Multi-GPU distributed training with PyTorch

<!DOCTYPE html> <html lang="en"> <head> <meta charset="utf-8"> <meta name="viewport" content="width=device-width, initial-scale=1"> <meta name="description" content="Keras documentation"> <meta name="author" content="Keras Team"> <link rel="shortcut icon" href="https://keras.io/img/favicon.ico"> <link rel="canonical" href="https://keras.io/guides/distributed_training_with_torch/" />  <meta property="og:title" content="Keras documentation: Multi-GPU distributed training with PyTorch"> <meta property="og:image" content="https://keras.io/img/logo-k-keras-wb.png"> <meta name="twitter:title" content="Keras documentation: Multi-GPU distributed training with PyTorch"> <meta name="twitter:image" content="https://keras.io/img/k-keras-social.png"> <meta name="twitter:card" content="summary"> <title>Multi-GPU distributed training with PyTorch</title>  <link href="/css/bootstrap.min.css" rel="stylesheet">  <link href="https://fonts.googleapis.com/css2?family=Open+Sans:wght@400;600;700;800&display=swap" rel="stylesheet">  <link href="/css/docs.css" rel="stylesheet"> <link href="/css/monokai.css" rel="stylesheet">  <script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start': new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0], j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src= 'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f); })(window,document,'script','dataLayer','GTM-5DNGF4N'); </script> <script> (function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){ (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o), m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m) })(window,document,'script','https://www.google-analytics.com/analytics.js','ga'); ga('create', 'UA-175165319-128', 'auto'); ga('send', 'pageview'); </script>  <script async defer src="https://buttons.github.io/buttons.js"></script> </head> <body>  <noscript><iframe src="https://www.googletagmanager.com/ns.html?id=GTM-5DNGF4N" height="0" width="0" style="display:none;visibility:hidden"></iframe></noscript>  <div class='k-page'> <div class="k-nav" id="nav-menu"> <a href='/'><img src='/img/logo-small.png' class='logo-small' /></a> <div class="nav flex-column nav-pills" role="tablist" aria-orientation="vertical"> <a class="nav-link" href="/about/" role="tab" aria-selected="">About Keras</a> <a class="nav-link" href="/getting_started/" role="tab" aria-selected="">Getting started</a> <a class="nav-link active" href="/guides/" role="tab" aria-selected="">Developer guides</a> <a class="nav-sublink" href="/guides/functional_api/">The Functional API</a> <a class="nav-sublink" href="/guides/sequential_model/">The Sequential model</a> <a class="nav-sublink" href="/guides/making_new_layers_and_models_via_subclassing/">Making new layers & models via subclassing</a> <a class="nav-sublink" href="/guides/training_with_built_in_methods/">Training & evaluation with the built-in methods</a> <a class="nav-sublink" href="/guides/custom_train_step_in_jax/">Customizing `fit()` with JAX</a> <a class="nav-sublink" href="/guides/custom_train_step_in_tensorflow/">Customizing `fit()` with TensorFlow</a> <a class="nav-sublink" href="/guides/custom_train_step_in_torch/">Customizing `fit()` with PyTorch</a> <a class="nav-sublink" href="/guides/writing_a_custom_training_loop_in_jax/">Writing a custom training loop in JAX</a> <a class="nav-sublink" href="/guides/writing_a_custom_training_loop_in_tensorflow/">Writing a custom training loop in TensorFlow</a> <a class="nav-sublink" href="/guides/writing_a_custom_training_loop_in_torch/">Writing a custom training loop in PyTorch</a> <a class="nav-sublink" href="/guides/serialization_and_saving/">Serialization & saving</a> <a class="nav-sublink" href="/guides/customizing_saving_and_serialization/">Customizing saving & serialization</a> <a class="nav-sublink" href="/guides/writing_your_own_callbacks/">Writing your own callbacks</a> <a class="nav-sublink" href="/guides/transfer_learning/">Transfer learning & fine-tuning</a> <a class="nav-sublink" href="/guides/distributed_training_with_jax/">Distributed training with JAX</a> <a class="nav-sublink" href="/guides/distributed_training_with_tensorflow/">Distributed training with TensorFlow</a> <a class="nav-sublink active" href="/guides/distributed_training_with_torch/">Distributed training with PyTorch</a> <a class="nav-sublink" href="/guides/distribution/">Distributed training with Keras 3</a> <a class="nav-sublink" href="/guides/migrating_to_keras_3/">Migrating Keras 2 code to Keras 3</a> <a class="nav-sublink" href="/guides/keras_tuner/">Hyperparameter Tuning</a> <a class="nav-sublink" href="/guides/keras_cv/">KerasCV</a> <a class="nav-sublink" href="/guides/keras_nlp/">KerasNLP</a> <a class="nav-sublink" href="/guides/keras_hub/">KerasHub</a> <a class="nav-link" href="/api/" role="tab" aria-selected="">Keras 3 API documentation</a> <a class="nav-link" href="/2.18/api/" role="tab" aria-selected="">Keras 2 API documentation</a> <a class="nav-link" href="/examples/" role="tab" aria-selected="">Code examples</a> <a class="nav-link" href="/keras_tuner/" role="tab" aria-selected="">KerasTuner: Hyperparameter Tuning</a> <a class="nav-link" href="/keras_hub/" role="tab" aria-selected="">KerasHub: Pretrained Models</a> <a class="nav-link" href="/keras_cv/" role="tab" aria-selected="">KerasCV: Computer Vision Workflows</a> <a class="nav-link" href="/keras_nlp/" role="tab" aria-selected="">KerasNLP: Natural Language Workflows</a> </div> </div> <div class='k-main'> <div class='k-main-top'> <script> function displayDropdownMenu() { e = document.getElementById("nav-menu"); if (e.style.display == "block") { e.style.display = "none"; } else { e.style.display = "block"; document.getElementById("dropdown-nav").style.display = "block"; } } function resetMobileUI() { if (window.innerWidth <= 840) { document.getElementById("nav-menu").style.display = "none"; document.getElementById("dropdown-nav").style.display = "block"; } else { document.getElementById("nav-menu").style.display = "block"; document.getElementById("dropdown-nav").style.display = "none"; } var navmenu = document.getElementById("nav-menu"); var menuheight = navmenu.clientHeight; var kmain = document.getElementById("k-main-id"); kmain.style.minHeight = (menuheight + 100) + 'px'; } window.onresize = resetMobileUI; window.addEventListener("load", (event) => { resetMobileUI() }); </script> <div id='dropdown-nav' onclick="displayDropdownMenu();"> <svg viewBox="-20 -20 120 120" width="60" height="60"> <rect width="100" height="20"></rect> <rect y="30" width="100" height="20"></rect> <rect y="60" width="100" height="20"></rect> </svg> </div> <form class="bd-search d-flex align-items-center k-search-form" id="search-form"> <input type="search" class="k-search-input" id="search-input" placeholder="Search Keras documentation..." aria-label="Search Keras documentation..." autocomplete="off"> <button class="k-search-btn"> <svg width="13" height="13" viewBox="0 0 13 13"><title>search</title><path d="m4.8495 7.8226c0.82666 0 1.5262-0.29146 2.0985-0.87438 0.57232-0.58292 0.86378-1.2877 0.87438-2.1144 0.010599-0.82666-0.28086-1.5262-0.87438-2.0985-0.59352-0.57232-1.293-0.86378-2.0985-0.87438-0.8055-0.010599-1.5103 0.28086-2.1144 0.87438-0.60414 0.59352-0.8956 1.293-0.87438 2.0985 0.021197 0.8055 0.31266 1.5103 0.87438 2.1144 0.56172 0.60414 1.2665 0.8956 2.1144 0.87438zm4.4695 0.2115 3.681 3.6819-1.259 1.284-3.6817-3.7 0.0019784-0.69479-0.090043-0.098846c-0.87973 0.76087-1.92 1.1413-3.1207 1.1413-1.3553 0-2.5025-0.46363-3.4417-1.3909s-1.4088-2.0686-1.4088-3.4239c0-1.3553 0.4696-2.4966 1.4088-3.4239 0.9392-0.92727 2.0864-1.3969 3.4417-1.4088 1.3553-0.011889 2.4906 0.45771 3.406 1.4088 0.9154 0.95107 1.379 2.0924 1.3909 3.4239 0 1.2126-0.38043 2.2588-1.1413 3.1385l0.098834 0.090049z"></path></svg> </button> </form> <script> var form = document.getElementById('search-form'); form.onsubmit = function(e) { e.preventDefault(); var query = document.getElementById('search-input').value; window.location.href = '/search.html?query=' + query; return False } </script> </div> <div class='k-main-inner' id='k-main-id'> <div class='k-location-slug'> ► <a href='/guides/'>Developer guides</a> / Multi-GPU distributed training with PyTorch </div> <div class='k-content'> <h1 id="multigpu-distributed-training-with-pytorch">Multi-GPU distributed training with PyTorch</h1> Author: <a href="https://twitter.com/fchollet">fchollet</a> Date created: 2023/06/29 Last modified: 2023/06/29 Description: Guide to multi-GPU training for Keras models with PyTorch. <img class="k-inline-icon" src="https://colab.research.google.com/img/colab_favicon.ico"/> <a href="https://colab.research.google.com/github/keras-team/keras-io/blob/master/guides/ipynb/distributed_training_with_torch.ipynb">View in Colab</a> •<img class="k-inline-icon" src="https://github.com/favicon.ico"/> <a href="https://github.com/keras-team/keras-io/blob/master/guides/distributed_training_with_torch.py">GitHub source</a> <hr /> <h2 id="introduction">Introduction</h2> There are generally two ways to distribute computation across multiple devices: Data parallelism, where a single model gets replicated on multiple devices or multiple machines. Each of them processes different batches of data, then they merge their results. There exist many variants of this setup, that differ in how the different model replicas merge results, in whether they stay in sync at every batch or whether they are more loosely coupled, etc. Model parallelism, where different parts of a single model run on different devices, processing a single batch of data together. This works best with models that have a naturally-parallel architecture, such as models that feature multiple branches. This guide focuses on data parallelism, in particular synchronous data parallelism, where the different replicas of the model stay in sync after each batch they process. Synchronicity keeps the model convergence behavior identical to what you would see for single-device training. Specifically, this guide teaches you how to use PyTorch's <code>DistributedDataParallel</code> module wrapper to train Keras, with minimal changes to your code, on multiple GPUs (typically 2 to 16) installed on a single machine (single host, multi-device training). This is the most common setup for researchers and small-scale industry workflows. <hr /> <h2 id="setup">Setup</h2> Let's start by defining the function that creates the model that we will train, and the function that creates the dataset we will train on (MNIST in this case). <div class="codehilite"><pre><code>import os os.environ["KERAS_BACKEND"] = "torch" import torch import numpy as np import keras def get_model(): # Make a simple convnet with batch normalization and dropout. inputs = keras.Input(shape=(28, 28, 1)) x = keras.layers.Rescaling(1.0 / 255.0)(inputs) x = keras.layers.Conv2D(filters=12, kernel_size=3, padding="same", use_bias=False)( x ) x = keras.layers.BatchNormalization(scale=False, center=True)(x) x = keras.layers.ReLU()(x) x = keras.layers.Conv2D( filters=24, kernel_size=6, use_bias=False, strides=2, )(x) x = keras.layers.BatchNormalization(scale=False, center=True)(x) x = keras.layers.ReLU()(x) x = keras.layers.Conv2D( filters=32, kernel_size=6, padding="same", strides=2, name="large_k", )(x) x = keras.layers.BatchNormalization(scale=False, center=True)(x) x = keras.layers.ReLU()(x) x = keras.layers.GlobalAveragePooling2D()(x) x = keras.layers.Dense(256, activation="relu")(x) x = keras.layers.Dropout(0.5)(x) outputs = keras.layers.Dense(10)(x) model = keras.Model(inputs, outputs) return model def get_dataset(): # Load the data and split it between train and test sets (x_train, y_train), (x_test, y_test) = keras.datasets.mnist.load_data() # Scale images to the [0, 1] range x_train = x_train.astype("float32") x_test = x_test.astype("float32") # Make sure images have shape (28, 28, 1) x_train = np.expand_dims(x_train, -1) x_test = np.expand_dims(x_test, -1) print("x_train shape:", x_train.shape) # Create a TensorDataset dataset = torch.utils.data.TensorDataset( torch.from_numpy(x_train), torch.from_numpy(y_train) ) return dataset </code></pre></div> Next, let's define a simple PyTorch training loop that targets a GPU (note the calls to <code>.cuda()</code>). <div class="codehilite"><pre><code>def train_model(model, dataloader, num_epochs, optimizer, loss_fn): for epoch in range(num_epochs): running_loss = 0.0 running_loss_count = 0 for batch_idx, (inputs, targets) in enumerate(dataloader): inputs = inputs.cuda(non_blocking=True) targets = targets.cuda(non_blocking=True) # Forward pass outputs = model(inputs) loss = loss_fn(outputs, targets) # Backward and optimize optimizer.zero_grad() loss.backward() optimizer.step() running_loss += loss.item() running_loss_count += 1 # Print loss statistics print( f"Epoch {epoch + 1}/{num_epochs}, " f"Loss: {running_loss / running_loss_count}" ) </code></pre></div> <hr /> <h2 id="singlehost-multidevice-synchronous-training">Single-host, multi-device synchronous training</h2> In this setup, you have one machine with several GPUs on it (typically 2 to 16). Each device will run a copy of your model (called a replica). For simplicity, in what follows, we'll assume we're dealing with 8 GPUs, at no loss of generality. How it works At each step of training: <ul> <li>The current batch of data (called global batch) is split into 8 different sub-batches (called local batches). For instance, if the global batch has 512 samples, each of the 8 local batches will have 64 samples.</li> <li>Each of the 8 replicas independently processes a local batch: they run a forward pass, then a backward pass, outputting the gradient of the weights with respect to the loss of the model on the local batch.</li> <li>The weight updates originating from local gradients are efficiently merged across the 8 replicas. Because this is done at the end of every step, the replicas always stay in sync.</li> </ul> In practice, the process of synchronously updating the weights of the model replicas is handled at the level of each individual weight variable. This is done through a mirrored variable object. How to use it To do single-host, multi-device synchronous training with a Keras model, you would use the <code>torch.nn.parallel.DistributedDataParallel</code> module wrapper. Here's how it works: <ul> <li>We use <code>torch.multiprocessing.start_processes</code> to start multiple Python processes, one per device. Each process will run the <code>per_device_launch_fn</code> function.</li> <li>The <code>per_device_launch_fn</code> function does the following: - It uses <code>torch.distributed.init_process_group</code> and <code>torch.cuda.set_device</code> to configure the device to be used for that process. - It uses <code>torch.utils.data.distributed.DistributedSampler</code> and <code>torch.utils.data.DataLoader</code> to turn our data into a distributed data loader. - It also uses <code>torch.nn.parallel.DistributedDataParallel</code> to turn our model into a distributed PyTorch module. - It then calls the <code>train_model</code> function.</li> <li>The <code>train_model</code> function will then run in each process, with the model using a separate device in each process.</li> </ul> Here's the flow, where each step is split into its own utility function: <div class="codehilite"><pre><code># Config num_gpu = torch.cuda.device_count() num_epochs = 2 batch_size = 64 print(f"Running on {num_gpu} GPUs") def setup_device(current_gpu_index, num_gpus): # Device setup os.environ["MASTER_ADDR"] = "localhost" os.environ["MASTER_PORT"] = "56492" device = torch.device("cuda:{}".format(current_gpu_index)) torch.distributed.init_process_group( backend="nccl", init_method="env://", world_size=num_gpus, rank=current_gpu_index, ) torch.cuda.set_device(device) def cleanup(): torch.distributed.destroy_process_group() def prepare_dataloader(dataset, current_gpu_index, num_gpus, batch_size): sampler = torch.utils.data.distributed.DistributedSampler( dataset, num_replicas=num_gpus, rank=current_gpu_index, shuffle=False, ) dataloader = torch.utils.data.DataLoader( dataset, sampler=sampler, batch_size=batch_size, shuffle=False, ) return dataloader def per_device_launch_fn(current_gpu_index, num_gpu): # Setup the process groups setup_device(current_gpu_index, num_gpu) dataset = get_dataset() model = get_model() # prepare the dataloader dataloader = prepare_dataloader(dataset, current_gpu_index, num_gpu, batch_size) # Instantiate the torch optimizer optimizer = torch.optim.Adam(model.parameters(), lr=1e-3) # Instantiate the torch loss function loss_fn = torch.nn.CrossEntropyLoss() # Put model on device model = model.to(current_gpu_index) ddp_model = torch.nn.parallel.DistributedDataParallel( model, device_ids=[current_gpu_index], output_device=current_gpu_index ) train_model(ddp_model, dataloader, num_epochs, optimizer, loss_fn) cleanup() </code></pre></div> <div class="k-default-codeblock"> <div class="codehilite"><pre><code>Running on 0 GPUs /opt/conda/envs/keras-torch/lib/python3.10/site-packages/torch/cuda/__init__.py:611: UserWarning: Can't initialize NVML warnings.warn("Can't initialize NVML") </code></pre></div> </div> Time to start multiple processes: <div class="codehilite"><pre><code>if __name__ == "__main__": # We use the "fork" method rather than "spawn" to support notebooks torch.multiprocessing.start_processes( per_device_launch_fn, args=(num_gpu,), nprocs=num_gpu, join=True, start_method="fork", ) </code></pre></div> That's it! </div> <div class='k-outline'> <div class='k-outline-depth-1'> <a href='#multigpu-distributed-training-with-pytorch'>Multi-GPU distributed training with PyTorch</a> </div> <div class='k-outline-depth-2'> ◆ <a href='#introduction'>Introduction</a> </div> <div class='k-outline-depth-2'> ◆ <a href='#setup'>Setup</a> </div> <div class='k-outline-depth-2'> ◆ <a href='#singlehost-multidevice-synchronous-training'>Single-host, multi-device synchronous training</a> </div> </div> </div> </div> </div> </body> <footer style="float: left; width: 100%; padding: 1em; border-top: solid 1px #bbb;"> <a href="https://policies.google.com/terms">Terms</a> | <a href="https://policies.google.com/privacy">Privacy</a> </footer> </html>

CINXE.COM

Multi-GPU distributed training with PyTorch