danjust published a site update

_modules/besskge/utils.html

@@ -140,7 +140,7 @@ <h1>Source code for besskge.utils</h1><div class="highlight"><pre>
 <span class="sd">        Row-wise rotated tensors.</span>
 <span class="sd">    &quot;&quot;&quot;</span>
     <span class="c1"># Always compute sin and cos in fp16, as faster on IPU</span>
-    <span class="k">if</span> <span class="n">r</span><span class="o">.</span><span class="n">dtype</span> <span class="o">==</span> <span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">:</span>
+    <span class="k">if</span> <span class="n">r</span><span class="o">.</span><span class="n">dtype</span> <span class="o">==</span> <span class="n">torch</span><span class="o">.</span><span class="n">float32</span> <span class="ow">and</span> <span class="n">r</span><span class="o">.</span><span class="n">device</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s2">&quot;ipu&quot;</span><span class="p">:</span>
         <span class="n">r_cos</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cos</span><span class="p">(</span><span class="n">r</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float16</span><span class="p">))</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
         <span class="n">r_sin</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">sin</span><span class="p">(</span><span class="n">r</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float16</span><span class="p">))</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
     <span class="k">else</span><span class="p">:</span>

generated/besskge.bess.TopKQueryBessKGE.html

@@ -119,17 +119,17 @@ <h1>besskge.bess.TopKQueryBessKGE<a class="headerlink" href="#besskge-bess-topkq
 <dl class="field-list simple">
 <dt class="field-odd">Parameters<span class="colon">:</span></dt>
 <dd class="field-odd"><ul class="simple">
-<li><p><strong>k</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.11)"><code class="xref py py-class docutils literal notranslate"><span class="pre">int</span></code></a>) – For each query return the top-k most likely predictions.</p></li>
-<li><p><strong>candidate_sampler</strong> (<a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.11)"><code class="xref py py-data docutils literal notranslate"><span class="pre">Union</span></code></a>[<a class="reference internal" href="besskge.negative_sampler.TripleBasedShardedNegativeSampler.html#besskge.negative_sampler.TripleBasedShardedNegativeSampler" title="besskge.negative_sampler.TripleBasedShardedNegativeSampler"><code class="xref py py-class docutils literal notranslate"><span class="pre">TripleBasedShardedNegativeSampler</span></code></a>, <a class="reference internal" href="besskge.negative_sampler.PlaceholderNegativeSampler.html#besskge.negative_sampler.PlaceholderNegativeSampler" title="besskge.negative_sampler.PlaceholderNegativeSampler"><code class="xref py py-class docutils literal notranslate"><span class="pre">PlaceholderNegativeSampler</span></code></a>]) – Sampler of candidate entities to score against queries.
+<li><p><strong>k</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><code class="xref py py-class docutils literal notranslate"><span class="pre">int</span></code></a>) – For each query return the top-k most likely predictions.</p></li>
+<li><p><strong>candidate_sampler</strong> (<a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Union" title="(in Python v3.12)"><code class="xref py py-data docutils literal notranslate"><span class="pre">Union</span></code></a>[<a class="reference internal" href="besskge.negative_sampler.TripleBasedShardedNegativeSampler.html#besskge.negative_sampler.TripleBasedShardedNegativeSampler" title="besskge.negative_sampler.TripleBasedShardedNegativeSampler"><code class="xref py py-class docutils literal notranslate"><span class="pre">TripleBasedShardedNegativeSampler</span></code></a>, <a class="reference internal" href="besskge.negative_sampler.PlaceholderNegativeSampler.html#besskge.negative_sampler.PlaceholderNegativeSampler" title="besskge.negative_sampler.PlaceholderNegativeSampler"><code class="xref py py-class docutils literal notranslate"><span class="pre">PlaceholderNegativeSampler</span></code></a>]) – Sampler of candidate entities to score against queries.
 Use <a class="reference internal" href="besskge.negative_sampler.PlaceholderNegativeSampler.html#besskge.negative_sampler.PlaceholderNegativeSampler" title="besskge.negative_sampler.PlaceholderNegativeSampler"><code class="xref py py-class docutils literal notranslate"><span class="pre">besskge.negative_sampler.PlaceholderNegativeSampler</span></code></a>
 to score queries against all entities in the knowledge graph, avoiding
 unnecessary loading of negative entities on device.</p></li>
 <li><p><strong>score_fn</strong> (<a class="reference internal" href="besskge.scoring.BaseScoreFunction.html#besskge.scoring.BaseScoreFunction" title="besskge.scoring.BaseScoreFunction"><code class="xref py py-class docutils literal notranslate"><span class="pre">BaseScoreFunction</span></code></a>) – Scoring function.</p></li>
-<li><p><strong>evaluation</strong> (<a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.11)"><code class="xref py py-data docutils literal notranslate"><span class="pre">Optional</span></code></a>[<a class="reference internal" href="besskge.metric.Evaluation.html#besskge.metric.Evaluation" title="besskge.metric.Evaluation"><code class="xref py py-class docutils literal notranslate"><span class="pre">Evaluation</span></code></a>]) – Evaluation module, for computing metrics on device.
+<li><p><strong>evaluation</strong> (<a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><code class="xref py py-data docutils literal notranslate"><span class="pre">Optional</span></code></a>[<a class="reference internal" href="besskge.metric.Evaluation.html#besskge.metric.Evaluation" title="besskge.metric.Evaluation"><code class="xref py py-class docutils literal notranslate"><span class="pre">Evaluation</span></code></a>]) – Evaluation module, for computing metrics on device.
 Default: None.</p></li>
-<li><p><strong>return_scores</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.11)"><code class="xref py py-class docutils literal notranslate"><span class="pre">bool</span></code></a>) – If True, return scores of the top-k best completions.
+<li><p><strong>return_scores</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#bool" title="(in Python v3.12)"><code class="xref py py-class docutils literal notranslate"><span class="pre">bool</span></code></a>) – If True, return scores of the top-k best completions.
 Default: False.</p></li>
-<li><p><strong>window_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.11)"><code class="xref py py-class docutils literal notranslate"><span class="pre">int</span></code></a>) – Size of the sliding window, namely the number of negative entities
+<li><p><strong>window_size</strong> (<a class="reference external" href="https://docs.python.org/3/library/functions.html#int" title="(in Python v3.12)"><code class="xref py py-class docutils literal notranslate"><span class="pre">int</span></code></a>) – Size of the sliding window, namely the number of negative entities
 scored against each query at each step of the on-device for-loop.
 Should be decreased with large batch sizes, to avoid an OOM error.
 Default: 100.</p></li>
@@ -149,22 +149,22 @@ <h1>besskge.bess.TopKQueryBessKGE<a class="headerlink" href="#besskge-bess-topkq
 <dl class="field-list simple">
 <dt class="field-odd">Parameters<span class="colon">:</span></dt>
 <dd class="field-odd"><ul class="simple">
-<li><p><strong>relation</strong> (<a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">Tensor</span></code></a>) – shape: (shard_bs,)
+<li><p><strong>relation</strong> (<a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.1)"><code class="xref py py-class docutils literal notranslate"><span class="pre">Tensor</span></code></a>) – shape: (shard_bs,)
 Relation indices.</p></li>
-<li><p><strong>head</strong> (<a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.11)"><code class="xref py py-data docutils literal notranslate"><span class="pre">Optional</span></code></a>[<a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">Tensor</span></code></a>]) – shape: (shard_bs,)
+<li><p><strong>head</strong> (<a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><code class="xref py py-data docutils literal notranslate"><span class="pre">Optional</span></code></a>[<a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.1)"><code class="xref py py-class docutils literal notranslate"><span class="pre">Tensor</span></code></a>]) – shape: (shard_bs,)
 Head indices, if known. Default: None.</p></li>
-<li><p><strong>tail</strong> (<a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.11)"><code class="xref py py-data docutils literal notranslate"><span class="pre">Optional</span></code></a>[<a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">Tensor</span></code></a>]) – shape: (shard_bs,)
+<li><p><strong>tail</strong> (<a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><code class="xref py py-data docutils literal notranslate"><span class="pre">Optional</span></code></a>[<a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.1)"><code class="xref py py-class docutils literal notranslate"><span class="pre">Tensor</span></code></a>]) – shape: (shard_bs,)
 Tail indices, if known. Default: None.</p></li>
-<li><p><strong>negative</strong> (<a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.11)"><code class="xref py py-data docutils literal notranslate"><span class="pre">Optional</span></code></a>[<a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">Tensor</span></code></a>]) – shape: (n_shard, B, padded_negative)
+<li><p><strong>negative</strong> (<a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><code class="xref py py-data docutils literal notranslate"><span class="pre">Optional</span></code></a>[<a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.1)"><code class="xref py py-class docutils literal notranslate"><span class="pre">Tensor</span></code></a>]) – shape: (n_shard, B, padded_negative)
 Candidates to score against the queries.
 It can be the same set for all queries (B=1),
 or specific for each query in the batch (B=shard_bs).
 If None, score each query against all entities in the knowledge
 graph. Default: None.</p></li>
-<li><p><strong>triple_mask</strong> (<a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.11)"><code class="xref py py-data docutils literal notranslate"><span class="pre">Optional</span></code></a>[<a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">Tensor</span></code></a>]) – shape: (shard_bs,)
+<li><p><strong>triple_mask</strong> (<a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><code class="xref py py-data docutils literal notranslate"><span class="pre">Optional</span></code></a>[<a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.1)"><code class="xref py py-class docutils literal notranslate"><span class="pre">Tensor</span></code></a>]) – shape: (shard_bs,)
 Mask to filter the triples in the micro-batch
 before computing metrics. Default: None.</p></li>
-<li><p><strong>negative_mask</strong> (<a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.11)"><code class="xref py py-data docutils literal notranslate"><span class="pre">Optional</span></code></a>[<a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.0)"><code class="xref py py-class docutils literal notranslate"><span class="pre">Tensor</span></code></a>]) – shape: (n_shard, B, padded_negative)
+<li><p><strong>negative_mask</strong> (<a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Optional" title="(in Python v3.12)"><code class="xref py py-data docutils literal notranslate"><span class="pre">Optional</span></code></a>[<a class="reference external" href="https://pytorch.org/docs/stable/tensors.html#torch.Tensor" title="(in PyTorch v2.1)"><code class="xref py py-class docutils literal notranslate"><span class="pre">Tensor</span></code></a>]) – shape: (n_shard, B, padded_negative)
 If candidates are provided, mask to discard padding
 negatives when computing best completions.
 Requires the use of <code class="code docutils literal notranslate"><span class="pre">mask_on_gather=True</span></code> in the candidate
@@ -174,7 +174,7 @@ <h1>besskge.bess.TopKQueryBessKGE<a class="headerlink" href="#besskge-bess-topkq
 </ul>
 </dd>
 <dt class="field-even">Return type<span class="colon">:</span></dt>
-<dd class="field-even"><p><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.11)"><code class="xref py py-class docutils literal notranslate"><span class="pre">Dict</span></code></a>[<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.11)"><code class="xref py py-class docutils literal notranslate"><span class="pre">str</span></code></a>, <a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.11)"><code class="xref py py-data docutils literal notranslate"><span class="pre">Any</span></code></a>]</p>
+<dd class="field-even"><p><a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Dict" title="(in Python v3.12)"><code class="xref py py-class docutils literal notranslate"><span class="pre">Dict</span></code></a>[<a class="reference external" href="https://docs.python.org/3/library/stdtypes.html#str" title="(in Python v3.12)"><code class="xref py py-class docutils literal notranslate"><span class="pre">str</span></code></a>, <a class="reference external" href="https://docs.python.org/3/library/typing.html#typing.Any" title="(in Python v3.12)"><code class="xref py py-data docutils literal notranslate"><span class="pre">Any</span></code></a>]</p>
 </dd>
 </dl>
 </dd></dl>