import mindspore.dataset.transforms as C2
import mindspore as ms
from mindspore import set_context, nn, Tensor, load_checkpoint, save_checkpoint, export
from mindspore.train import Model
from mindspore.train import Callback, LossMonitor, ModelCheckpoint, CheckpointConfigos.environ['GLOG_v'] = '3' # Log level includes 3(ERROR), 2(WARNING), 1(INFO), 0(DEBUG).
os.environ['GLOG_logtostderr'] = '0' # 0:输出到文件,1:输出到屏幕
os.environ['GLOG_log_dir'] = '../../log' # 日志目录
os.environ['GLOG_stderrthreshold'] = '2' # 输出到目录也输出到屏幕:3(ERROR), 2(WARNING), 1(INFO), 0(DEBUG).
set_context(mode=ms.GRAPH_MODE, device_target="CPU", device_id=0) # 设置采用图模式执行,设备为Ascend#
配置后续训练、验证、推理用到的参数:
# 垃圾分类数据集标签,以及用于标签映射的字典。
garbage_classes = {'干垃圾': ['贝壳', '打火机', '旧镜子', '扫把', '陶瓷碗', '牙刷', '一次性筷子', '脏污衣服'],'可回收物': ['报纸', '玻璃制品', '篮球', '塑料瓶', '硬纸板', '玻璃瓶', '金属制品', '帽子', '易拉罐', '纸张'],'湿垃圾': ['菜叶', '橙皮', '蛋壳', '香蕉皮'],'有害垃圾': ['电池', '药片胶囊', '荧光灯', '油漆桶']
}class_cn = ['贝壳', '打火机', '旧镜子', '扫把', '陶瓷碗', '牙刷', '一次性筷子', '脏污衣服','报纸', '玻璃制品', '篮球', '塑料瓶', '硬纸板', '玻璃瓶', '金属制品', '帽子', '易拉罐', '纸张','菜叶', '橙皮', '蛋壳', '香蕉皮','电池', '药片胶囊', '荧光灯', '油漆桶']
class_en = ['Seashell', 'Lighter','Old Mirror', 'Broom','Ceramic Bowl', 'Toothbrush','Disposable Chopsticks','Dirty Cloth','Newspaper', 'Glassware', 'Basketball', 'Plastic Bottle', 'Cardboard','Glass Bottle', 'Metalware', 'Hats', 'Cans', 'Paper','Vegetable Leaf','Orange Peel', 'Eggshell','Banana Peel','Battery', 'Tablet capsules','Fluorescent lamp', 'Paint bucket']index_en = {'Seashell': 0, 'Lighter': 1, 'Old Mirror': 2, 'Broom': 3, 'Ceramic Bowl': 4, 'Toothbrush': 5, 'Disposable Chopsticks': 6, 'Dirty Cloth': 7,'Newspaper': 8, 'Glassware': 9, 'Basketball': 10, 'Plastic Bottle': 11, 'Cardboard': 12, 'Glass Bottle': 13, 'Metalware': 14, 'Hats': 15, 'Cans': 16, 'Paper': 17,'Vegetable Leaf': 18, 'Orange Peel': 19, 'Eggshell': 20, 'Banana Peel': 21,'Battery': 22, 'Tablet capsules': 23, 'Fluorescent lamp': 24, 'Paint bucket': 25}# 训练超参
config = EasyDict({"num_classes": 26,"image_height": 224,"image_width": 224,#"data_split": [0.9, 0.1],"backbone_out_channels":1280,"batch_size": 16,"eval_batch_size": 8,"epochs": 10,"lr_max": 0.05,"momentum": 0.9,"weight_decay": 1e-4,"save_ckpt_epochs": 1,"dataset_path": "./data_en","class_index": index_en,"pretrained_ckpt": "./mobilenetV2-200_1067.ckpt" # mobilenetV2-200_1067.ckpt 
})
数据预处理操作

利用ImageFolderDataset方法读取垃圾分类数据集,并整体对数据集进行处理。

读取数据集时指定训练集和测试集,首先对整个数据集进行归一化,修改图像频道等预处理操作。然后对训练集的数据依次进行RandomCropDecodeResize、RandomHorizontalFlip、RandomColorAdjust、shuffle操作,以增加训练数据的丰富度;对测试集进行Decode、Resize、CenterCrop等预处理操作;最后返回处理后的数据集。

def create_dataset(dataset_path, config, training=True, buffer_size=1000):"""create a train or eval datasetArgs:dataset_path(string): the path of dataset.config(struct): the config of train and eval in diffirent platform.Returns:train_dataset, val_dataset"""data_path = os.path.join(dataset_path, 'train' if training else 'test')ds = de.ImageFolderDataset(data_path, num_parallel_workers=4, class_indexing=config.class_index)resize_height = config.image_heightresize_width = config.image_widthnormalize_op = C.Normalize(mean=[0.485*255, 0.456*255, 0.406*255], std=[0.229*255, 0.224*255, 0.225*255])change_swap_op = C.HWC2CHW()type_cast_op = C2.TypeCast(mstype.int32)if training:crop_decode_resize = C.RandomCropDecodeResize(resize_height, scale=(0.08, 1.0), ratio=(0.75, 1.333))horizontal_flip_op = C.RandomHorizontalFlip(prob=0.5)color_adjust = C.RandomColorAdjust(brightness=0.4, contrast=0.4, saturation=0.4)train_trans = [crop_decode_resize, horizontal_flip_op, color_adjust, normalize_op, change_swap_op]train_ds = ds.map(input_columns="image", operations=train_trans, num_parallel_workers=4)train_ds = train_ds.map(input_columns="label", operations=type_cast_op, num_parallel_workers=4)train_ds = train_ds.shuffle(buffer_size=buffer_size)ds = train_ds.batch(config.batch_size, drop_remainder=True)else:decode_op = C.Decode()resize_op = C.Resize((int(resize_width/0.875), int(resize_width/0.875)))center_crop = C.CenterCrop(resize_width)eval_trans = [decode_op, resize_op, center_crop, normalize_op, change_swap_op]eval_ds = ds.map(input_columns="image", operations=eval_trans, num_parallel_workers=4)eval_ds = eval_ds.map(input_columns="label", operations=type_cast_op, num_parallel_workers=4)ds = eval_ds.batch(config.eval_batch_size, drop_remainder=True)return ds
展示部分处理后的数据:
ds = create_dataset(dataset_path=config.dataset_path, config=config, training=False)
print(ds.get_dataset_size())
data = ds.create_dict_iterator(output_numpy=True)._get_next()
images = data['image']
labels = data['label']for i in range(1, 5):plt.subplot(2, 2, i)plt.imshow(np.transpose(images[i], (1,2,0)))plt.title('label: %s' % class_en[labels[i]])plt.xticks([])
plt.show()
Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers). Got range [-1.7240347..2.64].
Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers). Got range [-2.117904..2.64].
Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers). Got range [-2.117904..2.64].
Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers). Got range [-2.0357144..2.64].32

请添加图片描述

5、MobileNetV2模型搭建

使用MindSpore定义MobileNetV2网络的各模块时需要继承mindspore.nn.Cell。Cell是所有神经网络(Conv2d等)的基类。

神经网络的各层需要预先在__init__方法中定义,然后通过定义construct方法来完成神经网络的前向构造。原始模型激活函数为ReLU6,池化模块采用是全局平均池化层。

__all__ = ['MobileNetV2', 'MobileNetV2Backbone', 'MobileNetV2Head', 'mobilenet_v2']def _make_divisible(v, divisor, min_value=None):if min_value is None:min_value = divisornew_v = max(min_value, int(v + divisor / 2) // divisor * divisor)if new_v < 0.9 * v:new_v += divisorreturn new_vclass GlobalAvgPooling(nn.Cell):"""Global avg pooling definition.Args:Returns:Tensor, output tensor.Examples:>>> GlobalAvgPooling()"""def __init__(self):super(GlobalAvgPooling, self).__init__()def construct(self, x):x = P.mean(x, (2, 3))return xclass ConvBNReLU(nn.Cell):"""Convolution/Depthwise fused with Batchnorm and ReLU block definition.Args:in_planes (int): Input channel.out_planes (int): Output channel.kernel_size (int): Input kernel size.stride (int): Stride size for the first convolutional layer. Default: 1.groups (int): channel group. Convolution is 1 while Depthiwse is input channel. Default: 1.Returns:Tensor, output tensor.Examples:>>> ConvBNReLU(16, 256, kernel_size=1, stride=1, groups=1)"""def __init__(self, in_planes, out_planes, kernel_size=3, stride=1, groups=1):super(ConvBNReLU, self).__init__()padding = (kernel_size - 1) // 2in_channels = in_planesout_channels = out_planesif groups == 1:conv = nn.Conv2d(in_channels, out_channels, kernel_size, stride, pad_mode='pad', padding=padding)else:out_channels = in_planesconv = nn.Conv2d(in_channels, out_channels, kernel_size, stride, pad_mode='pad',padding=padding, group=in_channels)layers = [conv, nn.BatchNorm2d(out_planes), nn.ReLU6()]self.features = nn.SequentialCell(layers)def construct(self, x):output = self.features(x)return outputclass InvertedResidual(nn.Cell):"""Mobilenetv2 residual block definition.Args:inp (int): Input channel.oup (int): Output channel.stride (int): Stride size for the first convolutional layer. Default: 1.expand_ratio (int): expand ration of input channelReturns:Tensor, output tensor.Examples:>>> ResidualBlock(3, 256, 1, 1)"""def __init__(self, inp, oup, stride, expand_ratio):super(InvertedResidual, self).__init__()assert stride in [1, 2]hidden_dim = int(round(inp * expand_ratio))self.use_res_connect = stride == 1 and inp == ouplayers = []if expand_ratio != 1:layers.append(ConvBNReLU(inp, hidden_dim, kernel_size=1))layers.extend([ConvBNReLU(hidden_dim, hidden_dim,stride=stride, groups=hidden_dim),nn.Conv2d(hidden_dim, oup, kernel_size=1,stride=1, has_bias=False),nn.BatchNorm2d(oup),])self.conv = nn.SequentialCell(layers)self.cast = P.Cast()def construct(self, x):identity = xx = self.conv(x)if self.use_res_connect:return P.add(identity, x)return xclass MobileNetV2Backbone(nn.Cell):"""MobileNetV2 architecture.Args:class_num (int): number of classes.width_mult (int): Channels multiplier for round to 8/16 and others. Default is 1.has_dropout (bool): Is dropout used. Default is falseinverted_residual_setting (list): Inverted residual settings. Default is Noneround_nearest (list): Channel round to . Default is 8Returns:Tensor, output tensor.Examples:>>> MobileNetV2(num_classes=1000)"""def __init__(self, width_mult=1., inverted_residual_setting=None, round_nearest=8,input_channel=32, last_channel=1280):super(MobileNetV2Backbone, self).__init__()block = InvertedResidual# setting of inverted residual blocksself.cfgs = inverted_residual_settingif inverted_residual_setting is None:self.cfgs = [# t, c, n, s[1, 16, 1, 1],[6, 24, 2, 2],[6, 32, 3, 2],[6, 64, 4, 2],[6, 96, 3, 1],[6, 160, 3, 2],[6, 320, 1, 1],]# building first layerinput_channel = _make_divisible(input_channel * width_mult, round_nearest)self.out_channels = _make_divisible(last_channel * max(1.0, width_mult), round_nearest)features = [ConvBNReLU(3, input_channel, stride=2)]# building inverted residual blocksfor t, c, n, s in self.cfgs:output_channel = _make_divisible(c * width_mult, round_nearest)for i in range(n):stride = s if i == 0 else 1features.append(block(input_channel, output_channel, stride, expand_ratio=t))input_channel = output_channelfeatures.append(ConvBNReLU(input_channel, self.out_channels, kernel_size=1))self.features = nn.SequentialCell(features)self._initialize_weights()def construct(self, x):x = self.features(x)return xdef _initialize_weights(self):"""Initialize weights.Args:Returns:None.Examples:>>> _initialize_weights()"""self.init_parameters_data()for _, m in self.cells_and_names():if isinstance(m, nn.Conv2d):n = m.kernel_size[0] * m.kernel_size[1] * m.out_channelsm.weight.set_data(Tensor(np.random.normal(0, np.sqrt(2. / n),m.weight.data.shape).astype("float32")))if m.bias is not None:m.bias.set_data(Tensor(np.zeros(m.bias.data.shape, dtype="float32")))elif isinstance(m, nn.BatchNorm2d):m.gamma.set_data(Tensor(np.ones(m.gamma.data.shape, dtype="float32")))m.beta.set_data(Tensor(np.zeros(m.beta.data.shape, dtype="float32")))@propertydef get_features(self):return self.featuresclass MobileNetV2Head(nn.Cell):"""MobileNetV2 architecture.Args:class_num (int): Number of classes. Default is 1000.has_dropout (bool): Is dropout used. Default is falseReturns:Tensor, output tensor.Examples:>>> MobileNetV2(num_classes=1000)"""def __init__(self, input_channel=1280, num_classes=1000, has_dropout=False, activation="None"):super(MobileNetV2Head, self).__init__()# mobilenet headhead = ([GlobalAvgPooling(), nn.Dense(input_channel, num_classes, has_bias=True)] if not has_dropout else[GlobalAvgPooling(), nn.Dropout(0.2), nn.Dense(input_channel, num_classes, has_bias=True)])self.head = nn.SequentialCell(head)self.need_activation = Trueif activation == "Sigmoid":self.activation = nn.Sigmoid()elif activation == "Softmax":self.activation = nn.Softmax()else:self.need_activation = Falseself._initialize_weights()def construct(self, x):x = self.head(x)if self.need_activation:x = self.activation(x)return xdef _initialize_weights(self):"""Initialize weights.Args:Returns:None.Examples:>>> _initialize_weights()"""self.init_parameters_data()for _, m in self.cells_and_names():if isinstance(m, nn.Dense):m.weight.set_data(Tensor(np.random.normal(0, 0.01, m.weight.data.shape).astype("float32")))if m.bias is not None:m.bias.set_data(Tensor(np.zeros(m.bias.data.shape, dtype="float32")))@propertydef get_head(self):return self.headclass MobileNetV2(nn.Cell):"""MobileNetV2 architecture.Args:class_num (int): number of classes.width_mult (int): Channels multiplier for round to 8/16 and others. Default is 1.has_dropout (bool): Is dropout used. Default is falseinverted_residual_setting (list): Inverted residual settings. Default is Noneround_nearest (list): Channel round to . Default is 8Returns:Tensor, output tensor.Examples:>>> MobileNetV2(backbone, head)"""def __init__(self, num_classes=1000, width_mult=1., has_dropout=False, inverted_residual_setting=None, \round_nearest=8, input_channel=32, last_channel=1280):super(MobileNetV2, self).__init__()self.backbone = MobileNetV2Backbone(width_mult=width_mult, \inverted_residual_setting=inverted_residual_setting, \round_nearest=round_nearest, input_channel=input_channel, last_channel=last_channel).get_featuresself.head = MobileNetV2Head(input_channel=self.backbone.out_channel, num_classes=num_classes, \has_dropout=has_dropout).get_headdef construct(self, x):x = self.backbone(x)x = self.head(x)return xclass MobileNetV2Combine(nn.Cell):"""MobileNetV2Combine architecture.Args:backbone (Cell): the features extract layers.head (Cell):  the fully connected layers.Returns:Tensor, output tensor.Examples:>>> MobileNetV2(num_classes=1000)"""def __init__(self, backbone, head):super(MobileNetV2Combine, self).__init__(auto_prefix=False)self.backbone = backboneself.head = headdef construct(self, x):x = self.backbone(x)x = self.head(x)return xdef mobilenet_v2(backbone, head):return MobileNetV2Combine(backbone, head)

6、MobileNetV2模型的训练与测试

训练策略

一般情况下,模型训练时采用静态学习率,如0.01。随着训练步数的增加,模型逐渐趋于收敛,对权重参数的更新幅度应该逐渐降低,以减小模型训练后期的抖动。所以,模型训练时可以采用动态下降的学习率,常见的学习率下降策略有:

  • polynomial decay/square decay;
  • cosine decay;
  • exponential decay;
  • stage decay.

这里使用cosine decay下降策略:

def cosine_decay(total_steps, lr_init=0.0, lr_end=0.0, lr_max=0.1, warmup_steps=0):"""Applies cosine decay to generate learning rate array.Args:total_steps(int): all steps in training.lr_init(float): init learning rate.lr_end(float): end learning ratelr_max(float): max learning rate.warmup_steps(int): all steps in warmup epochs.Returns:list, learning rate array."""lr_init, lr_end, lr_max = float(lr_init), float(lr_end), float(lr_max)decay_steps = total_steps - warmup_stepslr_all_steps = []inc_per_step = (lr_max - lr_init) / warmup_steps if warmup_steps else 0for i in range(total_steps):if i < warmup_steps:lr = lr_init + inc_per_step * (i + 1)else:cosine_decay = 0.5 * (1 + math.cos(math.pi * (i - warmup_steps) / decay_steps))lr = (lr_max - lr_end) * cosine_decay + lr_endlr_all_steps.append(lr)return lr_all_steps

在模型训练过程中,可以添加检查点(Checkpoint)用于保存模型的参数,以便进行推理及中断后再训练使用。使用场景如下:

  • 训练后推理场景
  1. 模型训练完毕后保存模型的参数,用于推理或预测操作。
  2. 训练过程中,通过实时验证精度,把精度最高的模型参数保存下来,用于预测操作。
  • 再训练场景
  1. 进行长时间训练任务时,保存训练过程中的Checkpoint文件,防止任务异常退出后从初始状态开始训练。
  2. Fine-tuning(微调)场景,即训练一个模型并保存参数,基于该模型,面向第二个类似任务进行模型训练。

这里加载ImageNet数据上预训练的MobileNetv2进行Fine-tuning,只训练最后修改的FC层,并在训练过程中保存Checkpoint。

def switch_precision(net, data_type):if ms.get_context('device_target') == "Ascend":net.to_float(data_type)for _, cell in net.cells_and_names():if isinstance(cell, nn.Dense):cell.to_float(ms.float32)
模型训练与测试

在进行正式的训练之前,定义训练函数,读取数据并对模型进行实例化,定义优化器和损失函数。

首先简单介绍损失函数及优化器的概念:

  • 损失函数:又叫目标函数,用于衡量预测值与实际值差异的程度。深度学习通过不停地迭代来缩小损失函数的值。定义一个好的损失函数,可以有效提高模型的性能。

  • 优化器:用于最小化损失函数,从而在训练过程中改进模型。

定义了损失函数后,可以得到损失函数关于权重的梯度。梯度用于指示优化器优化权重的方向,以提高模型性能。

在训练MobileNetV2之前对MobileNetV2Backbone层的参数进行了固定,使其在训练过程中对该模块的权重参数不进行更新;只对MobileNetV2Head模块的参数进行更新。

MindSpore支持的损失函数有SoftmaxCrossEntropyWithLogits、L1Loss、MSELoss等。这里使用SoftmaxCrossEntropyWithLogits损失函数。

训练测试过程中会打印loss值,loss值会波动,但总体来说loss值会逐步减小,精度逐步提高。每个人运行的loss值有一定随机性,不一定完全相同。

每打印一个epoch后模型都会在测试集上的计算测试精度,从打印的精度值分析MobileNetV2模型的预测能力在不断提升。

from mindspore.amp import FixedLossScaleManager
import time
LOSS_SCALE = 1024train_dataset = create_dataset(dataset_path=config.dataset_path, config=config)
eval_dataset = create_dataset(dataset_path=config.dataset_path, config=config)
step_size = train_dataset.get_dataset_size()backbone = MobileNetV2Backbone() #last_channel=config.backbone_out_channels
# Freeze parameters of backbone. You can comment these two lines.
for param in backbone.get_parameters():param.requires_grad = False
# load parameters from pretrained model
load_checkpoint(config.pretrained_ckpt, backbone)head = MobileNetV2Head(input_channel=backbone.out_channels, num_classes=config.num_classes)
network = mobilenet_v2(backbone, head)# define loss, optimizer, and model
loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True, reduction='mean')
loss_scale = FixedLossScaleManager(LOSS_SCALE, drop_overflow_update=False)
lrs = cosine_decay(config.epochs * step_size, lr_max=config.lr_max)
opt = nn.Momentum(network.trainable_params(), lrs, config.momentum, config.weight_decay, loss_scale=LOSS_SCALE)# 定义用于训练的train_loop函数。
def train_loop(model, dataset, loss_fn, optimizer):# 定义正向计算函数def forward_fn(data, label):logits = model(data)loss = loss_fn(logits, label)return loss# 定义微分函数,使用mindspore.value_and_grad获得微分函数grad_fn,输出loss和梯度。# 由于是对模型参数求导,grad_position 配置为None,传入可训练参数。grad_fn = ms.value_and_grad(forward_fn, None, optimizer.parameters)# 定义 one-step training函数def train_step(data, label):loss, grads = grad_fn(data, label)optimizer(grads)return losssize = dataset.get_dataset_size()model.set_train()for batch, (data, label) in enumerate(dataset.create_tuple_iterator()):loss = train_step(data, label)if batch % 10 == 0:loss, current = loss.asnumpy(), batchprint(f"loss: {loss:>7f}  [{current:>3d}/{size:>3d}]")# 定义用于测试的test_loop函数。
def test_loop(model, dataset, loss_fn):num_batches = dataset.get_dataset_size()model.set_train(False)total, test_loss, correct = 0, 0, 0for data, label in dataset.create_tuple_iterator():pred = model(data)total += len(data)test_loss += loss_fn(pred, label).asnumpy()correct += (pred.argmax(1) == label).asnumpy().sum()test_loss /= num_batchescorrect /= totalprint(f"Test: \n Accuracy: {(100*correct):>0.1f}%, Avg loss: {test_loss:>8f} \n")print("============== Starting Training ==============")
# 由于时间问题,训练过程只进行了2个epoch ,可以根据需求调整。
epoch_begin_time = time.time()
epochs = 2
for t in range(epochs):begin_time = time.time()print(f"Epoch {t+1}\n-------------------------------")train_loop(network, train_dataset, loss, opt)ms.save_checkpoint(network, "save_mobilenetV2_model.ckpt")end_time = time.time()times = end_time - begin_timeprint(f"per epoch time: {times}s")test_loop(network, eval_dataset, loss)
epoch_end_time = time.time()
times = epoch_end_time - epoch_begin_time
print(f"total time:  {times}s")
print("============== Training Success ==============")
============== Starting Training ==============
Epoch 1
-------------------------------loss: 3.254856  [  0/162]
loss: 3.217916  [ 10/162]
loss: 3.186504  [ 20/162]
loss: 3.355182  [ 30/162]
loss: 3.336528  [ 40/162]
loss: 3.220652  [ 50/162]
loss: 3.253688  [ 60/162]
loss: 3.208350  [ 70/162]
loss: 3.215430  [ 80/162]
loss: 3.244096  [ 90/162]
loss: 3.188400  [100/162]
loss: 3.155391  [110/162]
loss: 3.304114  [120/162]
loss: 3.239458  [130/162]
loss: 3.225820  [140/162]
loss: 3.191090  [150/162]
loss: 3.188383  [160/162]
per epoch time: 70.34860825538635sTest: Accuracy: 11.4%, Avg loss: 3.189114 Epoch 2
-------------------------------loss: 3.156094  [  0/162]
loss: 3.262588  [ 10/162]
loss: 3.141063  [ 20/162]
loss: 3.195691  [ 30/162]
loss: 3.253536  [ 40/162]
loss: 3.236728  [ 50/162]
loss: 3.122650  [ 60/162]
loss: 3.165659  [ 70/162]
loss: 3.144341  [ 80/162]
loss: 3.140723  [ 90/162]
loss: 3.190994  [100/162]
loss: 3.177974  [110/162]
loss: 3.111208  [120/162]
loss: 3.126922  [130/162]
loss: 3.111721  [140/162]
loss: 3.135597  [150/162]
loss: 3.113985  [160/162]
per epoch time: 78.56036615371704s
Test: Accuracy: 20.6%, Avg loss: 3.109391 total time:  287.22266936302185s
============== Training Success ==============

7、模型推理

加载模型Checkpoint进行推理,使用load_checkpoint接口加载数据时,需要把数据传入给原始网络,而不能传递给带有优化器和损失函数的训练网络。

CKPT="save_mobilenetV2_model.ckpt"
def image_process(image):"""Precess one image per time.Args:image: shape (H, W, C)"""mean=[0.485*255, 0.456*255, 0.406*255]std=[0.229*255, 0.224*255, 0.225*255]image = (np.array(image) - mean) / stdimage = image.transpose((2,0,1))img_tensor = Tensor(np.array([image], np.float32))return img_tensordef infer_one(network, image_path):image = Image.open(image_path).resize((config.image_height, config.image_width))logits = network(image_process(image))pred = np.argmax(logits.asnumpy(), axis=1)[0]print(image_path, class_en[pred])def infer():backbone = MobileNetV2Backbone(last_channel=config.backbone_out_channels)head = MobileNetV2Head(input_channel=backbone.out_channels, num_classes=config.num_classes)network = mobilenet_v2(backbone, head)load_checkpoint(CKPT, network)for i in range(91, 100):infer_one(network, f'data_en/test/Cardboard/000{i}.jpg')
infer()
data_en/test/Cardboard/00091.jpg Old Mirror
data_en/test/Cardboard/00092.jpg Cardboard
data_en/test/Cardboard/00093.jpg Lighter
data_en/test/Cardboard/00094.jpg Glass Bottle
data_en/test/Cardboard/00095.jpg Lighter
data_en/test/Cardboard/00096.jpg Tablet capsules
data_en/test/Cardboard/00097.jpg Broom
data_en/test/Cardboard/00098.jpg Cardboard
data_en/test/Cardboard/00099.jpg Plastic Bottle

8、导出AIR/GEIR/ONNX模型文件

导出AIR模型文件,用于后续Atlas 200 DK上的模型转换与推理。当前仅支持MindSpore+Ascend环境。

backbone = MobileNetV2Backbone(last_channel=config.backbone_out_channels)
head = MobileNetV2Head(input_channel=backbone.out_channels, num_classes=config.num_classes)
network = mobilenet_v2(backbone, head)
load_checkpoint(CKPT, network)input = np.random.uniform(0.0, 1.0, size=[1, 3, 224, 224]).astype(np.float32)
# export(network, Tensor(input), file_name='mobilenetv2.air', file_format='AIR')
# export(network, Tensor(input), file_name='mobilenetv2.pb', file_format='GEIR')
export(network, Tensor(input), file_name='mobilenetv2.onnx', file_format='ONNX')
from datetime import datetime
import pytz
# 设置时区为北京时区
beijing_tz = pytz.timezone('Asia/shanghai')
# 获取当前时间,并转为北京时间
current_beijing_time = datetime.now(beijing_tz)
# 格式化时间输出
formatted_time = current_beijing_time.strftime('%Y-%m-%d %H:%M:%S')
print("当前北京时间:",formatted_time)
print('用户名:matpandas 显似')
当前北京时间: 2024-07-23 23:00:05
用户名:matpandas 显似

本文来自互联网用户投稿,该文观点仅代表作者本人,不代表本站立场。本站仅提供信息存储空间服务,不拥有所有权,不承担相关法律责任。如若转载,请注明出处:http://www.rhkb.cn/news/382260.html

如若内容造成侵权/违法违规/事实不符,请联系长河编程网进行投诉反馈email:809451989@qq.com,一经查实,立即删除!

相关文章

linux 网络子系统

__netif_receive_skb_core 是 Linux 内核网络子系统中一个非常重要的函数&#xff0c;它负责将网络设备驱动层接收到的数据包传递到上层协议栈进行处理。以下是对该函数的一些关键点的详细解析&#xff1a; 一、函数作用 __netif_receive_skb_core 函数是处理接收到的网络数据…

linux 解决端口占用

1.查询被占用的端口 netstat -tln | grep 60602.查询该端口对应的服务 lsof -i :60603.杀死该进程 //14868是第二步的PID kill -9 14868

ubuntu在命令行输出里查找内容,dmesg

直接执行查看日志指令会出来很多页。dmesg为开机日志信息。记录了开机时硬件的过程 sudo dmesg 执行结果&#xff1a; 可以用竖号“|”&#xff0c;在前一条命令返回的内容进行查找。下图为查找bluetooth sudo dmesg |grep -i bluetooth

算法-嵌套类递归解题套路

文章目录 理论基础 :1. 基本计算器2. 字符串解码3. 求原子数量 理论基础 : 嵌套类递归是指一种一个字符串形式的问题通过嵌套调用子函数从而求解出结果的一类问题, 解题方法相对来说比较的固定, 我们总结为下面的几部分 大概过程 : 定义全局变量where递归函数 f ( i ) : s [ i …

【C++】——初识模版

文章目录 前言函数模版函数模版的原理函数模版的实例化 类模版类模版的实例化 前言 当我们使用一个通用的函数&#xff1a; //为每一个类型都编写一个重载版本 void Swap(int& left, int& right) {int temp left;left right;right temp; } void Swap(double& …

C# 与C++ cli

cli CLI&#xff08;Command Line Interface&#xff09;是一种通过命令行界面与计算机系统进行交互的方式。它提供了一种以文本形式输入命令和接收系统输出的方法&#xff0c;用于执行各种操作和管理计算机系统。以下是CLI的详细解释&#xff1a; 一、定义与基本概念 定义&…

编程中的智慧四:设计模式总览

前面三篇我们通过从一些零散的例子&#xff0c;和简单应用来模糊的感受了下设计模式在编程中的智慧&#xff0c;从现在开始正式进入设计模式介绍&#xff0c;本篇将从设计模式的7大原则、设计模式的三大类型、与23种设计模式的进行总结&#xff0c;和描述具体意义。 设计模式体…

【中项】系统集成项目管理工程师-第4章 信息系统架构-4.5技术架构

前言&#xff1a;系统集成项目管理工程师专业&#xff0c;现分享一些教材知识点。觉得文章还不错的喜欢点赞收藏的同时帮忙点点关注。 软考同样是国家人社部和工信部组织的国家级考试&#xff0c;全称为“全国计算机与软件专业技术资格&#xff08;水平&#xff09;考试”&…

关卡1-2:Python关卡

关卡1-2&#xff1a;Python关卡 一、python实现wordcount二、通过本地VSCODE连接internStudio与debug2.1使用本地的VSCODE连接InternStudio2.2 debug插件安装2.3 debug进行时2.3.1 代码准备2.3.2 选择python解释器2.3.3 打断点 一、python实现wordcount 采用python实现经典任务…

虚拟机迁移报错:虚拟机版本与主机“x.x.x.x”的版本不兼容

1.虚拟机在VCenter上从一个ESXi迁移到另一个ESXi上时报错&#xff1a;虚拟机版本与主机“x.x.x.x”的版本不兼容。 2.例如从10.0.128.13的ESXi上迁移到10.0.128.11的ESXi上。点击10.0.128.10上的任意一台虚拟机&#xff0c;查看虚拟机版本。 3.确认要迁移的虚拟机磁盘所在位…

大厂面试-基本功

大厂面试第4季 服务可用性多少个9是什么意思遍历集合add或remove操作bughashcode冲突案例BigdecimalList去重复IDEA Debugger测试框架ThreaLocal父子线程数据同步 InheritableThreadLocal完美解决线程数据同步方案 TransmittableThreadLocal 服务可用性多少个9是什么意思 遍历集…

Android中systrace配置及注意问题

Android中systrace配置及注意问题 systrace配置的官方文档地址如下&#xff1a;优化启动时间 Systrace systrace 允许在启动期间收集内核和 Android 跟踪记录。systrace 的可视化可以帮助分析启动过程中的具体问题。&#xff08;不过&#xff0c;如果要查看整个启动过程中的平…

[Spring] Spring配置文件

&#x1f338;个人主页:https://blog.csdn.net/2301_80050796?spm1000.2115.3001.5343 &#x1f3f5;️热门专栏: &#x1f9ca; Java基本语法(97平均质量分)https://blog.csdn.net/2301_80050796/category_12615970.html?spm1001.2014.3001.5482 &#x1f355; Collection与…

DocRED数据集

DocRED数据集文件夹包含多个JSON文件&#xff0c;每个文件都有不同的用途。以下是这些文件的用途解释以及哪个文件是训练集&#xff1a; 文件解释 dev.json&#xff1a;包含开发集&#xff08;验证集&#xff09;的数据&#xff0c;通常用于模型调优和选择超参数。 label_map…

Java | Leetcode Java题解之第260题只出现一次的数字III

题目&#xff1a; 题解&#xff1a; class Solution {public int[] singleNumber(int[] nums) {int xorsum 0;for (int num : nums) {xorsum ^ num;}// 防止溢出int lsb (xorsum Integer.MIN_VALUE ? xorsum : xorsum & (-xorsum));int type1 0, type2 0;for (int n…

Java 中的异常

异常&#xff1a;就是出现的问题。 在Java中异常被当成对象进行处理&#xff0c;所有的异常类都继承于Throwable类&#xff0c;如果Java提供的异常类并不能满足需求&#xff0c;用户还可以自己定义一个异常类。 下面是异常体系结构&#xff1a; Throwable又分成了Error和Exce…

PHP框架详解- symfony框架

文心一言 Symfony框架是一个用PHP语言编写的开放源代码的Web应用框架&#xff0c;旨在加速Web应用程序的开发过程&#xff0c;提高代码的可维护性和可扩展性。以下是对Symfony框架的详细解析&#xff1a; 一、框架概述 起源与开发者&#xff1a; Symfony由SensioLabs&#…

音乐曲谱软件Guitar Pro 8.2 for Mac 中文破解版

Guitar Pro 8.2 for Mac 中文破解版是一款功能强大的音乐曲谱软件&#xff0c;非常适合学习如何玩&#xff0c;改进技巧&#xff0c;重现喜爱的歌曲或陪伴自己。 Guitar Pro for Mac 是一款功能强大的音乐曲谱软件&#xff0c;非常适合学习如何玩&#xff0c;改进技巧&#xf…

宠物经济纵深观察:口红效应显著,呈可持续发展态势

七月以来&#xff0c;全国各地陆续开启高温模式。和人一样&#xff0c;“毛孩子们”同样也难耐高温&#xff0c;由此&#xff0c;围绕猫猫狗狗的“宠物经济”迅速升温&#xff0c;宠物冰垫、宠物饮水机、宠物烘干机......一系列宠物单品掀起夏日消费热潮。 就在几天前&#xf…

Pytorch学习笔记day4——训练mnist数据集和初步研读

该来的还是来了hhhhhhhhhh&#xff0c;基本上机器学习的初学者都躲不开这个例子。开源&#xff0c;数据质量高&#xff0c;数据尺寸整齐&#xff0c;问题简单&#xff0c;实在太适合初学者食用了。 今天把代码跑通&#xff0c;趁着周末好好的琢磨一下里面的各种细节。 代码实…