Skip to main content

A PyTorch adapter running on MindSpore

Project description

torch-ms-adapter

PyPI version Python License

torch-ms-adapter 是一个在 MindSpore 上模拟 PyTorch API 的适配器库。
目标是让你几乎无需修改 PyTorch 代码,就能在 MindSpore 上运行。

特性

  • 支持 nn.Module, Conv2d, Linear, ReLU, CrossEntropyLoss
  • 封装了 DataLoader 和 Trainer(底层使用 MindSpore)
  • 提供 torch.nn.functional 基本操作
  • 自动屏蔽 PyTorch,兼容原始导入语句

安装

pip install torch-ms-adapter

使用示例

import torch_ms_adapter as torch
import torch_ms_adapter.nn as nn
import torch_ms_adapter.functional as F

class LeNet5(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(1, 6, 5)
        self.fc1 = nn.Linear(16*4*4, 120)
        self.fc2 = nn.Linear(120, 10)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        x = torch.flatten(x, 1)
        x = self.fc2(F.relu(self.fc1(x)))
        return x

运行方式与pytorch一致,只需替换

import torch

import torch_ms_adapter as torch

路线图

  • 支持基础层(linear,Conv2d,ReLU)
  • 支持Loss函数
  • 封装Optimizer
  • 支持更多PyTorch模块(YOLO,ResNet)
  • 完善Autograd

贡献

欢迎提交issue和PR,一起完善torch-ms-adopter

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

torch_ms_adapter-0.1.4.tar.gz (6.7 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

torch_ms_adapter-0.1.4-py3-none-any.whl (7.6 kB view details)

Uploaded Python 3

File details

Details for the file torch_ms_adapter-0.1.4.tar.gz.

File metadata

  • Download URL: torch_ms_adapter-0.1.4.tar.gz
  • Upload date:
  • Size: 6.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.9.23

File hashes

Hashes for torch_ms_adapter-0.1.4.tar.gz
Algorithm Hash digest
SHA256 b29c9e82c3e3ce8881f06f22a0cca08e674345bffa7186894bb97c40f981ad5b
MD5 73b4cbff2c19659d6b9773cc379bb567
BLAKE2b-256 4a0f687071bc9be06362e936bedd9c559c7fdef277b9c3eda098dc7373060cd3

See more details on using hashes here.

File details

Details for the file torch_ms_adapter-0.1.4-py3-none-any.whl.

File metadata

File hashes

Hashes for torch_ms_adapter-0.1.4-py3-none-any.whl
Algorithm Hash digest
SHA256 9bb086157f02156a22025282853005b591dbbe02b4ac6cdbcf8c35019d5ac2aa
MD5 03a8d2b66b618f212ed2ce7239bf03b4
BLAKE2b-256 fbca7bddfa9a316d678563beaef421b043148d5e4a6b193d18cff3d615fdf799

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page