deploy: cleanlab/cleanlab@8bebf6f

Author: cgnorthcutt

Diff for: .nojekyll

@@ -0,0 +1,4 @@
+# Sphinx build info version 1
+# This file hashes the configuration used when building these files. When it is not found, a full rebuild will be done.
+config: 95645162aad93a65d770e04aecce6f55
+tags: 645f666f9bcd5a90fca523b33c5a78b7

Diff for: master/.buildinfo

@@ -0,0 +1,19 @@
+:hide-toc:
+
+CIFAR CNN
+=========
+
+.. automodule:: cleanlab.models.cifar_cnn
+   :autosummary:
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+MNIST PyTorch
+=============
+
+.. automodule:: cleanlab.models.mnist_pytorch
+   :autosummary:
+   :members:
+   :undoc-members:
+   :show-inheritance:

Diff for: master/.doctrees/cleanlab.doctree

@@ -0,0 +1,8 @@
+CIFAR CNN
+=========
+
+.. automodule:: cleanlab.models.cifar_cnn
+   :autosummary:
+   :members:
+   :undoc-members:
+   :show-inheritance:

Diff for: master/.doctrees/cleanlab.models.doctree

@@ -0,0 +1,4 @@
+.. toctree::
+
+   cifar_cnn
+   mnist_pytorch

Diff for: master/.doctrees/cleanlab.models/cifar_cnn.doctree

@@ -0,0 +1,8 @@
+MNIST PyTorch
+=============
+
+.. automodule:: cleanlab.models.mnist_pytorch
+   :autosummary:
+   :members:
+   :undoc-members:
+   :show-inheritance:

Diff for: master/.doctrees/cleanlab.models/index.doctree

@@ -0,0 +1,90 @@
+:hide-toc:
+
+Classification
+==============
+
+.. automodule:: cleanlab.classification
+   :autosummary:
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+Latent Estimation
+=================
+
+.. automodule:: cleanlab.latent_estimation
+   :autosummary:
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+Noise Generation
+================
+
+.. automodule:: cleanlab.noise_generation
+   :autosummary:
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+Baseline Methods
+================
+
+.. automodule:: cleanlab.baseline_methods
+   :autosummary:
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+Co-Teaching
+===========
+
+.. automodule:: cleanlab.coteaching
+   :autosummary:
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+Latent Algebra
+==============
+
+.. automodule:: cleanlab.latent_algebra
+   :autosummary:
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+Pruning
+=======
+
+.. automodule:: cleanlab.pruning
+   :autosummary:
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+Utilities
+=========
+
+.. automodule:: cleanlab.util
+   :autosummary:
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+Polyplex
+========
+
+.. automodule:: cleanlab.polyplex
+   :autosummary:
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+Models
+======
+
+.. toctree::
+
+   cleanlab.models
+

Diff for: master/.doctrees/cleanlab.models/mnist_pytorch.doctree

@@ -0,0 +1,83 @@
+Introduction
+============
+
+**cleanlab automatically finds and fixes errors in your ML datasets.**
+
+| This reduces manual work needed to fix data issues and helps train reliable ML models on partially mislabeled datasets. ``cleanlab`` has already found thousands of `label errors <https://labelerrors.com>`_ in ImageNet, MNIST, and other popular ML benchmarking datasets, so let's get started with yours!
+
+Quickstart
+==========
+
+1. Install ``cleanlab``.
+------------------------
+
+.. tabs::
+
+   .. tab:: pip
+
+      .. code-block:: python
+
+         pip install cleanlab
+
+   .. tab:: conda
+
+      .. code-block:: python
+
+         conda install -c conda-forge cleanlab
+
+   .. tab:: source
+
+      .. code-block:: python
+
+         pip install git+https://github.com/cleanlab/cleanlab.git
+
+
+2. Find label errors with ``get_noise_indices``.
+------------------------------------------------
+
+``cleanlab``'s ``get_noise_indices`` function tells you which examples in your dataset are likely mislabeled. At a minimum, it expects two inputs - your data's given labels, ``y``, and predicted probabilities, ``pyx``, from some trained model (Note: these must be out-of-sample predictions where the data points were held out from the model during training, which can be obtained via cross-validation). 
+
+Setting ``sorted_index_method`` instructs ``cleanlab`` to return the indices of potential mislabeled examples, ordered by the likelihood of label error estimate via ``prob_given_label`` scores (predicted probability of given label according to the model).
+
+.. code-block:: python
+
+   from cleanlab.pruning import get_noise_indices
+
+   ordered_label_errors = get_noise_indices(
+      s=y, 
+      psx=pyx,
+      sorted_index_method='prob_given_label')
+
+.. important::
+   The predicted probabilities, ``pyx``, from your model **must be out-of-sample**! You should never provide predictions on the same data points used to train the model - this would reflect predictions of an overfitted model, making it unsuitable for finding label errors. To compute the out-of-sample predicted probabilities of the entire dataset, you can use cross-validation.
+
+..
+   todo - include the url for tf and torch beginner tutorials
+
+3. Train robust models with noisy labels using ``LearningWithNoisyLabels``.
+---------------------------------------------------------------------------
+
+``cleanlab``'s ``LearningWithNoisyLabels`` adapts any classification model, ``clf``, to a more reliable one by allowing it to train directly on partially mislabeled datasets. 
+
+When the ``.fit()`` method is called, it automatically identifies and removes any examples that are deemed "noisy" in the provided dataset before returning a final trained model.
+
+.. code-block:: python
+
+   from sklearn.linear_model import LogisticRegression
+   from cleanlab.classification import LearningWithNoisyLabels
+
+   clf = LogisticRegression() # Here we've used sklearn's Logistic Regression model, but this can be any classifier that implements sklearn's API.
+   lnl = LearningWithNoisyLabels(clf=clf)
+   lnl.fit(X=X, s=y)
+
+.. toctree::
+   :hidden:
+   :caption: Get Started
+
+   Quickstart <self>
+
+.. toctree::
+   :caption: API Reference
+   :hidden:
+
+   cleanlab