Skip to content

Commit

Permalink
Update course book
Browse files Browse the repository at this point in the history
  • Loading branch information
actions-user committed Feb 28, 2024
1 parent fe78469 commit 436ef1a
Show file tree
Hide file tree
Showing 252 changed files with 2,448 additions and 2,471 deletions.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Diff not rendered.
Diff not rendered.
Diff not rendered.
Diff not rendered.
Diff not rendered.
Diff not rendered.
Diff not rendered.
Diff not rendered.
44 changes: 22 additions & 22 deletions _sources/tutorials/Schedule/daily_schedules.md
Original file line number Diff line number Diff line change
Expand Up @@ -2,30 +2,30 @@

## Course overview

* Sun, July 9: Opening Ceremonies
* Sun, July 7: Opening Ceremonies
* 8:30 am UTC
* 5:30 pm UTC

* Week 1
* Mon, July 10: Model Types
* Tue, July 11: Model Fitting
* Wed, July 12: Generalized Linear Models
* Thu, July 13: Dimensionality Reduction
* Fri, July 14: Deep Learning
* Mon, July 8: Model Types
* Tue, July 9: Model Fitting
* Wed, July 10: Generalized Linear Models
* Thu, July 11: Dimensionality Reduction
* Fri, July 12: Deep Learning

* Week 2
* Mon, July 17: Modeling Practice & Half Project Day (Projects Proposal)
* Tue, July 18: Linear Systems
* Wed, July 19: Biological Neuron Models
* Thu, July 20: Dynamic Networks
* Fri, July 21: Project day (Abstract Writing)
* Mon, July 15: Modeling Practice & Half Project Day (Projects Proposal)
* Tue, July 15: Linear Systems
* Wed, July 17: Biological Neuron Models
* Thu, July 18: Dynamic Networks
* Fri, July 19: Project day (Abstract Writing)

* Week 3
* Mon, July 24: Bayesian Decisions
* Tue, July 25: Hidden Dynamics
* Wed, July 26: Optimal Control
* Thu, July 27: Reinforcement Learning
* Fri, July 28: Network Causality & Graduation
* Mon, July 22: Bayesian Decisions
* Tue, July 23: Hidden Dynamics
* Wed, July 24: Optimal Control
* Thu, July 25: Reinforcement Learning
* Fri, July 26: Network Causality & Graduation
----

## Daily coursework schedule
Expand Down Expand Up @@ -119,14 +119,13 @@ During project time:

| Time (Hour) | Component |
|------------------|---------------------------------------|
| 0:00-0:45\*, \** | Mentor meeting |
| 0:00-0:45\*, \** | Project work time |
| 0:45-1:45 | Pod abstract swap |
| 1:45-2:00 | Break |
| 2:00-3:00 | Finalize abstract |

\* : Note that this refers to the start of the projects time, not the coursework time like usual

\** : Timing of this mentor meeting may need to vary based on mentor availability

#### W3D5: Final Day of Course

Expand All @@ -151,11 +150,11 @@ During project time:

\** : Note that these events occur after the normal coursework time.

Course Review & Feedback: We will have a post course survey for both students and TAs (will share the link in discord) - please fill it out during this time!
Course Review & Feedback: We will have a post course survey for both students and TAs (will share the link in discord) - please fill it out during this time! Completing the final survey is required to access your certiicate after the coures.

Pod farewell: This is just a chance for your pod to say goodbye to each other. No structure imposed or requirements here!

Closing Ceremony/Graduation: We will do this all together on crowdcast.
Closing Ceremony/Graduation: We will do this all together on Zoom.


**Slot 3:**
Expand All @@ -178,11 +177,12 @@ Closing Ceremony/Graduation: We will do this all together on crowdcast.

\** : Note that these events occur after the normal coursework time.

Course Review & Feedback: We will have a post course survey for both students and TAs (will share the link in discord) - please fill it out during this time!
Course Review & Feedback: We will have a post course survey for both students and TAs (will share the link in discord) - please fill it out during this time! Completing the final survey is required to access your certiicate after the coures.


Pod farewell: This is just a chance for your pod to say goodbye to each other. No structure imposed or requirements here!

Closing Ceremony/Graduation: We will do this all together on crowdcast.
Closing Ceremony/Graduation: We will do this all together on Zoom.


### Professional Development Panels
Expand Down
5 changes: 2 additions & 3 deletions _sources/tutorials/TechnicalHelp/Links_Policy.md
Original file line number Diff line number Diff line change
Expand Up @@ -6,11 +6,10 @@ Course materials: [https://compneuro.neuromatch.io/](https://compneuro.neuromatc

Portal: [https://portal.neuromatchacademy.org/](https://portal.neuromatchacademy.org/)

Website: [https://academy.neuromatch.io/](https://academy.neuromatch.io/)
Website: [https://neuromatch.io/computational-neuroscience-course/](https://neuromatch.io/computational-neuroscience-course/)

Crowdcast: [https://www.crowdcast.io/e/neuromatch-academy-2022-](https://www.crowdcast.io/e/neuromatch-academy-2022-)

Code of Conduct and Code of Conduct Violations Form: [https://github.com/NeuromatchAcademy/precourse/blob/main/CODE_OF_CONDUCT.md](https://github.com/NeuromatchAcademy/precourse/blob/main/CODE_OF_CONDUCT.md)
Code of Conduct Violations Form: [https://github.com/NeuromatchAcademy/precourse/blob/main/CODE_OF_CONDUCT.md](https://github.com/NeuromatchAcademy/precourse/blob/main/CODE_OF_CONDUCT.md)

Project Exemption Form: [https://airtable.com/shrubhlgsWJ8DuA7E](https://airtable.com/shrubhlgsWJ8DuA7E)

Expand Down
8 changes: 4 additions & 4 deletions projects/behavior/behavior_videos.html

Large diffs are not rendered by default.

18 changes: 9 additions & 9 deletions projects/fMRI/fMRI_videos.html

Large diffs are not rendered by default.

20 changes: 10 additions & 10 deletions projects/modelingsteps/ModelingSteps_1through4.html

Large diffs are not rendered by default.

14 changes: 7 additions & 7 deletions projects/modelingsteps/ModelingSteps_5through10.html

Large diffs are not rendered by default.

18 changes: 9 additions & 9 deletions projects/modelingsteps/TrainIllusionDataProject.html
Original file line number Diff line number Diff line change
Expand Up @@ -2065,8 +2065,8 @@ <h1>Question<a class="headerlink" href="#question" title="Permalink to this head
<p><em>Part of Step 1</em></p>
<p>We assume that we have build the train illusion model (see the other example project colab). That model predicts that accumulated sensory evidence from vestibular signals determines the decision of whether self-motion is experienced or not. We now have vestibular neuron data (simulated in our case, but let’s pretend) and would like to see if that prediction holds true.</p>
<p>The data contains <span class="math notranslate nohighlight">\(N\)</span> neurons and <span class="math notranslate nohighlight">\(M\)</span> trials for each of 3 motion conditions: no self-motion, slowly accelerating self-motion and faster accelerating self-motion.</p>
<div class="amsmath math notranslate nohighlight" id="equation-20dc3725-0863-447e-bb99-08943807aa2f">
<span class="eqno">(470)<a class="headerlink" href="#equation-20dc3725-0863-447e-bb99-08943807aa2f" title="Permalink to this equation">#</a></span>\[\begin{align}
<div class="amsmath math notranslate nohighlight" id="equation-0350620c-730a-4d93-bf12-54e6cac5f54b">
<span class="eqno">(470)<a class="headerlink" href="#equation-0350620c-730a-4d93-bf12-54e6cac5f54b" title="Permalink to this equation">#</a></span>\[\begin{align}
N &amp;= 40\\
M &amp;= 400\\
\end{align}\]</div>
Expand Down Expand Up @@ -2111,7 +2111,7 @@ <h1>Background<a class="headerlink" href="#background" title="Permalink to this
</div>
</div>
<div class="cell_output docutils container">
<img alt="../../_images/995417fc3a013f54f97a875ff0e86f1c8f7825182cbfdecfa3d681968270e1a1.png" src="../../_images/995417fc3a013f54f97a875ff0e86f1c8f7825182cbfdecfa3d681968270e1a1.png" />
<img alt="../../_images/691998358ad0b84a3aa3e260513e493dea28f2884db29c3b72d2ae13d752d828.png" src="../../_images/691998358ad0b84a3aa3e260513e493dea28f2884db29c3b72d2ae13d752d828.png" />
</div>
</div>
<p>Blue is the no-motion condition, and produces flat average spike counts across the 3 s time interval. The orange and green line do show a bell-shaped curve that corresponds to the acceleration profile. But there also seems to be considerable noise: exactly what we need. Let’s see what the spike trains for a single trial look like:</p>
Expand All @@ -2123,9 +2123,9 @@ <h1>Background<a class="headerlink" href="#background" title="Permalink to this
</div>
</div>
<div class="cell_output docutils container">
<img alt="../../_images/0af6a7ce5b7fcfc66559e3abb9be1a363181008af25f16626064c8e0a08f4aec.png" src="../../_images/0af6a7ce5b7fcfc66559e3abb9be1a363181008af25f16626064c8e0a08f4aec.png" />
<img alt="../../_images/8464bd3763a148171d6c3a9457a1ead7354d5cbc4df1147aa55fb2784a55ced9.png" src="../../_images/8464bd3763a148171d6c3a9457a1ead7354d5cbc4df1147aa55fb2784a55ced9.png" />
<img alt="../../_images/8336499ecbe4946418db2379afdca992aae5b15d8d5bd46502e650fdf6c9963c.png" src="../../_images/8336499ecbe4946418db2379afdca992aae5b15d8d5bd46502e650fdf6c9963c.png" />
<img alt="../../_images/86d5ecbd5807f4033da9d40d5f7cda9fe5964ddc6b6f1329e5e61217afe379ff.png" src="../../_images/86d5ecbd5807f4033da9d40d5f7cda9fe5964ddc6b6f1329e5e61217afe379ff.png" />
<img alt="../../_images/54885e65427d84d7b01f9a435a4e143802b2c70c468ce6468848a91c2138d1ea.png" src="../../_images/54885e65427d84d7b01f9a435a4e143802b2c70c468ce6468848a91c2138d1ea.png" />
<img alt="../../_images/d8aea21b3d80d298463d9b5b90da9a832b7a4cfa982949836b3ac13c8ae399b7.png" src="../../_images/d8aea21b3d80d298463d9b5b90da9a832b7a4cfa982949836b3ac13c8ae399b7.png" />
</div>
</div>
<p>You can change the trial number in the bit of code above to compare what the rasterplots look like in different trials. You’ll notice that they all look kind of the same: the 3 conditions are very hard (impossible?) to distinguish by eye-balling.</p>
Expand Down Expand Up @@ -2273,7 +2273,7 @@ <h1>Model implementation<a class="headerlink" href="#model-implementation" title
</div>
</div>
<div class="cell_output docutils container">
<img alt="../../_images/c4da38692cebb95d43f122a69dccd23bc3e28537640465507366d41f5b438610.png" src="../../_images/c4da38692cebb95d43f122a69dccd23bc3e28537640465507366d41f5b438610.png" />
<img alt="../../_images/657b75caf157f5e8a44e7e14cd2f691d4f1560628aa3d1d8a2e085eeec1c3356.png" src="../../_images/657b75caf157f5e8a44e7e14cd2f691d4f1560628aa3d1d8a2e085eeec1c3356.png" />
</div>
</div>
<p>We asked for 8 cross validations, which show up as the blue dots in the graph (two have the same accuracy). Prediction accuracy ranges from 56% to 72%, with the average at 65%, and the orange line is the median. Given the noisy data, that is not too bad actually.</p>
Expand Down Expand Up @@ -2327,7 +2327,7 @@ <h1>Model implementation<a class="headerlink" href="#model-implementation" title
</div>
</div>
<div class="cell_output docutils container">
<img alt="../../_images/c4da38692cebb95d43f122a69dccd23bc3e28537640465507366d41f5b438610.png" src="../../_images/c4da38692cebb95d43f122a69dccd23bc3e28537640465507366d41f5b438610.png" />
<img alt="../../_images/657b75caf157f5e8a44e7e14cd2f691d4f1560628aa3d1d8a2e085eeec1c3356.png" src="../../_images/657b75caf157f5e8a44e7e14cd2f691d4f1560628aa3d1d8a2e085eeec1c3356.png" />
</div>
</div>
<p>This is the exact same figure as before, so our function <code class="docutils literal notranslate"><span class="pre">classifyMotionFromSpikes()</span></code> also works as intended.</p>
Expand Down Expand Up @@ -2463,7 +2463,7 @@ <h1>Model evaluation &amp; testing<a class="headerlink" href="#model-evaluation-
</div>
</div>
<div class="cell_output docutils container">
<img alt="../../_images/2d03961d6353a2a3ab0ad599a478f9c6f4149395c6cd7d7793558af41a669e98.png" src="../../_images/2d03961d6353a2a3ab0ad599a478f9c6f4149395c6cd7d7793558af41a669e98.png" />
<img alt="../../_images/92c8a721ca3f1972276d94a0fb7c12964f0ce72c90ab24090445dd10e9060044.png" src="../../_images/92c8a721ca3f1972276d94a0fb7c12964f0ce72c90ab24090445dd10e9060044.png" />
</div>
</div>
<p>Well, that’s interesting! The logistic regression doesn’t do a perfect job, but there is information in these results.</p>
Expand Down
26 changes: 8 additions & 18 deletions projects/modelingsteps/TrainIllusionModel.html
Original file line number Diff line number Diff line change
Expand Up @@ -1888,8 +1888,8 @@ <h1>Selected toolkit<a class="headerlink" href="#selected-toolkit" title="Permal
<p><em>Part of step 5</em></p>
<p>We chose to use a <a class="reference external" href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2474742/">Drift-Diffusion Model (DDM)</a> because it is a well-established framework that allows us to model decision making in the case of 2 alternative choices (here: self-motion vs. other train motion).</p>
<p>For our purposes simplest equation looks something like this:</p>
<div class="amsmath math notranslate nohighlight" id="equation-4dfa0d2b-6728-4600-b365-1310048d0420">
<span class="eqno">(469)<a class="headerlink" href="#equation-4dfa0d2b-6728-4600-b365-1310048d0420" title="Permalink to this equation">#</a></span>\[\begin{align}
<div class="amsmath math notranslate nohighlight" id="equation-7a63c18c-c139-425c-ba8d-90ecbad34591">
<span class="eqno">(469)<a class="headerlink" href="#equation-7a63c18c-c139-425c-ba8d-90ecbad34591" title="Permalink to this equation">#</a></span>\[\begin{align}
\dot e = \frac{de}{dt}= -c \cdot e + v
\end{align}\]</div>
<p>where <span class="math notranslate nohighlight">\(e\)</span> is the accumulated evidence and <span class="math notranslate nohighlight">\(v\)</span> is our vestibular input already containing the noise (so we don’t need to add more noise?). <span class="math notranslate nohighlight">\(c\)</span> is the leakage constant, i.e. <span class="math notranslate nohighlight">\(c=0\)</span> means perfect integration; <span class="math notranslate nohighlight">\(c=1\)</span> means no integration (perfect leakage).</p>
Expand Down Expand Up @@ -1965,7 +1965,7 @@ <h2>1. Vestibular signal generator<a class="headerlink" href="#vestibular-signal
<div class="output text_plain highlight-myst-ansi notranslate"><div class="highlight"><pre><span></span>Text(0, 0.5, &#39;vestibular signal (a.u.)&#39;)
</pre></div>
</div>
<img alt="../../_images/82439805bf8d1e03e7e41bf8552bf64be279d3f1e4a5d55b03cf58f150efb564.png" src="../../_images/82439805bf8d1e03e7e41bf8552bf64be279d3f1e4a5d55b03cf58f150efb564.png" />
<img alt="../../_images/2df68faf49e9d054e299b363ab2bde053494fb4e202d7f88c1a745f453b0a94c.png" src="../../_images/2df68faf49e9d054e299b363ab2bde053494fb4e202d7f88c1a745f453b0a94c.png" />
</div>
</div>
</section>
Expand Down Expand Up @@ -2008,7 +2008,7 @@ <h2>2. integrator (DDM mechanism)<a class="headerlink" href="#integrator-ddm-mec
<div class="output text_plain highlight-myst-ansi notranslate"><div class="highlight"><pre><span></span>Text(0, 0.5, &#39;integrated evidence&#39;)
</pre></div>
</div>
<img alt="../../_images/eacf2123b6b439ca68cb36af80096346accc141a39be6ec417e7040965b5b2db.png" src="../../_images/eacf2123b6b439ca68cb36af80096346accc141a39be6ec417e7040965b5b2db.png" />
<img alt="../../_images/9a6fcefd2d8764389ea83664b71778c9e68ddccdeb926be0397d300f00eed864.png" src="../../_images/9a6fcefd2d8764389ea83664b71778c9e68ddccdeb926be0397d300f00eed864.png" />
</div>
</div>
</section>
Expand Down Expand Up @@ -2076,7 +2076,7 @@ <h2>Assembling the model<a class="headerlink" href="#assembling-the-model" title
</div>
</div>
<div class="cell_output docutils container">
<div class="output text_plain highlight-myst-ansi notranslate"><div class="highlight"><pre><span></span>0
<div class="output text_plain highlight-myst-ansi notranslate"><div class="highlight"><pre><span></span>1
</pre></div>
</div>
</div>
Expand Down Expand Up @@ -2157,17 +2157,7 @@ <h1>Model evaluation &amp; testing<a class="headerlink" href="#model-evaluation-
</div>
</div>
<div class="cell_output docutils container">
<div class="output stderr highlight-myst-ansi notranslate"><div class="highlight"><pre><span></span>/tmp/ipykernel_3285/273880207.py:2: DeprecationWarning:
Pyarrow will become a required dependency of pandas in the next major release of pandas (pandas 3.0),
(to allow more performant data types, such as the Arrow string type, and better interoperability with other libraries)
but was not found to be installed on your system.
If this would cause problems for you,
please provide us feedback at https://github.com/pandas-dev/pandas/issues/54466

import pandas as pd
</pre></div>
</div>
<img alt="../../_images/7d134200083061d661e330bb2324310b611f901a2a0e8dc0f347fb94ddf8bb1b.png" src="../../_images/7d134200083061d661e330bb2324310b611f901a2a0e8dc0f347fb94ddf8bb1b.png" />
<img alt="../../_images/8de824fcc15d3d879c626263b2ef7886843ca0ceca851612e24b7ff4624d5f34.png" src="../../_images/8de824fcc15d3d879c626263b2ef7886843ca0ceca851612e24b7ff4624d5f34.png" />
</div>
</div>
<p>There seems to be some parameter redundancy. I.e. we could chose different parameter combinations to make the model do something sensible…</p>
Expand All @@ -2188,7 +2178,7 @@ <h1>Model evaluation &amp; testing<a class="headerlink" href="#model-evaluation-
</div>
</div>
<div class="cell_output docutils container">
<div class="output stream highlight-myst-ansi notranslate"><div class="highlight"><pre><span></span> Motion detected for no-motion = 30.3 %; and motion = 59.0 %
<div class="output stream highlight-myst-ansi notranslate"><div class="highlight"><pre><span></span> Motion detected for no-motion = 32.0 %; and motion = 60.3 %
</pre></div>
</div>
</div>
Expand Down Expand Up @@ -2225,7 +2215,7 @@ <h1>Model evaluation &amp; testing<a class="headerlink" href="#model-evaluation-
</div>
</div>
<div class="cell_output docutils container">
<img alt="../../_images/4f84cd33a3cea84c5adc81c0dd09d9a483557ca2c99addd3c5d93ff92e078b01.png" src="../../_images/4f84cd33a3cea84c5adc81c0dd09d9a483557ca2c99addd3c5d93ff92e078b01.png" />
<img alt="../../_images/8c210da3c6239a41e224933748c1abecf9b4514beedcdc45bc6219836ba97e5b.png" src="../../_images/8c210da3c6239a41e224933748c1abecf9b4514beedcdc45bc6219836ba97e5b.png" />
</div>
</div>
<p>Our <strong>hypothesis</strong> of linear increase of illusion strength with noise only holds true in a limited range of noise… It’s monotonic but saturating of course…</p>
Expand Down
8 changes: 4 additions & 4 deletions projects/neurons/neurons_videos.html

Large diffs are not rendered by default.

2 changes: 1 addition & 1 deletion searchindex.js

Large diffs are not rendered by default.

8 changes: 4 additions & 4 deletions tutorials/Bonus_Autoencoders/student/Bonus_Intro.html

Large diffs are not rendered by default.

8 changes: 4 additions & 4 deletions tutorials/Bonus_Autoencoders/student/Bonus_Outro.html

Large diffs are not rendered by default.

8 changes: 4 additions & 4 deletions tutorials/Bonus_Autoencoders/student/Bonus_Tutorial2.html
Original file line number Diff line number Diff line change
Expand Up @@ -2079,8 +2079,8 @@ <h1>Section 3: Spherical latent space<a class="headerlink" href="#section-3-sphe
<p><img alt="Unit sphere S2" src="https://github.com/mpbrigham/colaboratory-figures/raw/master/nma/autoencoders/unit_sphere.png"/></p>
<br/>
<p>Projecting to the surface of the sphere is implemented by dividing the coordinates <span class="math notranslate nohighlight">\((z_1, z_2, z_3)\)</span> by their <span class="math notranslate nohighlight">\(L_2\)</span> norm.</p>
<div class="amsmath math notranslate nohighlight" id="equation-e450b6e9-b214-4248-bcd0-994e46681259">
<span class="eqno">(262)<a class="headerlink" href="#equation-e450b6e9-b214-4248-bcd0-994e46681259" title="Permalink to this equation">#</a></span>\[\begin{equation}
<div class="amsmath math notranslate nohighlight" id="equation-12313b78-55e9-4016-aae3-11ac76a6a801">
<span class="eqno">(262)<a class="headerlink" href="#equation-12313b78-55e9-4016-aae3-11ac76a6a801" title="Permalink to this equation">#</a></span>\[\begin{equation}
(z_1, z_2, z_3)\longmapsto (s_1, s_2, s_3)=(z_1, z_2, z_3)/\|(z_1, z_2, z_3)\|_2=(z_1, z_2, z_3)/ \sqrt{z_1^2+z_2^2+z_3^2}
\end{equation}\]</div>
<p>This mapping projects to the surface of the <a class="reference external" href="https://en.wikipedia.org/wiki/N-sphere"><span class="math notranslate nohighlight">\(S_2\)</span> sphere</a> with unit radius. (Why?)</p>
Expand Down Expand Up @@ -2267,8 +2267,8 @@ <h2>Section 3.4: Train the autoencoder<a class="headerlink" href="#section-3-4-t
<section id="section-3-5-visualize-latent-space-on-surface-of-s-2">
<h2>Section 3.5: Visualize latent space on surface of <span class="math notranslate nohighlight">\(S_2\)</span><a class="headerlink" href="#section-3-5-visualize-latent-space-on-surface-of-s-2" title="Permalink to this heading">#</a></h2>
<p>The 3D coordinates <span class="math notranslate nohighlight">\((s_1, s_2, s_3)\)</span> on the surface of the unit sphere <span class="math notranslate nohighlight">\(S_2\)</span> can be mapped to <a class="reference external" href="https://en.wikipedia.org/wiki/Spherical_coordinate_system">spherical coordinates</a> <span class="math notranslate nohighlight">\((r, \theta, \phi)\)</span>, as follows:</p>
<div class="amsmath math notranslate nohighlight" id="equation-e4ab956d-efc4-43ce-8710-8a97e055c95b">
<span class="eqno">(263)<a class="headerlink" href="#equation-e4ab956d-efc4-43ce-8710-8a97e055c95b" title="Permalink to this equation">#</a></span>\[\begin{align}
<div class="amsmath math notranslate nohighlight" id="equation-7f34f6b8-bdba-4ecf-9c5d-1c820aa585c3">
<span class="eqno">(263)<a class="headerlink" href="#equation-7f34f6b8-bdba-4ecf-9c5d-1c820aa585c3" title="Permalink to this equation">#</a></span>\[\begin{align}
r &amp;= \sqrt{s_1^2 + s_2^2 + s_3^2} \\
\phi &amp;= \arctan \frac{s_2}{s_1} \\
\theta &amp;= \arccos\frac{s_3}{r}
Expand Down
Loading

0 comments on commit 436ef1a

Please sign in to comment.