Hivemind does not write how to connect to it

[drimeF0]

🐛 Bug

ValueError                                Traceback (most recent call last)
     46 train_loader = utils.data.DataLoader(dataset)
     47 
---> 48 trainer = Trainer(
     49   max_epochs=4,
     50   accelerator="auto",

[/usr/local/lib/python3.10/dist-packages/pytorch_lightning/trainer/connectors/accelerator_connector.py](https://localhost:8080/#) in _check_config_and_set_final_flags(self, strategy, accelerator, precision, plugins, sync_batchnorm)
    207 
    208         if strategy != "auto" and strategy not in self._registered_strategies and not isinstance(strategy, Strategy):
--> 209             raise ValueError(
    210                 f"You selected an invalid strategy name: `strategy={strategy!r}`."
    211                 " It must be either a string or an instance of `pytorch_lightning.strategies.Strategy`."

ValueError: You selected an invalid strategy name: `strategy=<lightning_hivemind.strategy.HivemindStrategy object at 0x7f3d206726e0>`. It must be either a string or an instance of `pytorch_lightning.strategies.Strategy`. Example choices: auto, ddp, ddp_spawn, deepspeed, ... Find a complete list of options in our documentation at https://lightning.ai/

Code sample

import os
from torch import optim, nn, utils, Tensor
from torchvision.datasets import MNIST
from torchvision.transforms import ToTensor
import lightning.pytorch as pl
from lightning_hivemind.strategy import HivemindStrategy

from pytorch_lightning import Trainer

import torch

# define any number of nn.Modules (or use your current ones)
encoder = nn.Sequential(nn.Linear(28 * 28, 64), nn.ReLU(), nn.Linear(64, 3))
decoder = nn.Sequential(nn.Linear(3, 64), nn.ReLU(), nn.Linear(64, 28 * 28))


# define the LightningModule
class LitAutoEncoder(pl.LightningModule):
    def __init__(self, encoder, decoder):
        super().__init__()
        self.encoder = encoder
        self.decoder = decoder

    def training_step(self, batch, batch_idx):
        # training_step defines the train loop.
        # it is independent of forward
        x, y = batch
        x = x.view(x.size(0), -1)
        z = self.encoder(x)
        x_hat = self.decoder(z)
        loss = nn.functional.mse_loss(x_hat, x)
        # Logging to TensorBoard (if installed) by default
        self.log("train_loss", loss)
        return loss

    def configure_optimizers(self):
        optimizer = optim.Adam(self.parameters(), lr=1e-3)
        return optimizer


# init the autoencoder
autoencoder = LitAutoEncoder(encoder, decoder)


dataset = MNIST(os.getcwd(), download=True, transform=ToTensor())
train_loader = utils.data.DataLoader(dataset)

trainer = Trainer(
  max_epochs=4,
  accelerator="auto",
  devices=1 if torch.cuda.is_available() else None,strategy=HivemindStrategy(target_batch_size=8192)
)
trainer.fit(model=autoencoder, train_dataloaders=train_loader)

Environment

• PyTorch Version (e.g., 1.0): 2.0.0+cu118
• OS (e.g., Linux): Ubuntu 20.04.5 LTS
• How you installed PyTorch (conda, pip, source): pip
• Python version: Python 3.10.11
• CUDA/cuDNN version: CUDA Version: 12.0
• GPU models and configuration: Tesla T4

[github-actions]

Hi! thanks for your contribution!, great first issue!

[drimeF0]

i fixed the previous mistake by replacing

py from pytorch_lightning import Trainer

to

from lightning import Trainer

but now I get another error:

May 04 14:49:11.323 [ERROR] [go-libp2p-daemon/daemon.go:190] error accepting connection: accept unix /tmp/hivemind-p2pd-XYPwBtT8_Qw.sock: use of closed network connection

and in the console I don't get data about where to connect