Skip to content

Commit

Permalink
modified: cmhe/cmhe_torch.html
Browse files Browse the repository at this point in the history 
	modified:   cmhe/cmhe_utilities.html
	modified:   cmhe/index.html
  • Loading branch information
@chiragnagpal
chiragnagpal committed Feb 23, 2022
1 parent 6cfc9fa commit e6bbaec
Show file tree
Hide file tree
Showing 3 changed files with 158 additions and 26 deletions.
12 changes: 6 additions & 6 deletions docs/cmhe/cmhe_torch.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -30,12 +30,6 @@ <h1 class="title">Module <code>auton_survival.models.cmhe.cmhe_torch</code></h1>
<div class="git-link-div"><a href="https://github.com/autonlab/auton-survival" class="git-link">Browse git</a></div>
</section>
<section>
</section>
<section>
</section>
<section>
</section>
<section>
<h2 class="section-title" id="header-classes">Classes</h2>
<dl>
<dt id="auton_survival.models.cmhe.cmhe_torch.DeepCMHETorch"><code class="flex name class">
Expand Down Expand Up @@ -76,6 +70,12 @@ <h3>Methods</h3>
</dd>
</dl>
</section>
<section>
</section>
<section>
</section>
<section>
</section>
</article>
<nav id="sidebar">
<img src="https://ndownloader.figshare.com/files/34052981">
Expand Down
4 changes: 2 additions & 2 deletions docs/cmhe/cmhe_utilities.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -34,6 +34,8 @@ <h1 class="title">Module <code>auton_survival.models.cmhe.cmhe_utilities</code><
<section>
</section>
<section>
</section>
<section>
<h2 class="section-title" id="header-functions">Functions</h2>
<dl>
<dt id="auton_survival.models.cmhe.cmhe_utilities.randargmax"><code class="name flex">
Expand Down Expand Up @@ -199,8 +201,6 @@ <h2 class="section-title" id="header-functions">Functions</h2>
</dd>
</dl>
</section>
<section>
</section>
</article>
<nav id="sidebar">
<img src="https://ndownloader.figshare.com/files/34052981">
Expand Down
168 changes: 150 additions & 18 deletions docs/cmhe/index.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -36,40 +36,161 @@ <h2 id="cox-mixtures-with-heterogenous-effects">Cox Mixtures With Heterogenous E
<a href="https://opensource.org/licenses/MIT"><img alt="License: MIT" src="https://img.shields.io/badge/License-MIT-yellow.svg"></a>
&nbsp;&nbsp;&nbsp;
<a href="https://github.com/autonlab/auton-survival"><img alt="GitHub Repo stars" src="https://img.shields.io/github/stars/autonlab/auton-survival?style=social"></a></p>
<p><strong>Cox Mixture with Heterogenous Effects (CMHE)</strong> is a fully parametric approach
to counterfactual phenotypes of individuals that demonstrate heterogneous
effects to an intervention in terms of Time-to-Event outcomes in the presence
of Censoring.</p>
<p><img
src="https://ndownloader.figshare.com/files/34056269"></p>
<p><img align="right" width=35% src="https://figshare.com/ndownloader/files/34056284"></p>
<p>In the context of Healthcare ML and Biostatistics, this is known as 'Survival
Analysis'. The key idea behind Deep Survival Machines is to model the
underlying event outcome distribution as a mixure of some fixed <span><span class="MathJax_Preview"> k </span><script type="math/tex"> k </script></span>
parametric distributions. The parameters of these mixture distributions as
well as the mixing weights are modelled using Neural Networks.</p>
<p><br><br><br><br></p>
<p><strong>Cox Mixture with Heterogenous Effects (CMHE)</strong> is a flexible approach to
recover counterfactual phenotypes of individuals that demonstrate heterogneous
effects to an intervention in terms of censored Time-to-Event outcomes.
CMHE is <strong>not</strong> restricted by the strong Cox Proportional Hazards assumption
or any parametric assumption on the time to event distributions. CMHE achieves
this by describing each individual as belonging to two different latent groups,
<span><span class="MathJax_Preview"> \mathcal{Z} </span><script type="math/tex"> \mathcal{Z} </script></span> that mediate the base survival rate and <span><span class="MathJax_Preview"> \phi </span><script type="math/tex"> \phi </script></span> the effect
of the treatment. CMHE can also be employed to model individual level
counterfactuals or for standard factual survival regression.</p>
<p>For full details on Cox Mixtures with Heterogenous Effects, please refer to
our preprint:</p>
<p><a href="https://arxiv.org/abs/2202.11089">Counterfactual Phenotyping with Censored Time-to-Events, arXiv preprint,
C. Nagpal, M. Goswami, K. Dufendach, A. Dubrawski</a></p>
<p><br></p>
<h2 id="example-usage">Example Usage</h2>
<pre><code class="language-python-repl">&gt;&gt;&gt; from auton_survival import CoxMixtureHeterogenousEffects
<pre><code class="language-python-repl">&gt;&gt;&gt; from auton_survival import DeepCoxMixturesHeterogenousEffects
&gt;&gt;&gt; from auton_survival import datasets
&gt;&gt;&gt; # load the SYNTHETIC dataset.
&gt;&gt;&gt; x, t, e, a = datasets.load_dataset('SYNTHETIC')
&gt;&gt;&gt; # instantiate a DeepSurvivalMachines model.
&gt;&gt;&gt; model = CoxMixtureHeterogenousEffects()
&gt;&gt;&gt; # instantiate a Cox Mixtures with Heterogenous Effects model.
&gt;&gt;&gt; model = DeepCoxMixturesHeterogenousEffects()
&gt;&gt;&gt; # fit the model to the dataset.
&gt;&gt;&gt; model.fit(x, t, e, a)
&gt;&gt;&gt; # estimate the predicted risks at the time
&gt;&gt;&gt; model.predict_risk(x, 10)
&gt;&gt;&gt; # estimate the treatment effect phenogroups
&gt;&gt;&gt; model.predict_latent_phi(x)
</code></pre>
<div class="git-link-div"><a href="https://github.com/autonlab/auton-survival" class="git-link">Browse git</a></div>
</section>
<section>
<h2 class="section-title" id="header-submodules">Sub-modules</h2>
<h2 class="section-title" id="header-classes">Classes</h2>
<dl>
<dt id="auton_survival.models.cmhe.DeepCoxMixturesHeterogenousEffects"><code class="flex name class">
<span>class <span class="ident">DeepCoxMixturesHeterogenousEffects</span></span>
<span>(</span><span>k, g, layers=None)</span>
</code></dt>
<dd>
<div class="desc"><p>A Deep Cox Mixtures with Heterogenous Effects model.</p>
<p>This is the main interface to a Deep Cox Mixture with Heterogenous Effects.
A model is instantiated with approporiate set of hyperparameters and
fit on numpy arrays consisting of the features, event/censoring times
and the event/censoring indicators.</p>
<p>For full details on Deep Cox Mixture, refer to the paper [1].</p>
<h2 id="references">References</h2>
<p>[1] Nagpal, C., Goswami M., Dufendach K., and Artur Dubrawski.
"Counterfactual phenotyping for censored Time-to-Events" (2022).</p>
<h2 id="parameters">Parameters</h2>
<dl>
<dt><strong><code>k</code></strong> :&ensp;<code>int</code></dt>
<dd>The number of underlying base survival phenotypes.</dd>
<dt><strong><code>g</code></strong> :&ensp;<code>int</code></dt>
<dd>The number of underlying treatment effect phenotypes.</dd>
<dt><strong><code>layers</code></strong> :&ensp;<code>list</code></dt>
<dd>A list of integers consisting of the number of neurons in each
hidden layer.</dd>
</dl>
<h2 id="example">Example</h2>
<pre><code class="language-python-repl">&gt;&gt;&gt; from auton_survival import DeepCoxMixturesHeterogenousEffects
&gt;&gt;&gt; model = DeepCoxMixturesHeterogenousEffects(k=2, g=3)
&gt;&gt;&gt; model.fit(x, t, e, a)
</code></pre></div>
<div class="git-link-div"><a href="https://github.com/autonlab/auton-survival" class="git-link">Browse git</a></div>
<h3>Methods</h3>
<dl>
<dt><code class="name"><a title="auton_survival.models.cmhe.cmhe_api" href="cmhe_api.html">auton_survival.models.cmhe.cmhe_api</a></code></dt>
<dt id="auton_survival.models.cmhe.DeepCoxMixturesHeterogenousEffects.fit"><code class="name flex">
<span>def <span class="ident">fit</span></span>(<span>self, x, t, e, a, vsize=0.15, val_data=None, iters=1, learning_rate=0.001, batch_size=100, optimizer='Adam', random_state=100)</span>
</code></dt>
<dd>
<div class="desc"><p>This method is used to train an instance of the DSM model.</p>
<h2 id="parameters">Parameters</h2>
<dl>
<dt><strong><code>x</code></strong> :&ensp;<code>np.ndarray</code></dt>
<dd>A numpy array of the input features, <span><span class="MathJax_Preview"> x </span><script type="math/tex"> x </script></span>.</dd>
<dt><strong><code>t</code></strong> :&ensp;<code>np.ndarray</code></dt>
<dd>A numpy array of the event/censoring times, <span><span class="MathJax_Preview"> t </span><script type="math/tex"> t </script></span>.</dd>
<dt><strong><code>e</code></strong> :&ensp;<code>np.ndarray</code></dt>
<dd>A numpy array of the event/censoring indicators, <span><span class="MathJax_Preview"> \delta </span><script type="math/tex"> \delta </script></span>.
<span><span class="MathJax_Preview"> \delta = 1 </span><script type="math/tex"> \delta = 1 </script></span> means the event took place.</dd>
<dt><strong><code>a</code></strong> :&ensp;<code>np.ndarray</code></dt>
<dd>A numpy array of the treatment assignment indicators, <span><span class="MathJax_Preview"> a </span><script type="math/tex"> a </script></span>.
<span><span class="MathJax_Preview"> a = 1 </span><script type="math/tex"> a = 1 </script></span> means the individual was treated.</dd>
<dt><strong><code>vsize</code></strong> :&ensp;<code>float</code></dt>
<dd>Amount of data to set aside as the validation set.</dd>
<dt><strong><code>val_data</code></strong> :&ensp;<code>tuple</code></dt>
<dd>A tuple of the validation dataset. If passed vsize is ignored.</dd>
<dt><strong><code>iters</code></strong> :&ensp;<code>int</code></dt>
<dd>The maximum number of training iterations on the training dataset.</dd>
<dt><strong><code>learning_rate</code></strong> :&ensp;<code>float</code></dt>
<dd>The learning rate for the <code>Adam</code> optimizer.</dd>
<dt><strong><code>batch_size</code></strong> :&ensp;<code>int</code></dt>
<dd>learning is performed on mini-batches of input data. this parameter
specifies the size of each mini-batch.</dd>
<dt><strong><code>optimizer</code></strong> :&ensp;<code>str</code></dt>
<dd>The choice of the gradient based optimization method. One of
'Adam', 'RMSProp' or 'SGD'.</dd>
<dt><strong><code>random_state</code></strong> :&ensp;<code>float</code></dt>
<dd>random seed that determines how the validation set is chosen.</dd>
</dl></div>
<div class="git-link-div"><a href="https://github.com/autonlab/auton-survival" class="git-link">Browse git</a></div>
</dd>
<dt id="auton_survival.models.cmhe.DeepCoxMixturesHeterogenousEffects.predict_risk"><code class="name flex">
<span>def <span class="ident">predict_risk</span></span>(<span>self, x, a, t=None)</span>
</code></dt>
<dd>
<div class="desc"></div>
<div class="git-link-div"><a href="https://github.com/autonlab/auton-survival" class="git-link">Browse git</a></div>
</dd>
<dt id="auton_survival.models.cmhe.DeepCoxMixturesHeterogenousEffects.predict_survival"><code class="name flex">
<span>def <span class="ident">predict_survival</span></span>(<span>self, x, a, t=None)</span>
</code></dt>
<dd>
<div class="desc"><p>Returns the estimated survival probability at time <span><span class="MathJax_Preview"> t </span><script type="math/tex"> t </script></span>,
<span><span class="MathJax_Preview"> \widehat{\mathbb{P}}(T &gt; t|X) </span><script type="math/tex"> \widehat{\mathbb{P}}(T > t|X) </script></span> for some input data <span><span class="MathJax_Preview"> x </span><script type="math/tex"> x </script></span>.</p>
<h2 id="parameters">Parameters</h2>
<dl>
<dt><strong><code>x</code></strong> :&ensp;<code>np.ndarray</code></dt>
<dd>A numpy array of the input features, <span><span class="MathJax_Preview"> x </span><script type="math/tex"> x </script></span>.</dd>
<dt><strong><code>a</code></strong> :&ensp;<code>np.ndarray</code></dt>
<dd>A numpy array of the treatmeant assignment, <span><span class="MathJax_Preview"> a </span><script type="math/tex"> a </script></span>.</dd>
<dt><strong><code>t</code></strong> :&ensp;<code>list</code> or <code>float</code></dt>
<dd>a list or float of the times at which survival probability is
to be computed</dd>
</dl>
<h2 id="returns">Returns</h2>
<dl>
<dt><code>np.array</code></dt>
<dd>numpy array of the survival probabilites at each time in t.</dd>
</dl></div>
<div class="git-link-div"><a href="https://github.com/autonlab/auton-survival" class="git-link">Browse git</a></div>
</dd>
<dt id="auton_survival.models.cmhe.DeepCoxMixturesHeterogenousEffects.predict_latent_z"><code class="name flex">
<span>def <span class="ident">predict_latent_z</span></span>(<span>self, x)</span>
</code></dt>
<dd>
<div class="desc"><p>Returns the estimated latent base survival group <span><span class="MathJax_Preview"> z </span><script type="math/tex"> z </script></span> given the confounders <span><span class="MathJax_Preview"> x </span><script type="math/tex"> x </script></span>.</p></div>
<div class="git-link-div"><a href="https://github.com/autonlab/auton-survival" class="git-link">Browse git</a></div>
</dd>
<dt id="auton_survival.models.cmhe.DeepCoxMixturesHeterogenousEffects.predict_latent_phi"><code class="name flex">
<span>def <span class="ident">predict_latent_phi</span></span>(<span>self, x)</span>
</code></dt>
<dd>
<div class="desc"><p>Returns the estimated latent treatment effect group <span><span class="MathJax_Preview"> \phi </span><script type="math/tex"> \phi </script></span> given the confounders <span><span class="MathJax_Preview"> x </span><script type="math/tex"> x </script></span>.</p></div>
<div class="git-link-div"><a href="https://github.com/autonlab/auton-survival" class="git-link">Browse git</a></div>
</dd>
</dl>
</dd>
</dl>
</section>
<section>
<h2 class="section-title" id="header-submodules">Sub-modules</h2>
<dl>
<dt><code class="name"><a title="auton_survival.models.cmhe.cmhe_torch" href="cmhe_torch.html">auton_survival.models.cmhe.cmhe_torch</a></code></dt>
<dd>
<div class="desc"></div>
Expand All @@ -84,8 +205,6 @@ <h2 class="section-title" id="header-submodules">Sub-modules</h2>
</section>
<section>
</section>
<section>
</section>
</article>
<nav id="sidebar">
<img src="https://ndownloader.figshare.com/files/34052981">
Expand All @@ -104,11 +223,24 @@ <h1>Index</h1>
</li>
<li><h3><a href="#header-submodules">Sub-modules</a></h3>
<ul>
<li><code><a title="auton_survival.models.cmhe.cmhe_api" href="cmhe_api.html">auton_survival.models.cmhe.cmhe_api</a></code></li>
<li><code><a title="auton_survival.models.cmhe.cmhe_torch" href="cmhe_torch.html">auton_survival.models.cmhe.cmhe_torch</a></code></li>
<li><code><a title="auton_survival.models.cmhe.cmhe_utilities" href="cmhe_utilities.html">auton_survival.models.cmhe.cmhe_utilities</a></code></li>
</ul>
</li>
<li><h3><a href="#header-classes">Classes</a></h3>
<ul>
<li>
<h4><code><a title="auton_survival.models.cmhe.DeepCoxMixturesHeterogenousEffects" href="#auton_survival.models.cmhe.DeepCoxMixturesHeterogenousEffects">DeepCoxMixturesHeterogenousEffects</a></code></h4>
<ul class="">
<li><code><a title="auton_survival.models.cmhe.DeepCoxMixturesHeterogenousEffects.fit" href="#auton_survival.models.cmhe.DeepCoxMixturesHeterogenousEffects.fit">fit</a></code></li>
<li><code><a title="auton_survival.models.cmhe.DeepCoxMixturesHeterogenousEffects.predict_risk" href="#auton_survival.models.cmhe.DeepCoxMixturesHeterogenousEffects.predict_risk">predict_risk</a></code></li>
<li><code><a title="auton_survival.models.cmhe.DeepCoxMixturesHeterogenousEffects.predict_survival" href="#auton_survival.models.cmhe.DeepCoxMixturesHeterogenousEffects.predict_survival">predict_survival</a></code></li>
<li><code><a title="auton_survival.models.cmhe.DeepCoxMixturesHeterogenousEffects.predict_latent_z" href="#auton_survival.models.cmhe.DeepCoxMixturesHeterogenousEffects.predict_latent_z">predict_latent_z</a></code></li>
<li><code><a title="auton_survival.models.cmhe.DeepCoxMixturesHeterogenousEffects.predict_latent_phi" href="#auton_survival.models.cmhe.DeepCoxMixturesHeterogenousEffects.predict_latent_phi">predict_latent_phi</a></code></li>
</ul>
</li>
</ul>
</li>
</ul>
</nav>
</main>
Expand Down

0 comments on commit e6bbaec

Please sign in to comment.