index.rst

Welcome to pytorch-lantern's documentation!

Lantern contains our process of bringing a project to fruition as efficiently as possible. This is subject to change as we iterate and improve. This package implements tools and missing features to help bridge the gap between frameworks and libraries that we utilize.

The main packages and tools that we build around are:

• pytorch
• pytorch-datastream
• guild

.. autoclass:: lantern.EarlyStopping
   :members:
   :undoc-members:
   :member-order: bysource

.. autoclass:: lantern.ProgressBar
   :members:
   :undoc-members:
   :member-order: bysource

.. autoclass:: lantern.MetricTable
   :members:
   :undoc-members:
   :member-order: bysource

.. autoclass:: lantern.MapMetric
   :members:
   :undoc-members:
   :member-order: bysource

.. autoclass:: lantern.ReduceMetric
   :members:
   :undoc-members:
   :member-order: bysource

.. autoclass:: lantern.FunctionalBase
   :members:
   :undoc-members:
   :member-order: bysource

.. autoclass:: lantern.Tensor
   :members:
   :undoc-members:
   :member-order: bysource

.. autoclass:: lantern.Numpy
   :members:
   :undoc-members:
   :member-order: bysource

.. autofunction:: lantern.Epochs

.. autofunction:: lantern.module_device

.. autofunction:: lantern.set_seeds

.. autofunction:: lantern.module_train

.. autofunction:: lantern.module_eval

.. autofunction:: lantern.worker_init_fn

.. autofunction:: lantern.set_learning_rate

.. autofunction:: lantern.star

.. autofunction:: lantern.numpy_from_matplotlib_figure

.. autofunction:: lantern.numpy_seed

.. autofunction:: lantern.git_info