Deploying to gh-pages from @ 5d9d660 🚀

Advantages_Data_Types.html

@@ -15151,10 +15151,10 @@ <h1 id="Why-%CE%A6ML-has-Precision-Management">Why &#934;<sub>ML</sub> has Preci
 
 
 <div class="jp-RenderedText jp-OutputArea-output" data-mime-type="application/vnd.jupyter.stderr">
-<pre>2024-08-03 18:17:24.645775: I tensorflow/tsl/cuda/cudart_stub.cc:28] Could not find cuda drivers on your machine, GPU will not be used.
-2024-08-03 18:17:24.683566: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
+<pre>2024-08-03 18:51:20.373509: I tensorflow/tsl/cuda/cudart_stub.cc:28] Could not find cuda drivers on your machine, GPU will not be used.
+2024-08-03 18:51:20.411625: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
 To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
-2024-08-03 18:17:25.451024: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
+2024-08-03 18:51:21.176404: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
 </pre>
 </div>
 </div>

Convert.html

@@ -15617,7 +15617,7 @@ <h2 id="Converting-Tensors">Converting Tensors<a class="anchor-link" href="#Conv
 
 
 <div class="jp-RenderedText jp-OutputArea-output" data-mime-type="application/vnd.jupyter.stderr">
-<pre>2024-08-03 18:17:37.023984: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
+<pre>2024-08-03 18:51:32.753942: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
 </pre>
 </div>
 </div>

Examples.html

@@ -15215,7 +15215,7 @@ <h3 id="Training-an-MLP">Training an MLP<a class="anchor-link" href="#Training-a
 
 
 <div class="jp-RenderedText jp-OutputArea-output" data-mime-type="application/vnd.jupyter.stderr">
-<pre>2024-08-03 18:17:52.619976: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
+<pre>2024-08-03 18:51:48.070782: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
 </pre>
 </div>
 </div>

Introduction.html

@@ -15349,7 +15349,7 @@ <h2 id="Usage-without-%CE%A6ML's-Tensors">Usage without &#934;<sub>ML</sub>'s Te
 
 
 <div class="jp-RenderedText jp-OutputArea-output" data-mime-type="application/vnd.jupyter.stderr">
-<pre>2024-08-03 18:18:09.812794: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
+<pre>2024-08-03 18:52:04.768523: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
 </pre>
 </div>
 </div>

Linear_Solves.html

@@ -16021,13 +16021,13 @@ <h2 id="Obtaining-Additional-Information-about-a-Solve">Obtaining Additional Inf
 
 
 <div class="jp-RenderedText jp-OutputArea-output" data-mime-type="text/plain">
-<pre>factor_ilu: auto-selecting iterations=1 (eager mode) for matrix <span class="ansi-blue-intense-fg">(2.000, 0.000)</span>; <span class="ansi-blue-intense-fg">(0.000, 1.000)</span> <span class="ansi-green-intense-fg">(b_vecᶜ=2, ~b_vecᵈ=2)</span> (DEBUG), 2024-08-03 18:18:25,561n
+<pre>factor_ilu: auto-selecting iterations=1 (eager mode) for matrix <span class="ansi-blue-intense-fg">(2.000, 0.000)</span>; <span class="ansi-blue-intense-fg">(0.000, 1.000)</span> <span class="ansi-green-intense-fg">(b_vecᶜ=2, ~b_vecᵈ=2)</span> (DEBUG), 2024-08-03 18:52:19,958n
 
-TorchScript -&gt; run compiled forward &#39;_matrix_solve_forward&#39; with args [((2,), False)] (DEBUG), 2024-08-03 18:18:25,576n
+TorchScript -&gt; run compiled forward &#39;_matrix_solve_forward&#39; with args [((2,), False)] (DEBUG), 2024-08-03 18:52:19,973n
 
-Running forward pass of custom op forward &#39;_matrix_solve_forward&#39; given args (&#39;y&#39;,) containing 1 native tensors (DEBUG), 2024-08-03 18:18:25,576n
+Running forward pass of custom op forward &#39;_matrix_solve_forward&#39; given args (&#39;y&#39;,) containing 1 native tensors (DEBUG), 2024-08-03 18:52:19,973n
 
-Performing linear solve scipy-CG with tolerance <span class="ansi-yellow-intense-fg">float64</span> <span class="ansi-blue-intense-fg">1e-05</span> (rel), <span class="ansi-yellow-intense-fg">float64</span> <span class="ansi-blue-intense-fg">1e-05</span> (abs), max_iterations=<span class="ansi-blue-intense-fg">1000</span> with backend torch (DEBUG), 2024-08-03 18:18:25,580n
+Performing linear solve scipy-CG with tolerance <span class="ansi-yellow-intense-fg">float64</span> <span class="ansi-blue-intense-fg">1e-05</span> (rel), <span class="ansi-yellow-intense-fg">float64</span> <span class="ansi-blue-intense-fg">1e-05</span> (abs), max_iterations=<span class="ansi-blue-intense-fg">1000</span> with backend torch (DEBUG), 2024-08-03 18:52:19,977n
 
 </pre>
 </div>
@@ -16155,11 +16155,11 @@ <h2 id="Linear-Solves-with-Native-Tensors">Linear Solves with Native Tensors<a c
 
 
 <div class="jp-RenderedText jp-OutputArea-output" data-mime-type="text/plain">
-<pre>TorchScript -&gt; run compiled forward &#39;_matrix_solve_forward&#39; with args [((2,), False), ((2, 2), False)] (DEBUG), 2024-08-03 18:18:25,603n
+<pre>TorchScript -&gt; run compiled forward &#39;_matrix_solve_forward&#39; with args [((2,), False), ((2, 2), False)] (DEBUG), 2024-08-03 18:52:20,000n
 
-Running forward pass of custom op forward &#39;_matrix_solve_forward&#39; given args (&#39;y&#39;, &#39;matrix&#39;) containing 2 native tensors (DEBUG), 2024-08-03 18:18:25,604n
+Running forward pass of custom op forward &#39;_matrix_solve_forward&#39; given args (&#39;y&#39;, &#39;matrix&#39;) containing 2 native tensors (DEBUG), 2024-08-03 18:52:20,000n
 
-Performing linear solve auto with tolerance <span class="ansi-yellow-intense-fg">float64</span> <span class="ansi-blue-intense-fg">1e-05</span> (rel), <span class="ansi-yellow-intense-fg">float64</span> <span class="ansi-blue-intense-fg">1e-05</span> (abs), max_iterations=<span class="ansi-blue-intense-fg">1000</span> with backend torch (DEBUG), 2024-08-03 18:18:25,608n
+Performing linear solve auto with tolerance <span class="ansi-yellow-intense-fg">float64</span> <span class="ansi-blue-intense-fg">1e-05</span> (rel), <span class="ansi-yellow-intense-fg">float64</span> <span class="ansi-blue-intense-fg">1e-05</span> (abs), max_iterations=<span class="ansi-blue-intense-fg">1000</span> with backend torch (DEBUG), 2024-08-03 18:52:20,004n
 
 </pre>
 </div>

N_Dimensional.html

@@ -15250,7 +15250,7 @@ <h2 id="Grids">Grids<a class="anchor-link" href="#Grids">&#182;</a></h2><p>Grids
 
 
 <div class="jp-RenderedText jp-OutputArea-output jp-OutputArea-executeResult" data-mime-type="text/plain">
-<pre><span class="ansi-blue-intense-fg">(0.656, 0.322, 0.769, 0.413, 0.392)</span> along <span class="ansi-green-intense-fg">xˢ</span></pre>
+<pre><span class="ansi-blue-intense-fg">(0.360, 0.426, 0.706, 0.334, 0.464)</span> along <span class="ansi-green-intense-fg">xˢ</span></pre>
 </div>
 
 </div>
@@ -15289,7 +15289,7 @@ <h2 id="Grids">Grids<a class="anchor-link" href="#Grids">&#182;</a></h2><p>Grids
 
 
 <div class="jp-RenderedText jp-OutputArea-output jp-OutputArea-executeResult" data-mime-type="text/plain">
-<pre><span class="ansi-green-intense-fg">(xˢ=3, yˢ=3)</span> <span class="ansi-blue-intense-fg">0.482 ± 0.110</span> <span class="ansi-white-fg">(3e-01...7e-01)</span></pre>
+<pre><span class="ansi-green-intense-fg">(xˢ=3, yˢ=3)</span> <span class="ansi-blue-intense-fg">0.619 ± 0.056</span> <span class="ansi-white-fg">(5e-01...7e-01)</span></pre>
 </div>
 
 </div>
@@ -15328,7 +15328,7 @@ <h2 id="Grids">Grids<a class="anchor-link" href="#Grids">&#182;</a></h2><p>Grids
 
 
 <div class="jp-RenderedText jp-OutputArea-output jp-OutputArea-executeResult" data-mime-type="text/plain">
-<pre><span class="ansi-green-intense-fg">(xˢ=16, yˢ=16, zˢ=16)</span> <span class="ansi-blue-intense-fg">0.493 ± 0.119</span> <span class="ansi-white-fg">(1e-01...9e-01)</span></pre>
+<pre><span class="ansi-green-intense-fg">(xˢ=16, yˢ=16, zˢ=16)</span> <span class="ansi-blue-intense-fg">0.502 ± 0.119</span> <span class="ansi-white-fg">(1e-01...9e-01)</span></pre>
 </div>
 
 </div>
@@ -15479,7 +15479,7 @@ <h2 id="Grids">Grids<a class="anchor-link" href="#Grids">&#182;</a></h2><p>Grids
 
 
 <div class="jp-RenderedText jp-OutputArea-output jp-OutputArea-executeResult" data-mime-type="text/plain">
-<pre><span class="ansi-blue-intense-fg">((2.55192+0j), (-0.25910917-0.45941594j), (-0.54797715-0.46981138j), (-0.54797715+0.46981138j), (-0.25910917+0.45941594j))</span> along <span class="ansi-green-intense-fg">xˢ</span> <span class="ansi-yellow-intense-fg">complex64</span></pre>
+<pre><span class="ansi-blue-intense-fg">((2.2901797+0j), (-0.6668361-0.5918753j), (0.04670691-0.46417135j), (0.04670691+0.46417135j), (-0.6668361+0.5918753j))</span> along <span class="ansi-green-intense-fg">xˢ</span> <span class="ansi-yellow-intense-fg">complex64</span></pre>
 </div>
 
 </div>
@@ -15518,7 +15518,7 @@ <h2 id="Grids">Grids<a class="anchor-link" href="#Grids">&#182;</a></h2><p>Grids
 
 
 <div class="jp-RenderedText jp-OutputArea-output jp-OutputArea-executeResult" data-mime-type="text/plain">
-<pre><span class="ansi-green-intense-fg">(xˢ=3, yˢ=3)</span> <span class="ansi-yellow-intense-fg">complex64</span> <span class="ansi-blue-intense-fg">|...| &lt; 4.33687686920166</span></pre>
+<pre><span class="ansi-green-intense-fg">(xˢ=3, yˢ=3)</span> <span class="ansi-yellow-intense-fg">complex64</span> <span class="ansi-blue-intense-fg">|...| &lt; 5.570593357086182</span></pre>
 </div>
 
 </div>
@@ -15557,7 +15557,7 @@ <h2 id="Grids">Grids<a class="anchor-link" href="#Grids">&#182;</a></h2><p>Grids
 
 
 <div class="jp-RenderedText jp-OutputArea-output jp-OutputArea-executeResult" data-mime-type="text/plain">
-<pre><span class="ansi-green-intense-fg">(xˢ=16, yˢ=16, zˢ=16)</span> <span class="ansi-yellow-intense-fg">complex64</span> <span class="ansi-blue-intense-fg">|...| &lt; 2019.8148193359375</span></pre>
+<pre><span class="ansi-green-intense-fg">(xˢ=16, yˢ=16, zˢ=16)</span> <span class="ansi-yellow-intense-fg">complex64</span> <span class="ansi-blue-intense-fg">|...| &lt; 2054.177978515625</span></pre>
 </div>
 
 </div>
@@ -15672,7 +15672,7 @@ <h2 id="Dimensions-as-Components">Dimensions as Components<a class="anchor-link"
 
 
 <div class="jp-RenderedText jp-OutputArea-output jp-OutputArea-executeResult" data-mime-type="text/plain">
-<pre><span class="ansi-green-intense-fg">(othersⁱ=4, pointsⁱ=4)</span> <span class="ansi-blue-intense-fg">0.327 ± 0.303</span> <span class="ansi-white-fg">(0e+00...9e-01)</span></pre>
+<pre><span class="ansi-green-intense-fg">(othersⁱ=4, pointsⁱ=4)</span> <span class="ansi-blue-intense-fg">0.376 ± 0.323</span> <span class="ansi-white-fg">(0e+00...8e-01)</span></pre>
 </div>
 
 </div>
@@ -15711,7 +15711,7 @@ <h2 id="Dimensions-as-Components">Dimensions as Components<a class="anchor-link"
 
 
 <div class="jp-RenderedText jp-OutputArea-output jp-OutputArea-executeResult" data-mime-type="text/plain">
-<pre><span class="ansi-green-intense-fg">(othersⁱ=4, pointsⁱ=4)</span> <span class="ansi-blue-intense-fg">0.307 ± 0.219</span> <span class="ansi-white-fg">(0e+00...7e-01)</span></pre>
+<pre><span class="ansi-green-intense-fg">(othersⁱ=4, pointsⁱ=4)</span> <span class="ansi-blue-intense-fg">0.293 ± 0.239</span> <span class="ansi-white-fg">(0e+00...7e-01)</span></pre>
 </div>
 
 </div>
@@ -15750,7 +15750,7 @@ <h2 id="Dimensions-as-Components">Dimensions as Components<a class="anchor-link"
 
 
 <div class="jp-RenderedText jp-OutputArea-output jp-OutputArea-executeResult" data-mime-type="text/plain">
-<pre><span class="ansi-green-intense-fg">(othersⁱ=4, pointsⁱ=4)</span> <span class="ansi-blue-intense-fg">0.535 ± 0.350</span> <span class="ansi-white-fg">(0e+00...1e+00)</span></pre>
+<pre><span class="ansi-green-intense-fg">(othersⁱ=4, pointsⁱ=4)</span> <span class="ansi-blue-intense-fg">0.544 ± 0.347</span> <span class="ansi-white-fg">(0e+00...1e+00)</span></pre>
 </div>
 
 </div>

Performance.html

@@ -15212,8 +15212,8 @@ <h2 id="Performance-and-JIT-compilation">Performance and JIT-compilation<a class
 
 
 <div class="jp-RenderedText jp-OutputArea-output" data-mime-type="text/plain">
-<pre>Φ-ML + torch JIT compilation: <span class="ansi-yellow-intense-fg">float64</span> <span class="ansi-blue-intense-fg">0.21197349</span>
-Φ-ML + torch execution average: 0.03549956902861595 +- 0.004727165214717388
+<pre>Φ-ML + torch JIT compilation: <span class="ansi-yellow-intense-fg">float64</span> <span class="ansi-blue-intense-fg">0.20342584</span>
+Φ-ML + torch execution average: 0.03442099690437317 +- 0.0051096719689667225
 </pre>
 </div>
 </div>
@@ -15233,8 +15233,8 @@ <h2 id="Performance-and-JIT-compilation">Performance and JIT-compilation<a class
 
 
 <div class="jp-RenderedText jp-OutputArea-output" data-mime-type="text/plain">
-<pre>Φ-ML + jax JIT compilation: <span class="ansi-yellow-intense-fg">float64</span> <span class="ansi-blue-intense-fg">0.1617278</span>
-Φ-ML + jax execution average: 0.012110414914786816 +- 0.0008309065597131848
+<pre>Φ-ML + jax JIT compilation: <span class="ansi-yellow-intense-fg">float64</span> <span class="ansi-blue-intense-fg">0.1580026</span>
+Φ-ML + jax execution average: 0.011942562647163868 +- 0.0009103829506784678
 </pre>
 </div>
 </div>
@@ -15244,7 +15244,7 @@ <h2 id="Performance-and-JIT-compilation">Performance and JIT-compilation<a class
 
 
 <div class="jp-RenderedText jp-OutputArea-output" data-mime-type="application/vnd.jupyter.stderr">
-<pre>2024-08-03 18:19:04.275435: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
+<pre>2024-08-03 18:52:57.862299: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
 </pre>
 </div>
 </div>
@@ -15254,8 +15254,8 @@ <h2 id="Performance-and-JIT-compilation">Performance and JIT-compilation<a class
 
 
 <div class="jp-RenderedText jp-OutputArea-output" data-mime-type="text/plain">
-<pre>Φ-ML + tensorflow JIT compilation: <span class="ansi-yellow-intense-fg">float64</span> <span class="ansi-blue-intense-fg">13.906091</span>
-Φ-ML + tensorflow execution average: 0.053461670875549316 +- 0.00282722688280046
+<pre>Φ-ML + tensorflow JIT compilation: <span class="ansi-yellow-intense-fg">float64</span> <span class="ansi-blue-intense-fg">13.694691</span>
+Φ-ML + tensorflow execution average: 0.05257116258144379 +- 0.0016889951657503843
 </pre>
 </div>
 </div>
@@ -15361,8 +15361,8 @@ <h2 id="Native-Implementations">Native Implementations<a class="anchor-link" hre
 
 
 <div class="jp-RenderedText jp-OutputArea-output" data-mime-type="text/plain">
-<pre>jax JIT compilation: 0.1335396100000139
-jax execution average: 0.010147904414141366
+<pre>jax JIT compilation: 0.13146938599999203
+jax execution average: 0.010196452373737267
 </pre>
 </div>
 </div>
@@ -15443,11 +15443,11 @@ <h2 id="Native-Implementations">Native Implementations<a class="anchor-link" hre
 
 
 <div class="jp-RenderedText jp-OutputArea-output" data-mime-type="application/vnd.jupyter.stderr">
-<pre>/tmp/ipykernel_2746/3571425526.py:12: TracerWarning: torch.tensor results are registered as constants in the trace. You can safely ignore this warning if you use this function to create tensors out of constant variables that would be the same every time you call this function. In any other case, this might cause the trace to be incorrect.
+<pre>/tmp/ipykernel_2692/3571425526.py:12: TracerWarning: torch.tensor results are registered as constants in the trace. You can safely ignore this warning if you use this function to create tensors out of constant variables that would be the same every time you call this function. In any other case, this might cause the trace to be incorrect.
   dist = torch.sqrt(torch.maximum(torch.sum(deltas ** 2, -1), torch.tensor(1e-4)))  # eps=1e-4 to avoid NaN during backprop of sqrt
-/tmp/ipykernel_2746/3571425526.py:20: TracerWarning: torch.tensor results are registered as constants in the trace. You can safely ignore this warning if you use this function to create tensors out of constant variables that would be the same every time you call this function. In any other case, this might cause the trace to be incorrect.
+/tmp/ipykernel_2692/3571425526.py:20: TracerWarning: torch.tensor results are registered as constants in the trace. You can safely ignore this warning if you use this function to create tensors out of constant variables that would be the same every time you call this function. In any other case, this might cause the trace to be incorrect.
   x_inc_contrib = torch.sum(torch.where(has_impact.unsqueeze(-1), torch.minimum(impact_time.unsqueeze(-1) - dt, torch.tensor(0.0)) * impulse, torch.tensor(0.0)), -2)
-/tmp/ipykernel_2746/3571425526.py:22: TracerWarning: torch.tensor results are registered as constants in the trace. You can safely ignore this warning if you use this function to create tensors out of constant variables that would be the same every time you call this function. In any other case, this might cause the trace to be incorrect.
+/tmp/ipykernel_2692/3571425526.py:22: TracerWarning: torch.tensor results are registered as constants in the trace. You can safely ignore this warning if you use this function to create tensors out of constant variables that would be the same every time you call this function. In any other case, this might cause the trace to be incorrect.
   v += torch.sum(torch.where(has_impact.unsqueeze(-1), impulse, torch.tensor(0.0)), -2)
 </pre>
 </div>
@@ -15458,8 +15458,8 @@ <h2 id="Native-Implementations">Native Implementations<a class="anchor-link" hre
 
 
 <div class="jp-RenderedText jp-OutputArea-output" data-mime-type="text/plain">
-<pre>torch JIT compilation: 0.036969639360904694
-torch execution average: 0.03337471932172775
+<pre>torch JIT compilation: 0.036141786724328995
+torch execution average: 0.03381171450018883
 </pre>
 </div>
 </div>
@@ -15469,7 +15469,7 @@ <h2 id="Native-Implementations">Native Implementations<a class="anchor-link" hre
 
 
 <div class="jp-RenderedText jp-OutputArea-output" data-mime-type="application/vnd.jupyter.stderr">
-<pre>/tmp/ipykernel_2746/3571425526.py:45: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
+<pre>/tmp/ipykernel_2692/3571425526.py:45: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
   print(f&#34;torch execution average: {torch.mean(torch.tensor(dt_torch[2:]))}&#34;)
 </pre>
 </div>
@@ -15545,8 +15545,8 @@ <h2 id="Native-Implementations">Native Implementations<a class="anchor-link" hre
 
 
 <div class="jp-RenderedText jp-OutputArea-output" data-mime-type="text/plain">
-<pre>tensorflow JIT compilation: 0.37863877415657043
-tensorflow execution average: 0.03856977820396423
+<pre>tensorflow JIT compilation: 0.20972128212451935
+tensorflow execution average: 0.03771407902240753
 </pre>
 </div>
 </div>