Documentation

Finding API Help

Every framework has thousands of functions and classes. You won’t memorize them — you’ll look them up.

Two Python builtins do most of the work:

  • dir(module) — what’s in here?
  • help(thing) (or ?thing in Jupyter) — how do I use it?

Plus the official docs: pytorch.org, jax.dev, tensorflow.org, mxnet.apache.org.

dir: discovering the API

Standard import:

import tensorflow as tf

dir(...) lists names in a module. Filter private names and show a small prefix on slides; in a notebook you can inspect the full list interactively:

print([name for name in dir(tf.random) if not name.startswith('_')][:20])
['Algorithm', 'Generator', 'all_candidate_sampler', 'categorical', 'create_rng_state', 'experimental', 'fixed_unigram_candidate_sampler', 'fold_in', ...

help: usage details

Once you have the name, help(...) prints the docstring with arguments, defaults, and a usage example:

help(tf.ones)
Help on function ones in module tensorflow.python.ops.array_ops:

ones(shape, dtype=tf.float32, name=None, layout=None)
    Creates a tensor with all elements set to one (1).

    See also `tf.ones_like`, `tf.zeros`, `tf.fill`, `tf.eye`.
...
      layout: Optional, `tf.experimental.dtensor.Layout`. If provided, the result
        is a [DTensor](https://www.tensorflow.org/guide/dtensor_overview) with the
        provided layout.

    Returns:
      A `Tensor` with all elements set to one (1).

Then run a one-liner to confirm the call:

tf.ones(4)
<tf.Tensor: shape=(4,), dtype=float32, numpy=array([1., 1., 1., 1.], dtype=float32)>

Recap

  • dir(module) — list contents.
  • help(symbol) (or symbol? in Jupyter) — show the docstring.
  • Notebook autocomplete (Tab) is your fastest discovery tool.
  • For prose-heavy explanations, deep links into the framework’s official documentation beat the inline help.