deploy: SimplexLab/TorchJD@3e3e0e6

Author: ValerianRey

latest/docs/aggregation/bases/index.html

@@ -242,7 +242,7 @@
 <h1>Aggregator (abstract)<a class="headerlink" href="#aggregator-abstract" title="Link to this heading">¶</a></h1>
 <dl class="py class">
 <dt class="sig sig-object py" id="torchjd.aggregation.Aggregator">
-<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchjd.aggregation.</span></span><span class="sig-name descname"><span class="pre">Aggregator</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference external" href="https://github.com/TorchJD/torchjd/blob/main/src/torchjd/aggregation/_aggregator_bases.py#L9-L45"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchjd.aggregation.Aggregator" title="Link to this definition">¶</a></dt>
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchjd.aggregation.</span></span><span class="sig-name descname"><span class="pre">Aggregator</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference external" href="https://github.com/TorchJD/torchjd/blob/main/src/torchjd/aggregation/_aggregator_bases.py#L9-L37"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchjd.aggregation.Aggregator" title="Link to this definition">¶</a></dt>
 <dd><p>Bases: <a class="reference external" href="https://docs.pytorch.org/docs/stable/generated/torch.nn.Module.html#torch.nn.Module" title="(in PyTorch v2.7)"><code class="xref py py-class docutils literal notranslate"><span class="pre">Module</span></code></a>, <a class="reference external" href="https://docs.python.org/3/library/abc.html#abc.ABC" title="(in Python v3.13)"><code class="xref py py-class docutils literal notranslate"><span class="pre">ABC</span></code></a></p>
 <p>Abstract base class for all aggregators. It has the role of aggregating matrices of dimension
 <span class="math notranslate nohighlight">\(m \times n\)</span> into row vectors of dimension <span class="math notranslate nohighlight">\(n\)</span>.</p>

latest/docs/aggregation/graddrop/index.html

@@ -242,7 +242,7 @@
 <h1>GradDrop<a class="headerlink" href="#graddrop" title="Link to this heading">¶</a></h1>
 <dl class="py class">
 <dt class="sig sig-object py" id="torchjd.aggregation.GradDrop">
-<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchjd.aggregation.</span></span><span class="sig-name descname"><span class="pre">GradDrop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">f=&lt;function</span> <span class="pre">_identity&gt;</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">leak=None</span></span></em><span class="sig-paren">)</span><a class="reference external" href="https://github.com/TorchJD/torchjd/blob/main/src/torchjd/aggregation/_graddrop.py#L14-L92"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchjd.aggregation.GradDrop" title="Link to this definition">¶</a></dt>
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchjd.aggregation.</span></span><span class="sig-name descname"><span class="pre">GradDrop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">f=&lt;function</span> <span class="pre">_identity&gt;</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">leak=None</span></span></em><span class="sig-paren">)</span><a class="reference external" href="https://github.com/TorchJD/torchjd/blob/main/src/torchjd/aggregation/_graddrop.py#L14-L91"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchjd.aggregation.GradDrop" title="Link to this definition">¶</a></dt>
 <dd><p><a class="reference internal" href="../bases/#torchjd.aggregation.Aggregator" title="torchjd.aggregation._aggregator_bases.Aggregator"><code class="xref py py-class docutils literal notranslate"><span class="pre">Aggregator</span></code></a> that applies the gradient combination
 steps from GradDrop, as defined in lines 10 to 15 of Algorithm 1 of <a class="reference external" href="https://arxiv.org/pdf/2010.06808.pdf">Just Pick a Sign:
 Optimizing Deep Multitask Models with Gradient Sign Dropout</a>.</p>

latest/docs/aggregation/trimmed_mean/index.html

@@ -242,7 +242,7 @@
 <h1>Trimmed Mean<a class="headerlink" href="#trimmed-mean" title="Link to this heading">¶</a></h1>
 <dl class="py class">
 <dt class="sig sig-object py" id="torchjd.aggregation.TrimmedMean">
-<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchjd.aggregation.</span></span><span class="sig-name descname"><span class="pre">TrimmedMean</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">trim_number</span></span></em><span class="sig-paren">)</span><a class="reference external" href="https://github.com/TorchJD/torchjd/blob/main/src/torchjd/aggregation/_trimmed_mean.py#L7-L73"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchjd.aggregation.TrimmedMean" title="Link to this definition">¶</a></dt>
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">torchjd.aggregation.</span></span><span class="sig-name descname"><span class="pre">TrimmedMean</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">trim_number</span></span></em><span class="sig-paren">)</span><a class="reference external" href="https://github.com/TorchJD/torchjd/blob/main/src/torchjd/aggregation/_trimmed_mean.py#L7-L72"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#torchjd.aggregation.TrimmedMean" title="Link to this definition">¶</a></dt>
 <dd><p><a class="reference internal" href="../bases/#torchjd.aggregation.Aggregator" title="torchjd.aggregation._aggregator_bases.Aggregator"><code class="xref py py-class docutils literal notranslate"><span class="pre">Aggregator</span></code></a> for adversarial federated learning,
 that trims the most extreme values of the input matrix, before averaging its rows, as defined in
 <a class="reference external" href="https://proceedings.mlr.press/v80/yin18a/yin18a.pdf">Byzantine-Robust Distributed Learning: Towards Optimal Statistical Rates</a>.</p>