lightning/docs/source-pytorch/common_usecases.rst

161 lines
4.2 KiB
ReStructuredText

:orphan:
################
Common Workflows
################
Customize and extend Lightning for things like custom hardware or distributed strategies.
.. raw:: html
<div class="display-card-container">
<div class="row">
.. Add callout items below this line
.. displayitem::
:header: Avoid overfitting
:description: Add a training and test loop.
:col_css: col-md-12
:button_link: common/evaluation.html
:height: 100
.. displayitem::
:header: Build a model
:description: Steps to build a model.
:col_css: col-md-12
:button_link: model/build_model.html
:height: 100
.. displayitem::
:header: Configure hyperparameters from the CLI
:description: Enable basic CLI with Lightning.
:col_css: col-md-12
:button_link: common/hyperparameters.html
:height: 100
.. displayitem::
:header: Customize the progress bar
:description: Change the progress bar behavior.
:col_css: col-md-12
:button_link: common/progress_bar.html
:height: 100
.. displayitem::
:header: Deploy models into production
:description: Deploy models with different levels of scale.
:col_css: col-md-12
:button_link: deploy/production.html
:height: 100
.. displayitem::
:header: Effective Training Techniques
:description: Explore advanced training techniques.
:col_css: col-md-12
:button_link: advanced/training_tricks.html
:height: 100
.. displayitem::
:header: Eliminate config boilerplate
:description: Control your training via CLI and YAML.
:col_css: col-md-12
:button_link: cli/lightning_cli.html
:height: 100
.. displayitem::
:header: Find bottlenecks in your code
:description: Learn to find bottlenecks in your code.
:col_css: col-md-12
:button_link: tuning/profiler.html
:height: 100
.. displayitem::
:header: Finetune a model
:description: Learn to use pretrained models
:col_css: col-md-12
:button_link: advanced/transfer_learning.html
:height: 100
.. displayitem::
:header: Manage Experiments
:description: Learn to track and visualize experiments
:col_css: col-md-12
:button_link: visualize/logging_intermediate.html
:height: 100
.. displayitem::
:header: Run on a multi-node cluster
:description: Learn to run multi-node in the cloud or on your cluster
:col_css: col-md-12
:button_link: clouds/cluster.html
:height: 100
.. displayitem::
:header: Save and load model progress
:description: Save and load progress with checkpoints.
:col_css: col-md-12
:button_link: common/checkpointing_basic.html
:height: 100
.. displayitem::
:header: Save memory with half-precision
:description: Enable half-precision to train faster and save memory.
:col_css: col-md-12
:button_link: common/precision.html
:height: 100
.. displayitem::
:header: Train models with billions of parameters
:description: Scale GPU training to models with billions of parameters
:col_css: col-md-12
:button_link: advanced/model_parallel/index.html
:height: 100
.. displayitem::
:header: Train in a notebook
:description: Train models in interactive notebooks (Jupyter, Colab, Kaggle, etc.)
:col_css: col-md-12
:button_link: common/notebooks.html
:height: 100
.. displayitem::
:header: Train on single or multiple GPUs
:description: Train models faster with GPUs.
:col_css: col-md-12
:button_link: accelerators/gpu.html
:height: 100
.. displayitem::
:header: Train on single or multiple HPUs
:description: Train models faster with HPUs.
:col_css: col-md-12
:button_link: integrations/hpu/index.html
:height: 100
.. displayitem::
:header: Train on single or multiple TPUs
:description: Train models faster with TPUs.
:col_css: col-md-12
:button_link: accelerators/tpu.html
:height: 100
.. displayitem::
:header: Track and Visualize Experiments
:description: Learn to track and visualize experiments
:col_css: col-md-12
:button_link: visualize/logging_intermediate.html
:height: 100
.. displayitem::
:header: Use a pure PyTorch training loop
:description: Run your pure PyTorch loop with Lightning.
:col_css: col-md-12
:button_link: model/own_your_loop.html
:height: 100
.. raw:: html
</div>
</div>