安装pytorch_lightning要点

安装pytorch_lightning要点

torch和lightning都使用pip安装

安装GPU版本的pytorch（注意cuda版本）：pip install torch==1.8.1+cu101 torchvision==0.9.1+cu101 torchaudio==0.8.1 -f https://download.pytorch.org/whl/torch_stable.html（直接官网找）

安装pytorch_lightning(我安装的是1.4.9版本)：python -m pip install lightning==1.4.9（或者是pip install lightning==1.4.9，不记得是哪一个了，重要的是后面跟版本号）（安装完了之后记得重新打印一下torch版本看一下，有可能安装lightning时把原先的GPU版torch卸载，安装了CPU版的torch，这也是坑所在，注意注意！）

此时出现报错：ImportError: cannot import name 'get_num_classes' from 'torchmetrics.utilities.data'

降低torchmetrics版本：pip install torchmetrics==0.6.0(0.6.0其实是瞎碰的，然后报错竟然没有了)

验证：import pytorch_lighrning as pl没出现报错就是安装成功了（或者直接运行下面的最简LigntningModule）

• 实现最简LigntningModule
  参考pytorch lightning最简上手

import os
import torch
from torch.nn import functional as F
from torch.utils.data import DataLoader
from torchvision.datasets import MNIST
from torchvision import transforms
import pytorch_lightning as pl


class MNISTModel(pl.LightningModule):
# 实现 training_step，train_dataloader, configure_optimizer，已经是最简单的 LightningModule 的实现了。如果连这三个方法都没有实现的话，将会报错：
# No `xxx` method defined. Lightning `Trainer` expects as minimum a `training_step()`, `train_dataloader()` and `configure_optimizers()` to be defined
    def __init__(self):
        super(MNISTModel, self).__init__()
        self.l1 = torch.nn.Linear(28 * 28, 10)

    def forward(self, x):
        return torch.relu(self.l1(x.view(x.size(0), -1)))
      
    def train_dataloader(self):
        return DataLoader(MNIST(os.getcwd(), train=True, download=True, transform=transforms.ToTensor()), batch_size=32)
      
    def training_step(self, batch, batch_nb):
        x, y = batch
        loss = F.cross_entropy(self(x), y)
        return loss

    def configure_optimizers(self):
        return torch.optim.Adam(self.parameters(), lr=0.02)

    
# 启动训练的最简实现非常简单，只需三行：实例化模型、实例化训练器、开始训练！
model = MNISTModel()
trainer = pl.Trainer(accelerator="auto", max_epochs=2)
trainer.fit(model)

运行结果显示：

GPU available: True, used: True
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]

  | Name | Type   | Params
--------------------------------
0 | l1   | Linear | 7.9 K 
--------------------------------
7.9 K     Trainable params
0         Non-trainable params
7.9 K     Total params
0.031     Total estimated model params size (MB)
/home/jhk/anaconda3/envs/openg/lib/python3.8/site-packages/pytorch_lightning/trainer/data_loading.py:105: UserWarning: The dataloader, train dataloader, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 16 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.
  rank_zero_warn(
Epoch 1: 100%|████████████████████████████████| 1875/1875 [00:05<00:00, 338.58it/s, loss=0.709, v_num=7]

至此，应该是安装成功了。
(折腾了一上午，简直崩溃！)
以上为个人经验记录，仅供参考。