# ----------------------------------# torchscript# ----------------------------------autoencoder = LitAutoEncoder()torch.jit.save(autoencoder.to_torchscript(), "model.pt")os.path.isfile("model.pt")
# ----------------------------------# onnx# ----------------------------------with tempfile.NamedTemporaryFile(suffix='.onnx', delete=False) as tmpfile:     autoencoder = LitAutoEncoder()     input_sample = torch.randn((1, 28 * 28))     autoencoder.to_onnx(tmpfile.name, input_sample, export_params=True)     os.path.isfile(tmpfile.name)

class LitModel(pl.LightningModule):    def __init__(self):        ...        self.train_acc = pl.metrics.Accuracy()        self.valid_acc = pl.metrics.Accuracy()
    def training_step(self, batch, batch_idx):        logits = self(x)        ...        self.train_acc(logits, y)        # log step metric        self.log('train_acc_step', self.train_acc)
    def validation_step(self, batch, batch_idx):        logits = self(x)        ...        self.valid_acc(logits, y)        # logs epoch metrics        self.log('valid_acc', self.valid_acc)

from pytorch_lightning.metrics import Metric
class MyAccuracy(Metric):
    def __init__(self, dist_sync_on_step=False):        super().__init__(dist_sync_on_step=dist_sync_on_step)        self.add_state("correct", default=torch.tensor(0), dist_reduce_fx="sum")        self.add_state("total", default=torch.tensor(0), dist_reduce_fx="sum")            def update(self, preds: torch.Tensor, target: torch.Tensor):        preds, target = self._input_format(preds, target)        assert preds.shape == target.shape        self.correct += torch.sum(preds == target)        self.total += target.numel()     def compute(self):        return self.correct.float() / self.total

def training_step(self, batch, batch_idx):    loss = self.encoder(batch[0])    return loss

trainer?=?Trainer(automatic_optimization=False)

from pytorch_lightning.metrics import Metric
class MyAccuracy(Metric):
    def __init__(self, dist_sync_on_step=False):        super().__init__(dist_sync_on_step=dist_sync_on_step)        self.add_state("correct", default=torch.tensor(0), dist_reduce_fx="sum")        self.add_state("total", default=torch.tensor(0), dist_reduce_fx="sum")            def update(self, preds: torch.Tensor, target: torch.Tensor):        preds, target = self._input_format(preds, target)        assert preds.shape == target.shape        self.correct += torch.sum(preds == target)        self.total += target.numel()     def compute(self):        return self.correct.float() / self.total

def training_step(self, batch, batch_idx):   self.log('my_metric', x)

def training_step(self, batch, batch_idx): self.log('my_loss',?loss,?on_step=True,?on_epoch=True,?prog_bar=True,?logger=True)

outs = []for batch in data:  out = training_step(batch)  outs.append(out)training_epoch_end(outs)

def training_step(self, batch, batch_idx):  prediction = …  return {'loss': loss, 'preds': prediction}
def training_epoch_end(self, training_step_outputs):  for out in training_step_outputs:    prediction = out['preds']  # do something with these

from pytorch_lightning.callbacks import ModelCheckpoint
class LitAutoEncoder(pl.LightningModule):    def validation_step(self, batch, batch_idx):        x, y = batch        y_hat = self.backbone(x)
        # 1. calculate loss        loss = F.cross_entropy(y_hat, y)
        # 2. log `val_loss`        self.log('val_loss', loss)
# 3. Init ModelCheckpoint callback, monitoring 'val_loss'checkpoint_callback = ModelCheckpoint(monitor='val_loss')
# 4. Pass your callback to checkpoint_callback trainer flagtrainer = Trainer(checkpoint_callback=checkpoint_callback)