mindtorch 0.2.0

Introduction

MindTorch is MindSpore tool for adapting the PyTorch interface, which is designed to make PyTorch code perform efficiently on Ascend without changing the habits of the original PyTorch users.

Install

MindTorch has some prerequisites that need to be installed first, including MindSpore, PIL, NumPy.

# for last stable version
pip install mindtorch

# for latest release candidate
pip install --upgrade --pre mindtorch

Alternatively, you can install the latest or development version by directly pulling from OpenI:

pip3 install git+https://openi.pcl.ac.cn/OpenI/MSAdapter.git

User guide

For data processing and model building, MindTorch can be used in the same way as PyTorch, while the model training part of the code needs to be customized, as shown in the following example.

1. Data processing (only modify the import package)

from mindtorch.torch.utils.data import DataLoader
from mindtorch.torchvision import datasets, transforms

transform = transforms.Compose([transforms.Resize((224, 224), interpolation=InterpolationMode.BICUBIC),
                                transforms.ToTensor(),
                                transforms.Normalize(mean=[0.4914, 0.4822, 0.4465], std=[0.247, 0.2435, 0.2616])
                               ])
train_images = datasets.CIFAR10('./', train=True, download=True, transform=transform)
train_data = DataLoader(train_images, batch_size=128, shuffle=True, num_workers=2, drop_last=True)

2. Model construction (modify import package only)

from mindtorch.torch.nn import Module, Linear, Flatten

class MLP(Module):
    def __init__(self):
        super(MLP, self).__init__()
        self.flatten = Flatten()
        self.line1 = Linear(in_features=1024, out_features=64)
        self.line2 = Linear(in_features=64, out_features=128, bias=False)
        self.line3 = Linear(in_features=128, out_features=10)

    def forward(self, inputs):
        x = self.flatten(inputs)
        x = self.line1(x)
        x = self.line2(x)
        x = self.line3(x)
        return x

3.Model training (custom training)

import mindtorch.torch as torch
import mindtorch.torch.nn as nn
import mindspore as ms

net = MLP()
net.train()
epochs = 500
criterion = nn.CrossEntropyLoss()
optimizer = ms.nn.SGD(net.trainable_params(), learning_rate=0.01, momentum=0.9, weight_decay=0.0005)

# Define the training process
loss_net = ms.nn.WithLossCell(net, criterion)
train_net = ms.nn.TrainOneStepCell(loss_net, optimizer)

for i in range(epochs):
    for X, y in train_data:
        res = train_net(X, y)
        print("epoch:{}, loss:{:.6f}".format(i, res.asnumpy()))
# Save model
ms.save_checkpoint(net, "save_path.ckpt")

License

MindTorch is released under the Apache 2.0 license.