2）统一的中间表示（Intermediate Representation，IR）的保存，MindIR同时保存了Checkpoint和模型结构，因此需要定义输入Tensor来获取输入shape。mindspore.export(model, inputs, file_name="model", file_format="MINDIR")

6 模型推理

- 两种加载方式：

1）模型参数加载：

> model = network()

> param_dict = mindspore.load_checkpoint("model.ckpt");

> param_not_load, _ = mindspore.load_param_into_net(model, param_dict)

2）统一的中间表示（Intermediate Representation，IR）的加载：

> mindspore.set_context(mode=mindspore.GRAPH_MODE)
> graph = mindspore.load("model.mindir")
> model = nn.GraphCell(graph)  ## nn.GraphCell 仅支持图模式。
> outputs = model(inputs)

保存与加载 — MindSpore master 文档

代码

安装

pip/conda均可：

pip install -i https://pypi.mirrors.ustc.edu.cn/simple mindspore==2.3.0rc1

从模型训练到预测推理

训练：

python self_main_train_and_save.py

推理：

python self_predict.py

self_main_train_and_save.py


import mindspore
from mindspore import nn
from mindspore.dataset import vision, transforms
from mindspore.dataset import MnistDataset
 
# 用download库从公开华为云obs桶下载 MINIST 数据集并解压。因为mindspore.dataset 提供的接口仅支持解压后的数据文件 
from download import download
url = "https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/notebook/datasets/MNIST_Data.zip"
path = download(url, "./", kind="zip", replace=True)    
 
## 1 加载数据集
train_dataset = MnistDataset('MNIST_Data/train', shuffle=False)
test_dataset = MnistDataset('MNIST_Data/test')
print(train_dataset.get_col_names())   # 打印数据集中包含的数据列名，用于dataset的预处理。输出['image', 'label']
 
 
## 2 MindSpore的dataset使用数据处理流水线，这里将处理好的数据集打包为大小为64的batch。
from self_dataprocess import datapipe
# Map vision transforms and batch dataset
train_dataset = datapipe(train_dataset, 64)  
test_dataset = datapipe(test_dataset, 64)  
 
## 3 数据集加载后，一般以迭代方式获取数据，然后送入神经网络中进行训练。可使用create_tuple_iterator 或create_dict_iterator对数据集进行迭代访问，查看数据和标签的shape和datatype。
for image, label in test_dataset.create_tuple_iterator():
    print(f"Shape of image [N, C, H, W]: {image.shape} {image.dtype}")
    print(f"Shape of label: {label.shape} {label.dtype}")
    break
    “”“
    Shape of image [N, C, H, W]: (64, 1, 28, 28) Float32
    Shape of label: (64,) Int32
    ”“”
for data in test_dataset.create_dict_iterator():
    print(f"Shape of image [N, C, H, W]: {data['image'].shape} {data['image'].dtype}")
    print(f"Shape of label: {data['label'].shape} {data['label'].dtype}")
    break
 
 
## 4 模型训练
from self_network import Network
from self_modeltrain import train, loss_fn 
from self_modelteset import test
model = Network()
epochs = 3
for t in range(epochs):
    print(f"Epoch {t+1}\n-------------------------------")
    train(model, train_dataset)
    test(model, test_dataset, loss_fn)
print("Done!")
 
 
## 5 保存模型
# Save checkpoint
mindspore.save_checkpoint(model, "model.ckpt")
print("Saved Model to model.ckpt")

`self_dataprocess.py`


from mindspore.dataset import vision, transforms
def datapipe(dataset, batch_size):
    image_transforms = [
        vision.Rescale(1.0 / 255.0, 0),
        vision.Normalize(mean=(0.1307,), std=(0.3081,)),
        vision.HWC2CHW()
    ]
    label_transform = transforms.TypeCast(mindspore.int32)
    dataset = dataset.map(image_transforms, 'image')
    dataset = dataset.map(label_transform, 'label')
    dataset = dataset.batch(batch_size)
    return dataset

self_network.py


# Define model
from mindspore import nn
 
class Network(nn.Cell): 
    def __init__(self):
        super().__init__()
        self.flatten = nn.Flatten()
        self.dense_relu_sequential = nn.SequentialCell(
            nn.Dense(28*28, 512),
            nn.ReLU(),
            nn.Dense(512, 512),
            nn.ReLU(),
            nn.Dense(512, 10)
        )
    def construct(self, x):
        x = self.flatten(x)
        logits = self.dense_relu_sequential(x)
        return logits
 
 
def check_network():
    model = Network()
    print(model)

self_modeltrain.py


# Instantiate loss function and optimizer
from mindspore import nn
 
loss_fn = nn.CrossEntropyLoss()
optimizer = nn.SGD(model.trainable_params(), 1e-2)
 
# 1. Define forward function
def forward_fn(data, label):
    logits = model(data)
    loss = loss_fn(logits, label)
    return loss, logits
 
# 2. Get gradient function
grad_fn = mindspore.value_and_grad(forward_fn, None, optimizer.parameters, has_aux=True)
 
# 3. Define function of one-step training
def train_step(data, label):
    (loss, _), grads = grad_fn(data, label)
    optimizer(grads)
    return loss
 
 
def train(model, dataset):
    size = dataset.get_dataset_size()
    model.set_train()     ## 设置当前Cell和所有子Cell的训练模式。对于训练和预测具有不同结构的网络层(如 BatchNorm)，将通过这个属性区分分支。如果设置为True，则执行训练分支，否则执行另一个分支。默认True
    for batch, (data, label) in enumerate(dataset.create_tuple_iterator()):
        loss = train_step(data, label)
        if batch % 100 == 0:
            loss, current = loss.asnumpy(), batch
            print(f"loss: {loss:>7f}  [{current:>3d}/{size:>3d}]")

self_modeltest.py


from mindspore import nn 
 
def test(model, dataset, loss_fn):
    num_batches = dataset.get_dataset_size()
    model.set_train(False)
    total, test_loss, correct = 0, 0, 0
    for 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_batches
    correct /= total
    print(f"Test: \n Accuracy: {(100*correct):>0.1f}%, Avg loss: {test_loss:>8f} \n")

self_predict.py


## 加载模型
from self_network import Network
 
# Instantiate a random initialized model
model = Network()
 
# Load checkpoint and load parameter to model
param_dict = mindspore.load_checkpoint("model.ckpt")
param_not_load, _ = mindspore.load_param_into_net(model, param_dict)  
print(param_not_load)   ## param_not_load是未被加载的参数列表，为空时代表所有参数均加载成功。
 
## 加载后的模型可以直接用于预测推理。
model.set_train(False)
for data, label in test_dataset:
    pred = model(data)
    predicted = pred.argmax(1)
    print(f'Predicted: "{predicted[:10]}", Actual: "{label[:10]}"')
    break

声明：本文内容由网友自发贡献，不代表【wpsshop博客】立场，版权归原作者所有，本站不承担相应法律责任。如您发现有侵权的内容，请联系我们。转载请注明出处：https://www.wpsshop.cn/w/繁依Fanyi0/article/detail/906753

昇思25天学习打卡营第2天|MindSpore快速入门_昇思25天学习打卡营第2天|快速入门

打卡

快速入门案例：minist图像数据识别任务

案例任务说明

流程

1 加载并处理数据集

2 模型网络构建与定义

3 模型约束定义

4 模型训练

5 模型保存

6 模型推理

相关参考文档入门理解

MindSpore数据处理引擎

模型网络参数初始化

模型优化器

损失函数

代码