lightning/docs/source/new-project.rst

.. testsetup:: *

    import os
    import torch
    from torch.nn import functional as F
    from torch.utils.data import DataLoader
    from torch.utils.data import random_split
    import pytorch_lightning as pl
    from pytorch_lightning.core.datamodule import LightningDataModule
    from pytorch_lightning.core.lightning import LightningModule
    from pytorch_lightning.trainer.trainer import Trainer

.. _new_project:

####################
Lightning in 2 steps
####################

**In this guide we'll show you how to organize your PyTorch code into Lightning in 2 steps.**

Organizing your code with PyTorch Lightning makes your code:

* Keep all the flexibility (this is all pure PyTorch), but removes a ton of boilerplate
* More readable by decoupling the research code from the engineering
* Easier to reproduce
* Less error-prone by automating most of the training loop and tricky engineering
* Scalable to any hardware without changing your model

----------

Here's a 3 minute conversion guide for PyTorch projects:

.. raw:: html

    <video width="100%" max-width="800px" controls autoplay muted playsinline
    src="https://pl-bolts-doc-images.s3.us-east-2.amazonaws.com/pl_docs/pl_docs_animation_final.m4v"></video>

----------

*********************************
Step 0: Install PyTorch Lightning
*********************************


You can install using `pip <https://pypi.org/project/pytorch-lightning/>`_

.. code-block:: bash

    pip install pytorch-lightning

Or with `conda <https://anaconda.org/conda-forge/pytorch-lightning>`_ (see how to install conda `here <https://docs.conda.io/projects/conda/en/latest/user-guide/install/>`_):

.. code-block:: bash

    conda install pytorch-lightning -c conda-forge

You could also use conda environments

.. code-block:: bash

    conda activate my_env
    pip install pytorch-lightning

----------

Import the following:

.. testcode::
    :skipif: not _TORCHVISION_AVAILABLE

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

******************************
Step 1: Define LightningModule
******************************

.. testcode::

    class LitAutoEncoder(LightningModule):

        def __init__(self):
            super().__init__()
            self.encoder = nn.Sequential(
                nn.Linear(28*28, 64),
                nn.ReLU(),
                nn.Linear(64, 3)
            )
            self.decoder = nn.Sequential(
                nn.Linear(3, 64),
                nn.ReLU(),
                nn.Linear(64, 28*28)
            )

        def forward(self, x):
            # in lightning, forward defines the prediction/inference actions
            embedding = self.encoder(x)
            return embedding

        def training_step(self, batch, batch_idx):
            # training_step defined 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 = F.mse_loss(x_hat, x)
            # Logging to TensorBoard by default
            self.log('train_loss', loss)
            return loss

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

**SYSTEM VS MODEL**

A :ref:`lightning_module` defines a *system* not a model.

.. figure:: https://pl-bolts-doc-images.s3.us-east-2.amazonaws.com/pl_docs/model_system.png
    :width: 400

Examples of systems are:

- `Autoencoder <https://github.com/PyTorchLightning/pytorch-lightning-bolts/blob/master/pl_bolts/models/autoencoders/basic_ae/basic_ae_module.py>`_
- `BERT <https://colab.research.google.com/drive/1F_RNcHzTfFuQf-LeKvSlud6x7jXYkG31#scrollTo=yr7eaxkF-djf>`_
- `DQN <https://colab.research.google.com/drive/1F_RNcHzTfFuQf-LeKvSlud6x7jXYkG31#scrollTo=IAlT0-75T_Kv>`_
- `GAN <https://github.com/PyTorchLightning/pytorch-lightning-bolts/blob/master/pl_bolts/models/gans/basic/basic_gan_module.py>`_
- `Image classifier <https://colab.research.google.com/drive/1F_RNcHzTfFuQf-LeKvSlud6x7jXYkG31#scrollTo=gEulmrbxwaYL>`_
- Seq2seq
- `SimCLR <https://github.com/PyTorchLightning/pytorch-lightning-bolts/blob/master/pl_bolts/models/self_supervised/simclr/simclr_module.py>`_
- `VAE <https://github.com/PyTorchLightning/pytorch-lightning-bolts/blob/master/pl_bolts/models/autoencoders/basic_vae/basic_vae_module.py>`_

Under the hood a LightningModule is still just a :class:`torch.nn.Module` that groups all research code into a single file to make it self-contained:

- The Train loop
- The Validation loop
- The Test loop
- The Model or system of Models
- The Optimizer

You can customize any part of training (such as the backward pass) by overriding any
of the 20+ hooks found in :ref:`hooks`

.. testcode::

    class LitAutoEncoder(LightningModule):

        def backward(self, loss, optimizer, optimizer_idx):
            loss.backward()

**FORWARD vs TRAINING_STEP**

In Lightning we separate training from inference. The training_step defines
the full training loop. We encourage users to use the forward to define inference
actions.

For example, in this case we could define the autoencoder to act as an embedding extractor:

.. code-block:: python

    def forward(self, x):
        embeddings = self.encoder(x)
        return embeddings

Of course, nothing is stopping you from using forward from within the training_step.

.. code-block:: python

    def training_step(self, batch, batch_idx):
        ...
        z = self(x)

It really comes down to your application. We do, however, recommend that you keep both intents separate.

* Use forward for inference (predicting).
* Use training_step for training.

More details in :ref:`lightning_module` docs.

----------

**********************************
Step 2: Fit with Lightning Trainer
**********************************

First, define the data however you want. Lightning just needs a :class:`~torch.utils.data.DataLoader` for the train/val/test splits.

.. code-block:: python

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

Next, init the :ref:`lightning_module` and the PyTorch Lightning :class:`~pytorch_lightning.trainer.Trainer`,
then call fit with both the data and model.

.. code-block:: python

    # init model
    autoencoder = LitAutoEncoder()

    # most basic trainer, uses good defaults (auto-tensorboard, checkpoints, logs, and more)
    # trainer = pl.Trainer(gpus=8) (if you have GPUs)
    trainer = pl.Trainer()
    trainer.fit(autoencoder, train_loader)

The :class:`~pytorch_lightning.trainer.Trainer` automates:

* Epoch and batch iteration
* Calling of optimizer.step(), backward, zero_grad()
* Calling of .eval(), enabling/disabling grads
* :ref:`weights_loading`
* Tensorboard (see :ref:`loggers` options)
* :ref:`multi_gpu` support
* :ref:`tpu`
* :ref:`amp` support

.. tip:: If you prefer to manually manage optimizers you can use the :ref:`manual_opt` mode  (ie: RL, GANs, etc...).


---------

**That's it!**

These are the main 2 concepts you need to know in Lightning. All the other features of lightning are either
features of the Trainer or LightningModule.

-----------

**************
Basic features
**************

Manual vs automatic optimization
================================

Automatic optimization
----------------------
With Lightning, you don't need to worry about when to enable/disable grads, do a backward pass, or update optimizers
as long as you return a loss with an attached graph from the `training_step`, Lightning will automate the optimization.

.. code-block:: python

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

.. _manual_opt:

Manual optimization
-------------------
However, for certain research like GANs, reinforcement learning, or something with multiple optimizers
or an inner loop, you can turn off automatic optimization and fully control the training loop yourself.

First, turn off automatic optimization:

.. testcode::

    trainer = Trainer(automatic_optimization=False)

Now you own the train loop!

.. code-block:: python

    def training_step(self, batch, batch_idx, opt_idx):
        # access your optimizers with use_pl_optimizer=False. Default is True
        (opt_a, opt_b, opt_c) = self.optimizers(use_pl_optimizer=True)

        loss_a = self.generator(batch[0])

        # use this instead of loss.backward so we can automate half precision, etc...
        self.manual_backward(loss_a, opt_a, retain_graph=True)
        self.manual_backward(loss_a, opt_a)
        opt_a.step()
        opt_a.zero_grad()

        loss_b = self.discriminator(batch[0])
        self.manual_backward(loss_b, opt_b)
        ...


Predict or Deploy
=================
When you're done training, you have 3 options to use your LightningModule for predictions.

Option 1: Sub-models
--------------------
Pull out any model inside your system for predictions.

.. code-block:: python

    # ----------------------------------
    # to use as embedding extractor
    # ----------------------------------
    autoencoder = LitAutoEncoder.load_from_checkpoint('path/to/checkpoint_file.ckpt')
    encoder_model = autoencoder.encoder
    encoder_model.eval()

    # ----------------------------------
    # to use as image generator
    # ----------------------------------
    decoder_model = autoencoder.decoder
    decoder_model.eval()

Option 2: Forward
-----------------
You can also add a forward method to do predictions however you want.

.. testcode::

    # ----------------------------------
    # using the AE to extract embeddings
    # ----------------------------------
    class LitAutoEncoder(LightningModule):
        def __init__(self):
            super().__init__()
            self.encoder = nn.Sequential()

        def forward(self, x):
            embedding = self.encoder(x)
            return embedding

    autoencoder = LitAutoEncoder()
    autoencoder = autoencoder(torch.rand(1, 28 * 28))


.. code-block:: python

    # ----------------------------------
    # or using the AE to generate images
    # ----------------------------------
    class LitAutoEncoder(LightningModule):
        def __init__(self):
            super().__init__()
            self.decoder = nn.Sequential()

        def forward(self):
            z = torch.rand(1, 3)
            image = self.decoder(z)
            image = image.view(1, 1, 28, 28)
            return image

    autoencoder = LitAutoEncoder()
    image_sample = autoencoder()

Option 3: Production
--------------------
For production systems, onnx or torchscript are much faster. Make sure you have added
a forward method or trace only the sub-models you need.

.. code-block:: python

    # ----------------------------------
    # torchscript
    # ----------------------------------
    autoencoder = LitAutoEncoder()
    torch.jit.save(autoencoder.to_torchscript(), "model.pt")
    os.path.isfile("model.pt")

.. code-block:: python

    # ----------------------------------
    # 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)

--------------------

Using CPUs/GPUs/TPUs
====================
It's trivial to use CPUs, GPUs or TPUs in Lightning. There's **NO NEED** to change your code, simply change the :class:`~pytorch_lightning.trainer.Trainer` options.

.. testcode::

    # train on CPU
    trainer = Trainer()

.. testcode::

    # train on 8 CPUs
    trainer = Trainer(num_processes=8)

.. code-block:: python

    # train on 1024 CPUs across 128 machines
    trainer = pl.Trainer(
        num_processes=8,
        num_nodes=128
    )

.. code-block:: python

    # train on 1 GPU
    trainer = pl.Trainer(gpus=1)

.. code-block:: python

    # train on multiple GPUs across nodes (32 gpus here)
    trainer = pl.Trainer(
        gpus=4,
        num_nodes=8
    )

.. code-block:: python

    # train on gpu 1, 3, 5 (3 gpus total)
    trainer = pl.Trainer(gpus=[1, 3, 5])

.. code-block:: python

    # Multi GPU with mixed precision
    trainer = pl.Trainer(gpus=2, precision=16)

.. code-block:: python

    # Train on TPUs
    trainer = pl.Trainer(tpu_cores=8)

Without changing a SINGLE line of your code, you can now do the following with the above code:

.. code-block:: python

    # train on TPUs using 16 bit precision
    # using only half the training data and checking validation every quarter of a training epoch
    trainer = pl.Trainer(
        tpu_cores=8,
        precision=16,
        limit_train_batches=0.5,
        val_check_interval=0.25
    )

-----------

Checkpoints
===========
Lightning automatically saves your model. Once you've trained, you can load the checkpoints as follows:

.. code-block:: python

    model = LitModel.load_from_checkpoint(path)

The above checkpoint contains all the arguments needed to init the model and set the state dict.
If you prefer to do it manually, here's the equivalent

.. code-block:: python

    # load the ckpt
    ckpt = torch.load('path/to/checkpoint.ckpt')

    # equivalent to the above
    model = LitModel()
    model.load_state_dict(ckpt['state_dict'])

---------

Data flow
=========
Each loop (training, validation, test) has three hooks you can implement:

- x_step
- x_step_end
- x_epoch_end

To illustrate how data flows, we'll use the training loop (ie: x=training)

.. code-block:: python

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

The equivalent in Lightning is:

.. code-block:: python

    def training_step(self, batch, batch_idx):
        prediction = ...
        return prediction

    def training_epoch_end(self, training_step_outputs):
        for prediction in predictions:
            # do something with these

In the event that you use DP or DDP2 distributed modes (ie: split a batch across GPUs),
use the x_step_end to manually aggregate (or don't implement it to let lightning auto-aggregate for you).

.. code-block:: python

    for batch in data:
        model_copies = copy_model_per_gpu(model, num_gpus)
        batch_split = split_batch_per_gpu(batch, num_gpus)

        gpu_outs = []
        for model, batch_part in zip(model_copies, batch_split):
            # LightningModule hook
            gpu_out = model.training_step(batch_part)
            gpu_outs.append(gpu_out)

        # LightningModule hook
        out = training_step_end(gpu_outs)

The lightning equivalent is:

.. code-block:: python

    def training_step(self, batch, batch_idx):
        loss = ...
        return loss

    def training_step_end(self, losses):
        gpu_0_loss = losses[0]
        gpu_1_loss = losses[1]
        return (gpu_0_loss + gpu_1_loss) * 1/2

.. tip:: The validation and test loops have the same structure.

-----------------

Logging
=======
To log to Tensorboard, your favorite logger, and/or the progress bar, use the
:func:`~~pytorch_lightning.core.lightning.LightningModule.log` method which can be called from
any method in the LightningModule.

.. code-block:: python

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

The :func:`~~pytorch_lightning.core.lightning.LightningModule.log` method has a few options:

- on_step (logs the metric at that step in training)
- on_epoch (automatically accumulates and logs at the end of the epoch)
- prog_bar (logs to the progress bar)
- logger (logs to the logger like Tensorboard)

Depending on where the log is called from, Lightning auto-determines the correct mode for you. But of course
you can override the default behavior by manually setting the flags

.. note:: Setting on_epoch=True will accumulate your logged values over the full training epoch.

.. code-block:: python

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

.. note::
    The loss value shown in the progress bar is smoothed (averaged) over the last values,
    so it differs from the actual loss returned in the train/validation step.

You can also use any method of your logger directly:

.. code-block:: python

    def training_step(self, batch, batch_idx):
        tensorboard = self.logger.experiment
        tensorboard.any_summary_writer_method_you_want())

Once your training starts, you can view the logs by using your favorite logger or booting up the Tensorboard logs:

.. code-block:: bash

    tensorboard --logdir ./lightning_logs

.. note::
    Lightning automatically shows the loss value returned from ``training_step`` in the progress bar.
    So, no need to explicitly log like this ``self.log('loss', loss, prog_bar=True)``.

Read more about :ref:`loggers`.

----------------

Optional extensions
===================

Callbacks
---------
A callback is an arbitrary self-contained program that can be executed at arbitrary parts of the training loop.

Here's an example adding a not-so-fancy learning rate decay rule:

.. testcode::

    from pytorch_lightning.callbacks import Callback

    class DecayLearningRate(Callback):

        def __init__(self):
            self.old_lrs = []

        def on_train_start(self, trainer, pl_module):
            # track the initial learning rates
            for opt_idx, optimizer in enumerate(trainer.optimizers):
                group = [param_group['lr'] for param_group in optimizer.param_groups]
                self.old_lrs.append(group)

        def on_train_epoch_end(self, trainer, pl_module, outputs):
            for opt_idx, optimizer in enumerate(trainer.optimizers):
                old_lr_group = self.old_lrs[opt_idx]
                new_lr_group = []
                for p_idx, param_group in enumerate(optimizer.param_groups):
                    old_lr = old_lr_group[p_idx]
                    new_lr = old_lr * 0.98
                    new_lr_group.append(new_lr)
                    param_group['lr'] = new_lr
                self.old_lrs[opt_idx] = new_lr_group

    # And pass the callback to the Trainer
    decay_callback = DecayLearningRate()
    trainer = Trainer(callbacks=[decay_callback])

Things you can do with a callback:

- Send emails at some point in training
- Grow the model
- Update learning rates
- Visualize gradients
- ...
- You are only limited by your imagination

:ref:`Learn more about custom callbacks <callbacks>`.


LightningDataModules
--------------------
DataLoaders and data processing code tends to end up scattered around.
Make your data code reusable by organizing it into a :class:`~pytorch_lightning.core.datamodule.LightningDataModule`.

.. testcode::

  class MNISTDataModule(LightningDataModule):

        def __init__(self, batch_size=32):
            super().__init__()
            self.batch_size = batch_size

        # When doing distributed training, Datamodules have two optional arguments for
        # granular control over download/prepare/splitting data:

        # OPTIONAL, called only on 1 GPU/machine
        def prepare_data(self):
            MNIST(os.getcwd(), train=True, download=True)
            MNIST(os.getcwd(), train=False, download=True)

        # OPTIONAL, called for every GPU/machine (assigning state is OK)
        def setup(self, stage):
            # transforms
            transform=transforms.Compose([
                transforms.ToTensor(),
                transforms.Normalize((0.1307,), (0.3081,))
            ])
            # split dataset
            if stage == 'fit':
                mnist_train = MNIST(os.getcwd(), train=True, transform=transform)
                self.mnist_train, self.mnist_val = random_split(mnist_train, [55000, 5000])
            if stage == 'test':
                self.mnist_test = MNIST(os.getcwd(), train=False, transform=transform)

        # return the dataloader for each split
        def train_dataloader(self):
            mnist_train = DataLoader(self.mnist_train, batch_size=self.batch_size)
            return mnist_train

        def val_dataloader(self):
            mnist_val = DataLoader(self.mnist_val, batch_size=self.batch_size)
            return mnist_val

        def test_dataloader(self):
            mnist_test = DataLoader(self.mnist_test, batch_size=self.batch_size)
            return mnist_test

:class:`~pytorch_lightning.core.datamodule.LightningDataModule` is designed to enable sharing and reusing data splits
and transforms across different projects. It encapsulates all the steps needed to process data: downloading,
tokenizing, processing etc.

Now you can simply pass your :class:`~pytorch_lightning.core.datamodule.LightningDataModule` to
the :class:`~pytorch_lightning.trainer.Trainer`:

.. code-block:: python

    # init model
    model = LitModel()

    # init data
    dm = MNISTDataModule()

    # train
    trainer = pl.Trainer()
    trainer.fit(model, dm)

    # test
    trainer.test(datamodule=dm)

DataModules are specifically useful for building models based on data. Read more on :ref:`datamodules`.

------

Debugging
=========
Lightning has many tools for debugging. Here is an example of just a few of them:

.. testcode::

    # use only 10 train batches and 3 val batches
    trainer = Trainer(limit_train_batches=10, limit_val_batches=3)

.. testcode::

    # Automatically overfit the sane batch of your model for a sanity test
    trainer = Trainer(overfit_batches=1)

.. testcode::

    # unit test all the code- hits every line of your code once to see if you have bugs,
    # instead of waiting hours to crash on validation
    trainer = Trainer(fast_dev_run=True)

.. testcode::

   # train only 20% of an epoch
   trainer = Trainer(limit_train_batches=0.2)

.. testcode::

    # run validation every 25% of a training epoch
    trainer = Trainer(val_check_interval=0.25)

.. testcode::

    # Profile your code to find speed/memory bottlenecks
    Trainer(profiler=True)

---------------

********************
Other coool features
********************

Once you define and train your first Lightning model, you might want to try other cool features like

- :ref:`Automatic early stopping <early_stopping>`
- :ref:`Automatic truncated-back-propagation-through-time <trainer:truncated_bptt_steps>`
- :ref:`Automatically scale your batch size <training_tricks:Auto scaling of batch size>`
- :ref:`Automatically find a good learning rate <lr_finder>`
- :ref:`Load checkpoints directly from S3 <weights_loading:Checkpoint Loading>`
- :ref:`Scale to massive compute clusters <slurm>`
- :ref:`Use multiple dataloaders per train/val/test loop <multiple_loaders>`
- :ref:`Use multiple optimizers to do reinforcement learning or even GANs <optimizers:Use multiple optimizers (like GANs)>`

Or read our :ref:`introduction_guide` to learn more!

-------------

Grid AI
=======
Grid AI is our native solution for large scale training and tuning on the cloud provider of your choice.

`Click here to request early-access <https://www.grid.ai/>`_.

------------

**********
Community
**********
Our community of core maintainers and thousands of expert researchers is active on our
`Slack <https://join.slack.com/t/pytorch-lightning/shared_invite/zt-f6bl2l0l-JYMK3tbAgAmGRrlNr00f1A>`_
and `Forum <https://forums.pytorchlightning.ai/>`_. Drop by to hang out, ask Lightning questions or even discuss research!


-------------

Masterclass
===========
We also offer a Masterclass to teach you the advanced uses of Lightning.

.. image:: _images/general/PTL101_youtube_thumbnail.jpg
    :width: 500
    :align: center
    :alt: Masterclass
    :target: https://www.youtube.com/playlist?list=PLaMu-SDt_RB5NUm67hU2pdE75j6KaIOv2