diff --git a/doc/pub/week15/html/week15-bs.html b/doc/pub/week15/html/week15-bs.html index ce009b71..1d750038 100644 --- a/doc/pub/week15/html/week15-bs.html +++ b/doc/pub/week15/html/week15-bs.html @@ -309,10 +309,10 @@

Plans for the
  1. Summary of Variational Autoencoders
  2. -
  3. Generative Adversarial Networks (GANs)
  4. -
  5. Start discussion of diffusion models - -
  6. +
  7. Generative Adversarial Networks (GANs), see https://lilianweng.github.io/posts/2017-08-20-gan/ for nice overview
  8. +
  9. Start discussion of diffusion models
  10. +
  11. Video of lecture
  12. +
  13. Whiteboard notes
diff --git a/doc/pub/week15/html/week15-reveal.html b/doc/pub/week15/html/week15-reveal.html index e3242f36..c496b5a9 100644 --- a/doc/pub/week15/html/week15-reveal.html +++ b/doc/pub/week15/html/week15-reveal.html @@ -202,10 +202,10 @@

Plans for the week of April 2

  1. Summary of Variational Autoencoders
  2. -

  3. Generative Adversarial Networks (GANs)
  4. -

  5. Start discussion of diffusion models - -
  6. +

  7. Generative Adversarial Networks (GANs), see https://lilianweng.github.io/posts/2017-08-20-gan/ for nice overview
  8. +

  9. Start discussion of diffusion models
  10. +

  11. Video of lecture
  12. +

  13. Whiteboard notes
diff --git a/doc/pub/week15/html/week15-solarized.html b/doc/pub/week15/html/week15-solarized.html index 3ac7a0da..8c5fd574 100644 --- a/doc/pub/week15/html/week15-solarized.html +++ b/doc/pub/week15/html/week15-solarized.html @@ -244,10 +244,10 @@

Plans for the week of April 2

  1. Summary of Variational Autoencoders
  2. -
  3. Generative Adversarial Networks (GANs)
  4. -
  5. Start discussion of diffusion models - -
  6. +
  7. Generative Adversarial Networks (GANs), see https://lilianweng.github.io/posts/2017-08-20-gan/ for nice overview
  8. +
  9. Start discussion of diffusion models
  10. +
  11. Video of lecture
  12. +
  13. Whiteboard notes
diff --git a/doc/pub/week15/html/week15.html b/doc/pub/week15/html/week15.html index 56e61571..9bb01f5f 100644 --- a/doc/pub/week15/html/week15.html +++ b/doc/pub/week15/html/week15.html @@ -321,10 +321,10 @@

Plans for the week of April 2

  1. Summary of Variational Autoencoders
  2. -
  3. Generative Adversarial Networks (GANs)
  4. -
  5. Start discussion of diffusion models - -
  6. +
  7. Generative Adversarial Networks (GANs), see https://lilianweng.github.io/posts/2017-08-20-gan/ for nice overview
  8. +
  9. Start discussion of diffusion models
  10. +
  11. Video of lecture
  12. +
  13. Whiteboard notes
diff --git a/doc/pub/week15/ipynb/ipynb-week15-src.tar.gz b/doc/pub/week15/ipynb/ipynb-week15-src.tar.gz index 6e2fa832..831ba04f 100644 Binary files a/doc/pub/week15/ipynb/ipynb-week15-src.tar.gz and b/doc/pub/week15/ipynb/ipynb-week15-src.tar.gz differ diff --git a/doc/pub/week15/ipynb/week15.ipynb b/doc/pub/week15/ipynb/week15.ipynb index 20853fc8..46204ccc 100644 --- a/doc/pub/week15/ipynb/week15.ipynb +++ b/doc/pub/week15/ipynb/week15.ipynb @@ -2,8 +2,10 @@ "cells": [ { "cell_type": "markdown", - "id": "2e110748", - "metadata": {}, + "id": "bbc251d3", + "metadata": { + "editable": true + }, "source": [ "\n", @@ -12,8 +14,10 @@ }, { "cell_type": "markdown", - "id": "9e12bd6e", - "metadata": {}, + "id": "8154bff2", + "metadata": { + "editable": true + }, "source": [ "# Advanced machine learning and data analysis for the physical sciences\n", "**Morten Hjorth-Jensen**, Department of Physics and Center for Computing in Science Education, University of Oslo, Norway and Department of Physics and Astronomy and Facility for Rare Isotope Beams, Michigan State University, East Lansing, Michigan, USA\n", @@ -23,8 +27,10 @@ }, { "cell_type": "markdown", - "id": "372a7ed5", - "metadata": {}, + "id": "30bdfd80", + "metadata": { + "editable": true + }, "source": [ "## Plans for the week of April 29- May 3, 2024\n", "\n", @@ -32,17 +38,21 @@ "\n", "1. Summary of Variational Autoencoders\n", "\n", - "2. Generative Adversarial Networks (GANs)\n", + "2. Generative Adversarial Networks (GANs), see for nice overview\n", "\n", "3. Start discussion of diffusion models\n", - "\n", - "" + "\n", + "4. [Video of lecture](https://youtu.be/Cg8n9aWwHuU)\n", + "\n", + "5. [Whiteboard notes](https://github.com/CompPhysics/AdvancedMachineLearning/blob/main/doc/HandwrittenNotes/2024/NotesApril30.pdf)" ] }, { "cell_type": "markdown", - "id": "60ea87ab", - "metadata": {}, + "id": "b0ea3794", + "metadata": { + "editable": true + }, "source": [ "## Readings\n", "\n", @@ -55,8 +65,10 @@ }, { "cell_type": "markdown", - "id": "72e6ab04", - "metadata": {}, + "id": "32939787", + "metadata": { + "editable": true + }, "source": [ "## Summary of Variational Autoencoders (VAEs)\n", "\n", @@ -69,16 +81,20 @@ }, { "cell_type": "markdown", - "id": "87ac135f", - "metadata": {}, + "id": "89bf1911", + "metadata": { + "editable": true + }, "source": [ "## Boltzmann machines and energy-based models and contrastive optimization" ] }, { "cell_type": "markdown", - "id": "f7d948fc", - "metadata": {}, + "id": "f04c1132", + "metadata": { + "editable": true + }, "source": [ "## Energy models\n", "\n", @@ -87,8 +103,10 @@ }, { "cell_type": "markdown", - "id": "0eafbcb2", - "metadata": {}, + "id": "306c3a9e", + "metadata": { + "editable": true + }, "source": [ "$$\n", "p(\\boldsymbol{X})=\\prod_{x_i\\in \\boldsymbol{X}}p(x_i),\n", @@ -97,16 +115,20 @@ }, { "cell_type": "markdown", - "id": "59ce6766", - "metadata": {}, + "id": "38d85424", + "metadata": { + "editable": true + }, "source": [ "where we have assumed that the random varaibles $x_i$ are all independent and identically distributed (iid)." ] }, { "cell_type": "markdown", - "id": "1d085906", - "metadata": {}, + "id": "72d0e5a4", + "metadata": { + "editable": true + }, "source": [ "## Probability model\n", "\n", @@ -115,8 +137,10 @@ }, { "cell_type": "markdown", - "id": "8ef3a999", - "metadata": {}, + "id": "edc35f6e", + "metadata": { + "editable": true + }, "source": [ "$$\n", "p(x_i,h_j;\\boldsymbol{\\Theta}) = \\frac{f(x_i,h_j;\\boldsymbol{\\Theta})}{Z(\\boldsymbol{\\Theta})},\n", @@ -125,8 +149,10 @@ }, { "cell_type": "markdown", - "id": "d95523ac", - "metadata": {}, + "id": "eab97d57", + "metadata": { + "editable": true + }, "source": [ "where $f(x_i,h_j;\\boldsymbol{\\Theta})$ is a function which we assume is larger or\n", "equal than zero and obeys all properties required for a probability\n", @@ -137,8 +163,10 @@ }, { "cell_type": "markdown", - "id": "446b9219", - "metadata": {}, + "id": "e36868b7", + "metadata": { + "editable": true + }, "source": [ "$$\n", "Z(\\boldsymbol{\\Theta})=\\sum_{x_i\\in \\boldsymbol{X}}\\sum_{h_j\\in \\boldsymbol{H}} f(x_i,h_j;\\boldsymbol{\\Theta}).\n", @@ -147,8 +175,10 @@ }, { "cell_type": "markdown", - "id": "4cce8c03", - "metadata": {}, + "id": "05ccd2ab", + "metadata": { + "editable": true + }, "source": [ "## Marginal and conditional probabilities\n", "\n", @@ -157,8 +187,10 @@ }, { "cell_type": "markdown", - "id": "70e792b7", - "metadata": {}, + "id": "e5f7b578", + "metadata": { + "editable": true + }, "source": [ "$$\n", "p(x_i;\\boldsymbol{\\Theta}) = \\frac{\\sum_{h_j\\in \\boldsymbol{H}}f(x_i,h_j;\\boldsymbol{\\Theta})}{Z(\\boldsymbol{\\Theta})},\n", @@ -167,16 +199,20 @@ }, { "cell_type": "markdown", - "id": "d904edae", - "metadata": {}, + "id": "fdc48043", + "metadata": { + "editable": true + }, "source": [ "and" ] }, { "cell_type": "markdown", - "id": "a0ec7c21", - "metadata": {}, + "id": "32dd3ff0", + "metadata": { + "editable": true + }, "source": [ "$$\n", "p(h_i;\\boldsymbol{\\Theta}) = \\frac{\\sum_{x_i\\in \\boldsymbol{X}}f(x_i,h_j;\\boldsymbol{\\Theta})}{Z(\\boldsymbol{\\Theta})}.\n", @@ -185,8 +221,10 @@ }, { "cell_type": "markdown", - "id": "808da9e3", - "metadata": {}, + "id": "5e333dc2", + "metadata": { + "editable": true + }, "source": [ "## Partition function\n", "\n", @@ -198,8 +236,10 @@ }, { "cell_type": "markdown", - "id": "1f727294", - "metadata": {}, + "id": "9d4996e0", + "metadata": { + "editable": true + }, "source": [ "$$\n", "Z(\\boldsymbol{\\Theta})=\\sum_{x_i\\in \\boldsymbol{X}}\\sum_{h_j\\in \\boldsymbol{H}} f(x_i,h_j;\\boldsymbol{\\Theta}),\n", @@ -208,16 +248,20 @@ }, { "cell_type": "markdown", - "id": "9af6f8af", - "metadata": {}, + "id": "bfbcdfd5", + "metadata": { + "editable": true + }, "source": [ "changes to" ] }, { "cell_type": "markdown", - "id": "7859ac9d", - "metadata": {}, + "id": "2bac65fd", + "metadata": { + "editable": true + }, "source": [ "$$\n", "Z(\\boldsymbol{\\Theta})=\\sum_{\\boldsymbol{x}}\\sum_{\\boldsymbol{h}} f(\\boldsymbol{x},\\boldsymbol{h};\\boldsymbol{\\Theta}).\n", @@ -226,8 +270,10 @@ }, { "cell_type": "markdown", - "id": "3974d552", - "metadata": {}, + "id": "f513d787", + "metadata": { + "editable": true + }, "source": [ "If we have a binary set of variable $x_i$ and $h_j$ and $M$ values of $x_i$ and $N$ values of $h_j$ we have in total $2^M$ and $2^N$ possible $\\boldsymbol{x}$ and $\\boldsymbol{h}$ configurations, respectively.\n", "\n", @@ -237,8 +283,10 @@ }, { "cell_type": "markdown", - "id": "0e0dd828", - "metadata": {}, + "id": "7344a35e", + "metadata": { + "editable": true + }, "source": [ "## Optimization problem\n", "\n", @@ -247,8 +295,10 @@ }, { "cell_type": "markdown", - "id": "f85c772f", - "metadata": {}, + "id": "135b8bb8", + "metadata": { + "editable": true + }, "source": [ "$$\n", "p(\\boldsymbol{X};\\boldsymbol{\\Theta})=\\prod_{x_i\\in \\boldsymbol{X}}p(x_i;\\boldsymbol{\\Theta})=\\prod_{x_i\\in \\boldsymbol{X}}\\left(\\frac{\\sum_{h_j\\in \\boldsymbol{H}}f(x_i,h_j;\\boldsymbol{\\Theta})}{Z(\\boldsymbol{\\Theta})}\\right),\n", @@ -257,16 +307,20 @@ }, { "cell_type": "markdown", - "id": "2dc86863", - "metadata": {}, + "id": "aea87bfd", + "metadata": { + "editable": true + }, "source": [ "which we rewrite as" ] }, { "cell_type": "markdown", - "id": "1a2874e9", - "metadata": {}, + "id": "cce8712d", + "metadata": { + "editable": true + }, "source": [ "$$\n", "p(\\boldsymbol{X};\\boldsymbol{\\Theta})=\\frac{1}{Z(\\boldsymbol{\\Theta})}\\prod_{x_i\\in \\boldsymbol{X}}\\left(\\sum_{h_j\\in \\boldsymbol{H}}f(x_i,h_j;\\boldsymbol{\\Theta})\\right).\n", @@ -275,8 +329,10 @@ }, { "cell_type": "markdown", - "id": "6ad8c91c", - "metadata": {}, + "id": "5bcfef5b", + "metadata": { + "editable": true + }, "source": [ "## Further simplifications\n", "\n", @@ -285,8 +341,10 @@ }, { "cell_type": "markdown", - "id": "0574c65a", - "metadata": {}, + "id": "665445d0", + "metadata": { + "editable": true + }, "source": [ "$$\n", "p(\\boldsymbol{X};\\boldsymbol{\\Theta})=\\frac{1}{Z(\\boldsymbol{\\Theta})}\\prod_{x_i\\in \\boldsymbol{X}}f(x_i;\\boldsymbol{\\Theta}),\n", @@ -295,8 +353,10 @@ }, { "cell_type": "markdown", - "id": "71e7a68f", - "metadata": {}, + "id": "bdc994e9", + "metadata": { + "editable": true + }, "source": [ "where we used $p(x_i;\\boldsymbol{\\Theta}) = \\sum_{h_j\\in \\boldsymbol{H}}f(x_i,h_j;\\boldsymbol{\\Theta})$.\n", "The optimization problem is then" @@ -304,8 +364,10 @@ }, { "cell_type": "markdown", - "id": "4391e56d", - "metadata": {}, + "id": "97db2763", + "metadata": { + "editable": true + }, "source": [ "$$\n", "{\\displaystyle \\mathrm{arg} \\hspace{0.1cm}\\max_{\\boldsymbol{\\boldsymbol{\\Theta}}\\in {\\mathbb{R}}^{p}}} \\hspace{0.1cm}p(\\boldsymbol{X};\\boldsymbol{\\Theta}).\n", @@ -314,8 +376,10 @@ }, { "cell_type": "markdown", - "id": "5190ce80", - "metadata": {}, + "id": "6274f6f9", + "metadata": { + "editable": true + }, "source": [ "## Optimizing the logarithm instead\n", "\n", @@ -326,8 +390,10 @@ }, { "cell_type": "markdown", - "id": "7ff51e54", - "metadata": {}, + "id": "a8c268f3", + "metadata": { + "editable": true + }, "source": [ "$$\n", "{\\displaystyle \\mathrm{arg} \\hspace{0.1cm}\\max_{\\boldsymbol{\\boldsymbol{\\Theta}}\\in {\\mathbb{R}}^{p}}} \\hspace{0.1cm}\\log{p(\\boldsymbol{X};\\boldsymbol{\\Theta})},\n", @@ -336,16 +402,20 @@ }, { "cell_type": "markdown", - "id": "2d40eebb", - "metadata": {}, + "id": "5fb324c3", + "metadata": { + "editable": true + }, "source": [ "which leads to" ] }, { "cell_type": "markdown", - "id": "822d6a93", - "metadata": {}, + "id": "3ab2b400", + "metadata": { + "editable": true + }, "source": [ "$$\n", "\\nabla_{\\boldsymbol{\\Theta}}\\log{p(\\boldsymbol{X};\\boldsymbol{\\Theta})}=0.\n", @@ -354,8 +424,10 @@ }, { "cell_type": "markdown", - "id": "89492d82", - "metadata": {}, + "id": "ad53bab5", + "metadata": { + "editable": true + }, "source": [ "## Expression for the gradients\n", "\n", @@ -364,8 +436,10 @@ }, { "cell_type": "markdown", - "id": "ff456b90", - "metadata": {}, + "id": "80022c49", + "metadata": { + "editable": true + }, "source": [ "$$\n", "\\nabla_{\\boldsymbol{\\Theta}}\\log{p(\\boldsymbol{X};\\boldsymbol{\\Theta})}=\\nabla_{\\boldsymbol{\\Theta}}\\left(\\sum_{x_i\\in \\boldsymbol{X}}\\log{f(x_i;\\boldsymbol{\\Theta})}\\right)-\\nabla_{\\boldsymbol{\\Theta}}\\log{Z(\\boldsymbol{\\Theta})}=0.\n", @@ -374,8 +448,10 @@ }, { "cell_type": "markdown", - "id": "d364a111", - "metadata": {}, + "id": "ab06bb70", + "metadata": { + "editable": true + }, "source": [ "The first term is called the positive phase and we assume that we have a model for the function $f$ from which we can sample values. Below we will develop an explicit model for this.\n", "The second term is called the negative phase and is the one which leads to more difficulties." @@ -383,8 +459,10 @@ }, { "cell_type": "markdown", - "id": "102180ca", - "metadata": {}, + "id": "7f086749", + "metadata": { + "editable": true + }, "source": [ "## Contrastive optimization\n", "The evaluation of these two terms leads to what in the literature is called contrastive optimization.\n", @@ -396,8 +474,10 @@ }, { "cell_type": "markdown", - "id": "ae88482f", - "metadata": {}, + "id": "65f4974d", + "metadata": { + "editable": true + }, "source": [ "## The derivative of the partition function\n", "\n", @@ -406,8 +486,10 @@ }, { "cell_type": "markdown", - "id": "33407b19", - "metadata": {}, + "id": "01b6074d", + "metadata": { + "editable": true + }, "source": [ "$$\n", "Z(\\boldsymbol{\\Theta})=\\sum_{x_i\\in \\boldsymbol{X}}\\sum_{h_j\\in \\boldsymbol{H}} f(x_i,h_j;\\boldsymbol{\\Theta}),\n", @@ -416,16 +498,20 @@ }, { "cell_type": "markdown", - "id": "ae502918", - "metadata": {}, + "id": "abc68191", + "metadata": { + "editable": true + }, "source": [ "is in general the most problematic term. In principle both $x$ and $h$ can span large degrees of freedom, if not even infinitely many ones, and computing the partition function itself is often not desirable or even feasible. The above derivative of the partition function can however be written in terms of an expectation value which is in turn evaluated using Monte Carlo sampling and the theory of Markov chains, popularly shortened to MCMC (or just MC$^2$)." ] }, { "cell_type": "markdown", - "id": "ba65f40a", - "metadata": {}, + "id": "42d55ede", + "metadata": { + "editable": true + }, "source": [ "## Explicit expression for the derivative\n", "We can rewrite" @@ -433,8 +519,10 @@ }, { "cell_type": "markdown", - "id": "3ac7d02a", - "metadata": {}, + "id": "2713bc10", + "metadata": { + "editable": true + }, "source": [ "$$\n", "\\nabla_{\\boldsymbol{\\Theta}}\\log{Z(\\boldsymbol{\\Theta})}=\\frac{\\nabla_{\\boldsymbol{\\Theta}}Z(\\boldsymbol{\\Theta})}{Z(\\boldsymbol{\\Theta})},\n", @@ -443,16 +531,20 @@ }, { "cell_type": "markdown", - "id": "5fe20b8e", - "metadata": {}, + "id": "bcbc944d", + "metadata": { + "editable": true + }, "source": [ "which reads in more detail" ] }, { "cell_type": "markdown", - "id": "ba850619", - "metadata": {}, + "id": "f470f0b7", + "metadata": { + "editable": true + }, "source": [ "$$\n", "\\nabla_{\\boldsymbol{\\Theta}}\\log{Z(\\boldsymbol{\\Theta})}=\\frac{\\nabla_{\\boldsymbol{\\Theta}} \\sum_{x_i\\in \\boldsymbol{X}}f(x_i;\\boldsymbol{\\Theta}) }{Z(\\boldsymbol{\\Theta})}.\n", @@ -461,8 +553,10 @@ }, { "cell_type": "markdown", - "id": "771c8ff5", - "metadata": {}, + "id": "6cc4dc33", + "metadata": { + "editable": true + }, "source": [ "We can rewrite the function $f$ (we have assumed that is larger or\n", "equal than zero) as $f=\\exp{\\log{f}}$. We can then reqrite the last\n", @@ -471,8 +565,10 @@ }, { "cell_type": "markdown", - "id": "8a348ff2", - "metadata": {}, + "id": "b8d79f98", + "metadata": { + "editable": true + }, "source": [ "$$\n", "\\nabla_{\\boldsymbol{\\Theta}}\\log{Z(\\boldsymbol{\\Theta})}=\\frac{ \\sum_{x_i\\in \\boldsymbol{X}} \\nabla_{\\boldsymbol{\\Theta}}\\exp{\\log{f(x_i;\\boldsymbol{\\Theta})}} }{Z(\\boldsymbol{\\Theta})}.\n", @@ -481,8 +577,10 @@ }, { "cell_type": "markdown", - "id": "3d8854e8", - "metadata": {}, + "id": "133d8b3f", + "metadata": { + "editable": true + }, "source": [ "## Final expression\n", "\n", @@ -491,8 +589,10 @@ }, { "cell_type": "markdown", - "id": "4175d17f", - "metadata": {}, + "id": "9fc2ebb7", + "metadata": { + "editable": true + }, "source": [ "$$\n", "\\nabla_{\\boldsymbol{\\Theta}}\\log{Z(\\boldsymbol{\\Theta})}=\\frac{ \\sum_{x_i\\in \\boldsymbol{X}}f(x_i;\\boldsymbol{\\Theta}) \\nabla_{\\boldsymbol{\\Theta}}\\log{f(x_i;\\boldsymbol{\\Theta})} }{Z(\\boldsymbol{\\Theta})},\n", @@ -501,16 +601,20 @@ }, { "cell_type": "markdown", - "id": "efbbb764", - "metadata": {}, + "id": "431a67c4", + "metadata": { + "editable": true + }, "source": [ "which is the expectation value of $\\log{f}$" ] }, { "cell_type": "markdown", - "id": "5655190f", - "metadata": {}, + "id": "3df67560", + "metadata": { + "editable": true + }, "source": [ "$$\n", "\\nabla_{\\boldsymbol{\\Theta}}\\log{Z(\\boldsymbol{\\Theta})}=\\sum_{x_i\\in \\boldsymbol{X}}p(x_i;\\boldsymbol{\\Theta}) \\nabla_{\\boldsymbol{\\Theta}}\\log{f(x_i;\\boldsymbol{\\Theta})},\n", @@ -519,16 +623,20 @@ }, { "cell_type": "markdown", - "id": "bee7cbd0", - "metadata": {}, + "id": "0838e65c", + "metadata": { + "editable": true + }, "source": [ "that is" ] }, { "cell_type": "markdown", - "id": "b38e042d", - "metadata": {}, + "id": "822bf5ac", + "metadata": { + "editable": true + }, "source": [ "$$\n", "\\nabla_{\\boldsymbol{\\Theta}}\\log{Z(\\boldsymbol{\\Theta})}=\\mathbb{E}(\\log{f(x_i;\\boldsymbol{\\Theta})}).\n", @@ -537,8 +645,10 @@ }, { "cell_type": "markdown", - "id": "84b5a5c8", - "metadata": {}, + "id": "a93722a2", + "metadata": { + "editable": true + }, "source": [ "This quantity is evaluated using Monte Carlo sampling, with Gibbs\n", "sampling as the standard sampling rule." @@ -546,8 +656,10 @@ }, { "cell_type": "markdown", - "id": "60912295", - "metadata": {}, + "id": "d6d8c42e", + "metadata": { + "editable": true + }, "source": [ "## Generative model, basic overview (Borrowed from Rashcka et al)\n", "\n", @@ -560,8 +672,10 @@ }, { "cell_type": "markdown", - "id": "fa29e674", - "metadata": {}, + "id": "f17356a3", + "metadata": { + "editable": true + }, "source": [ "## Reminder on VAEs\n", "\n", @@ -577,8 +691,10 @@ }, { "cell_type": "markdown", - "id": "7c5cf97d", - "metadata": {}, + "id": "27fb099f", + "metadata": { + "editable": true + }, "source": [ "$$\n", "p(\\boldsymbol{x}) = \\int p(\\boldsymbol{x}, \\boldsymbol{h})d\\boldsymbol{h}\n", @@ -587,16 +703,20 @@ }, { "cell_type": "markdown", - "id": "42b0c6a4", - "metadata": {}, + "id": "17548236", + "metadata": { + "editable": true + }, "source": [ "or, we could also appeal to the chain rule of probability" ] }, { "cell_type": "markdown", - "id": "93d97d7b", - "metadata": {}, + "id": "3f05b416", + "metadata": { + "editable": true + }, "source": [ "$$\n", "p(\\boldsymbol{x}) = \\frac{p(\\boldsymbol{x}, \\boldsymbol{h})}{p(\\boldsymbol{h}|\\boldsymbol{x})}\n", @@ -605,16 +725,20 @@ }, { "cell_type": "markdown", - "id": "da374052", - "metadata": {}, + "id": "2cf43c02", + "metadata": { + "editable": true + }, "source": [ "We suppress here the dependence\ton the optimization parameters $\\boldsymbol{\\Theta}$." ] }, { "cell_type": "markdown", - "id": "e124fca0", - "metadata": {}, + "id": "4c9b8c03", + "metadata": { + "editable": true + }, "source": [ "## Evidence Lower Bound\n", "Directly computing and maximizing the likelihood $p(\\boldsymbol{x})$ is\n", @@ -633,8 +757,10 @@ }, { "cell_type": "markdown", - "id": "f0555155", - "metadata": {}, + "id": "587e829b", + "metadata": { + "editable": true + }, "source": [ "## ELBO equations\n", "Formally, the equation of the ELBO is" @@ -642,8 +768,10 @@ }, { "cell_type": "markdown", - "id": "5f61104a", - "metadata": {}, + "id": "6cec0936", + "metadata": { + "editable": true + }, "source": [ "$$\n", "\\mathbb{E}_{q_{\\boldsymbol{\\phi}}(\\boldsymbol{h}|\\boldsymbol{x})}\\left[\\log\\frac{p(\\boldsymbol{x}, \\boldsymbol{h})}{q_{\\boldsymbol{\\phi}}(\\boldsymbol{h}|\\boldsymbol{x})}\\right]\n", @@ -652,16 +780,20 @@ }, { "cell_type": "markdown", - "id": "4c3f6cec", - "metadata": {}, + "id": "008b0d93", + "metadata": { + "editable": true + }, "source": [ "To make the relationship with the evidence explicit, we can mathematically write:" ] }, { "cell_type": "markdown", - "id": "9740dc8e", - "metadata": {}, + "id": "03ec09b6", + "metadata": { + "editable": true + }, "source": [ "$$\n", "\\log p(\\boldsymbol{x}) \\geq \\mathbb{E}_{q_{\\boldsymbol{\\phi}}(\\boldsymbol{h}|\\boldsymbol{x})}\\left[\\log\\frac{p(\\boldsymbol{x}, \\boldsymbol{h})}{q_{\\boldsymbol{\\phi}}(\\boldsymbol{h}|\\boldsymbol{x})}\\right]\n", @@ -670,8 +802,10 @@ }, { "cell_type": "markdown", - "id": "2a5402dd", - "metadata": {}, + "id": "ee878746", + "metadata": { + "editable": true + }, "source": [ "## Introducing the encoder function\n", "\n", @@ -689,8 +823,10 @@ }, { "cell_type": "markdown", - "id": "3aa0f72f", - "metadata": {}, + "id": "be988b7e", + "metadata": { + "editable": true + }, "source": [ "## The derivation from last week\n", "\n", @@ -699,8 +835,10 @@ }, { "cell_type": "markdown", - "id": "ee186ef7", - "metadata": {}, + "id": "d06829bb", + "metadata": { + "editable": true + }, "source": [ "$$\n", "\\begin{align*}\n", @@ -719,8 +857,10 @@ }, { "cell_type": "markdown", - "id": "00908b7d", - "metadata": {}, + "id": "f165c01c", + "metadata": { + "editable": true + }, "source": [ "## Analysis\n", "\n", @@ -735,8 +875,10 @@ }, { "cell_type": "markdown", - "id": "d1df666e", - "metadata": {}, + "id": "a5a5ba44", + "metadata": { + "editable": true + }, "source": [ "## The VAE\n", "\n", @@ -751,8 +893,10 @@ }, { "cell_type": "markdown", - "id": "134efdf5", - "metadata": {}, + "id": "5a0edb20", + "metadata": { + "editable": true + }, "source": [ "## Dissecting the equations\n", "To make\n", @@ -761,8 +905,10 @@ }, { "cell_type": "markdown", - "id": "71c93e7e", - "metadata": {}, + "id": "066fb151", + "metadata": { + "editable": true + }, "source": [ "$$\n", "\\begin{align*}\n", @@ -776,8 +922,10 @@ }, { "cell_type": "markdown", - "id": "d50e4e2d", - "metadata": {}, + "id": "b1c188ac", + "metadata": { + "editable": true + }, "source": [ "## Bottlenecking distribution\n", "\n", @@ -792,8 +940,10 @@ }, { "cell_type": "markdown", - "id": "290915db", - "metadata": {}, + "id": "61b1725e", + "metadata": { + "editable": true + }, "source": [ "## Decoder and encoder\n", "The two terms in the last equation each have intuitive descriptions: the first\n", @@ -810,8 +960,10 @@ }, { "cell_type": "markdown", - "id": "6f7a335b", - "metadata": {}, + "id": "d134dd70", + "metadata": { + "editable": true + }, "source": [ "## Defining feature of VAEs\n", "\n", @@ -820,8 +972,10 @@ }, { "cell_type": "markdown", - "id": "865be5c3", - "metadata": {}, + "id": "c914ac07", + "metadata": { + "editable": true + }, "source": [ "$$\n", "\\begin{align*}\n", @@ -833,8 +987,10 @@ }, { "cell_type": "markdown", - "id": "504fe084", - "metadata": {}, + "id": "182be149", + "metadata": { + "editable": true + }, "source": [ "## Analytical evaluation\n", "\n", @@ -843,8 +999,10 @@ }, { "cell_type": "markdown", - "id": "60423328", - "metadata": {}, + "id": "d26c7329", + "metadata": { + "editable": true + }, "source": [ "$$\n", "\\begin{align*}\n", @@ -855,16 +1013,20 @@ }, { "cell_type": "markdown", - "id": "bf6ec7ac", - "metadata": {}, + "id": "2a89d0f3", + "metadata": { + "editable": true + }, "source": [ "where latents $\\{\\boldsymbol{h}^{(l)}\\}_{l=1}^L$ are sampled from $q_{\\boldsymbol{\\phi}}(\\boldsymbol{h}|\\boldsymbol{x})$, for every observation $\\boldsymbol{x}$ in the dataset." ] }, { "cell_type": "markdown", - "id": "1432c75e", - "metadata": {}, + "id": "7666826e", + "metadata": { + "editable": true + }, "source": [ "## Reparameterization trick\n", "\n", @@ -878,8 +1040,10 @@ }, { "cell_type": "markdown", - "id": "5578e8d1", - "metadata": {}, + "id": "a6ff50b2", + "metadata": { + "editable": true + }, "source": [ "## Actual implementation\n", "\n", @@ -893,8 +1057,10 @@ }, { "cell_type": "markdown", - "id": "e64959bb", - "metadata": {}, + "id": "1300c538", + "metadata": { + "editable": true + }, "source": [ "$$\n", "\\begin{align*}\n", @@ -905,8 +1071,10 @@ }, { "cell_type": "markdown", - "id": "2748e807", - "metadata": {}, + "id": "ebd27757", + "metadata": { + "editable": true + }, "source": [ "## Interpretation\n", "An arbitrary Gaussian distributions can be interpreted as\n", @@ -921,8 +1089,10 @@ }, { "cell_type": "markdown", - "id": "e75a9ac2", - "metadata": {}, + "id": "624b1963", + "metadata": { + "editable": true + }, "source": [ "## Deterministic function\n", "\n", @@ -931,8 +1101,10 @@ }, { "cell_type": "markdown", - "id": "44da9a14", - "metadata": {}, + "id": "eaf66572", + "metadata": { + "editable": true + }, "source": [ "$$\n", "\\begin{align*}\n", @@ -943,8 +1115,10 @@ }, { "cell_type": "markdown", - "id": "036c51f1", - "metadata": {}, + "id": "f4d09b90", + "metadata": { + "editable": true + }, "source": [ "where $\\odot$ represents an element-wise product. Under this\n", "reparameterized version of $\\boldsymbol{h}$, gradients can then be computed\n", @@ -957,8 +1131,10 @@ }, { "cell_type": "markdown", - "id": "28ad53de", - "metadata": {}, + "id": "df030b4e", + "metadata": { + "editable": true + }, "source": [ "## After training\n", "\n", @@ -974,8 +1150,10 @@ }, { "cell_type": "markdown", - "id": "efdb56d2", - "metadata": {}, + "id": "179d5913", + "metadata": { + "editable": true + }, "source": [ "## What is a GAN?\n", "\n", @@ -990,8 +1168,10 @@ }, { "cell_type": "markdown", - "id": "9edbf5b2", - "metadata": {}, + "id": "e954e4d0", + "metadata": { + "editable": true + }, "source": [ "## What is a generator network?\n", "\n", @@ -1009,8 +1189,10 @@ }, { "cell_type": "markdown", - "id": "4a9f2d71", - "metadata": {}, + "id": "9912338d", + "metadata": { + "editable": true + }, "source": [ "## And what is a discriminator network?\n", "\n", @@ -1019,8 +1201,10 @@ }, { "cell_type": "markdown", - "id": "64218644", - "metadata": {}, + "id": "9fadae1d", + "metadata": { + "editable": true + }, "source": [ "## Appplications of GANs\n", "\n", @@ -1042,8 +1226,10 @@ }, { "cell_type": "markdown", - "id": "815730d9", - "metadata": {}, + "id": "d205f4c0", + "metadata": { + "editable": true + }, "source": [ "## Discriminator versus generator (Borrowed from Rashcka et al)\n", "\n", @@ -1056,8 +1242,10 @@ }, { "cell_type": "markdown", - "id": "cd7735fd", - "metadata": {}, + "id": "8afeef95", + "metadata": { + "editable": true + }, "source": [ "## Generative Adversarial Networks\n", "\n", @@ -1072,8 +1260,10 @@ }, { "cell_type": "markdown", - "id": "ff76bbce", - "metadata": {}, + "id": "6e728830", + "metadata": { + "editable": true + }, "source": [ "$$\n", "x = g(z; \\theta^{(g)}).\n", @@ -1082,8 +1272,10 @@ }, { "cell_type": "markdown", - "id": "2ab9c200", - "metadata": {}, + "id": "5318fb35", + "metadata": { + "editable": true + }, "source": [ "## Discriminator\n", "\n", @@ -1096,8 +1288,10 @@ }, { "cell_type": "markdown", - "id": "dc4ca054", - "metadata": {}, + "id": "70ed2bca", + "metadata": { + "editable": true + }, "source": [ "$$\n", "d(x; \\theta^{(d)}).\n", @@ -1106,8 +1300,10 @@ }, { "cell_type": "markdown", - "id": "a7a1c19c", - "metadata": {}, + "id": "b2f662ca", + "metadata": { + "editable": true + }, "source": [ "indicating the probability that $x$ is a real training example rather than a\n", "fake sample the generator has generated." @@ -1115,8 +1311,10 @@ }, { "cell_type": "markdown", - "id": "492c4755", - "metadata": {}, + "id": "ca69c67d", + "metadata": { + "editable": true + }, "source": [ "## Zero-sum game\n", "\n", @@ -1127,8 +1325,10 @@ }, { "cell_type": "markdown", - "id": "6784d23c", - "metadata": {}, + "id": "4d8a5c1f", + "metadata": { + "editable": true + }, "source": [ "$$\n", "v(\\theta^{(g)}, \\theta^{(d)}),\n", @@ -1137,8 +1337,10 @@ }, { "cell_type": "markdown", - "id": "41ee9cb1", - "metadata": {}, + "id": "eaeafa28", + "metadata": { + "editable": true + }, "source": [ "determines the reward for the discriminator, while the generator gets the\n", "conjugate reward" @@ -1146,8 +1348,10 @@ }, { "cell_type": "markdown", - "id": "ff393ede", - "metadata": {}, + "id": "f35d0439", + "metadata": { + "editable": true + }, "source": [ "$$\n", "-v(\\theta^{(g)}, \\theta^{(d)})\n", @@ -1156,8 +1360,10 @@ }, { "cell_type": "markdown", - "id": "baef9203", - "metadata": {}, + "id": "e9df164f", + "metadata": { + "editable": true + }, "source": [ "## Maximizing reward\n", "\n", @@ -1177,8 +1383,10 @@ }, { "cell_type": "markdown", - "id": "22ed48ac", - "metadata": {}, + "id": "ac025bad", + "metadata": { + "editable": true + }, "source": [ "## Progression in training\n", "\n", @@ -1192,8 +1400,10 @@ }, { "cell_type": "markdown", - "id": "6153915a", - "metadata": {}, + "id": "1548d332", + "metadata": { + "editable": true + }, "source": [ "$$\n", "g^* = \\underset{g}{\\mathrm{argmin}}\\hspace{2pt}\n", @@ -1203,8 +1413,10 @@ }, { "cell_type": "markdown", - "id": "b69cf466", - "metadata": {}, + "id": "415e7020", + "metadata": { + "editable": true + }, "source": [ "## Deafault choice\n", "The default choice for $v$ is" @@ -1212,8 +1424,10 @@ }, { "cell_type": "markdown", - "id": "e989c0e6", - "metadata": {}, + "id": "afa15d31", + "metadata": { + "editable": true + }, "source": [ "$$\n", "v(\\theta^{(g)}, \\theta^{(d)}) = \\mathbb{E}_{x\\sim p_\\mathrm{data}}\\log d(x)\n", @@ -1224,8 +1438,10 @@ }, { "cell_type": "markdown", - "id": "f1e73429", - "metadata": {}, + "id": "db584443", + "metadata": { + "editable": true + }, "source": [ "## Design of GANs\n", "The main motivation for the design of GANs is that the learning process requires\n", @@ -1235,8 +1451,10 @@ }, { "cell_type": "markdown", - "id": "2fc2d44a", - "metadata": {}, + "id": "65ecceb4", + "metadata": { + "editable": true + }, "source": [ "$$\n", "\\underset{d}{\\mathrm{max}}v(\\theta^{(g)}, \\theta^{(d)})\n", @@ -1245,8 +1463,10 @@ }, { "cell_type": "markdown", - "id": "bb108909", - "metadata": {}, + "id": "93110eb4", + "metadata": { + "editable": true + }, "source": [ "is convex in $\\theta^{(g)}$ then the procedure is guaranteed to converge and is\n", "asymptotically consistent\n", @@ -1258,8 +1478,10 @@ }, { "cell_type": "markdown", - "id": "2fc42186", - "metadata": {}, + "id": "f5789340", + "metadata": { + "editable": true + }, "source": [ "## Steps in building a GAN (Borrowed from Rashcka et al)\n", "\n", @@ -1272,8 +1494,10 @@ }, { "cell_type": "markdown", - "id": "6e4bf240", - "metadata": {}, + "id": "4bbe2c2b", + "metadata": { + "editable": true + }, "source": [ "## More references\n", "\n", @@ -1289,8 +1513,10 @@ }, { "cell_type": "markdown", - "id": "375542b5", - "metadata": {}, + "id": "c0522fb3", + "metadata": { + "editable": true + }, "source": [ "## Writing Our First Generative Adversarial Network\n", "\n", @@ -1301,8 +1527,10 @@ }, { "cell_type": "markdown", - "id": "f451c1ae", - "metadata": {}, + "id": "fc2b9ad9", + "metadata": { + "editable": true + }, "source": [ "## Implementing the networks (Borrowed from Rashcka et al)\n", "\n", @@ -1315,8 +1543,10 @@ }, { "cell_type": "markdown", - "id": "420d32a6", - "metadata": {}, + "id": "45c07637", + "metadata": { + "editable": true + }, "source": [ "## Code elements" ] @@ -1324,18 +1554,12 @@ { "cell_type": "code", "execution_count": 1, - "id": "78c885cd", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "2.2.0\n", - "GPU Available: False\n" - ] - } - ], + "id": "b4dc82a1", + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], "source": [ "%matplotlib inline\n", "\n", @@ -1355,8 +1579,10 @@ }, { "cell_type": "markdown", - "id": "8c7a259d", - "metadata": {}, + "id": "f87b63c1", + "metadata": { + "editable": true + }, "source": [ "## Setting up the GAN" ] @@ -1364,8 +1590,11 @@ { "cell_type": "code", "execution_count": 2, - "id": "801940ac", - "metadata": {}, + "id": "139f0467", + "metadata": { + "collapsed": false, + "editable": true + }, "outputs": [], "source": [ "## define a function for the generator:\n", @@ -1410,8 +1639,10 @@ }, { "cell_type": "markdown", - "id": "de405acb", - "metadata": {}, + "id": "2c1f7b6f", + "metadata": { + "editable": true + }, "source": [ "## Printing the model" ] @@ -1419,22 +1650,12 @@ { "cell_type": "code", "execution_count": 3, - "id": "7c2211d1", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Sequential(\n", - " (fc_g0): Linear(in_features=20, out_features=100, bias=True)\n", - " (relu_g0): LeakyReLU(negative_slope=0.01)\n", - " (fc_g1): Linear(in_features=100, out_features=784, bias=True)\n", - " (tanh_g): Tanh()\n", - ")\n" - ] - } - ], + "id": "ad486a71", + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], "source": [ "image_size = (28, 28)\n", "z_size = 20\n", @@ -1458,23 +1679,12 @@ { "cell_type": "code", "execution_count": 4, - "id": "4aad7e10", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Sequential(\n", - " (fc_d0): Linear(in_features=784, out_features=100, bias=False)\n", - " (relu_d0): LeakyReLU(negative_slope=0.01)\n", - " (dropout): Dropout(p=0.5, inplace=False)\n", - " (fc_d1): Linear(in_features=100, out_features=1, bias=True)\n", - " (sigmoid): Sigmoid()\n", - ")\n" - ] - } - ], + "id": "632212e4", + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], "source": [ "disc_model = make_discriminator_network(\n", " input_size=np.prod(image_size),\n", @@ -1486,8 +1696,10 @@ }, { "cell_type": "markdown", - "id": "606bd08f", - "metadata": {}, + "id": "e204ed36", + "metadata": { + "editable": true + }, "source": [ "## Defining the training set" ] @@ -1495,28 +1707,12 @@ { "cell_type": "code", "execution_count": 5, - "id": "a521030b", - "metadata": {}, - "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/Users/mhjensen/miniforge3/envs/myenv/lib/python3.9/site-packages/torchvision/io/image.py:13: UserWarning: Failed to load image Python extension: 'dlopen(/Users/mhjensen/miniforge3/envs/myenv/lib/python3.9/site-packages/torchvision/image.so, 0x0006): Symbol not found: __ZN3c1017RegisterOperatorsD1Ev\n", - " Referenced from: <2D1B8D5C-7891-3680-9CF9-F771AE880676> /Users/mhjensen/miniforge3/envs/myenv/lib/python3.9/site-packages/torchvision/image.so\n", - " Expected in: /Users/mhjensen/miniforge3/envs/myenv/lib/python3.9/site-packages/torch/lib/libtorch_cpu.dylib'If you don't plan on using image functionality from `torchvision.io`, you can ignore this warning. Otherwise, there might be something wrong with your environment. Did you have `libjpeg` or `libpng` installed before building `torchvision` from source?\n", - " warn(\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Min: -1.0 Max: 1.0\n", - "torch.Size([1, 28, 28])\n" - ] - } - ], + "id": "0ca2ac08", + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], "source": [ "import torchvision \n", "from torchvision import transforms \n", @@ -1539,8 +1735,10 @@ }, { "cell_type": "markdown", - "id": "5e5ce63c", - "metadata": {}, + "id": "1791b464", + "metadata": { + "editable": true + }, "source": [ "## Defining the training set, part 2" ] @@ -1548,21 +1746,12 @@ { "cell_type": "code", "execution_count": 6, - "id": "0dfeaf46", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "input-z -- shape: torch.Size([32, 20])\n", - "input-real -- shape: torch.Size([32, 784])\n", - "Output of G -- shape: torch.Size([32, 784])\n", - "Disc. (real) -- shape: torch.Size([32, 1])\n", - "Disc. (fake) -- shape: torch.Size([32, 1])\n" - ] - } - ], + "id": "386b381a", + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], "source": [ "def create_noise(batch_size, z_size, mode_z):\n", " if mode_z == 'uniform':\n", @@ -1598,8 +1787,10 @@ }, { "cell_type": "markdown", - "id": "ad562406", - "metadata": {}, + "id": "f28eaf09", + "metadata": { + "editable": true + }, "source": [ "## Training the GAN" ] @@ -1607,18 +1798,12 @@ { "cell_type": "code", "execution_count": 7, - "id": "a09230d1", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Generator Loss: 0.6944\n", - "Discriminator Losses: Real 0.7758 Fake 0.6924\n" - ] - } - ], + "id": "352db36d", + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], "source": [ "loss_fn = nn.BCELoss()\n", "\n", @@ -1638,8 +1823,10 @@ }, { "cell_type": "markdown", - "id": "f4fde94c", - "metadata": {}, + "id": "d02e0c9a", + "metadata": { + "editable": true + }, "source": [ "## More on training" ] @@ -1647,8 +1834,11 @@ { "cell_type": "code", "execution_count": 8, - "id": "4f2f69bc", - "metadata": {}, + "id": "3c157e73", + "metadata": { + "collapsed": false, + "editable": true + }, "outputs": [], "source": [ "batch_size = 64\n", @@ -1726,118 +1916,12 @@ { "cell_type": "code", "execution_count": 9, - "id": "1c71de9b", - "metadata": { - "scrolled": true - }, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Epoch 001 | Avg Losses >> G/D 0.9169/0.8954 [D-Real: 0.8062 D-Fake: 0.4670]\n", - "Epoch 002 | Avg Losses >> G/D 1.0160/1.0904 [D-Real: 0.6298 D-Fake: 0.4150]\n", - "Epoch 003 | Avg Losses >> G/D 0.9259/1.2066 [D-Real: 0.5767 D-Fake: 0.4275]\n", - "Epoch 004 | Avg Losses >> G/D 0.8778/1.2460 [D-Real: 0.5619 D-Fake: 0.4424]\n", - "Epoch 005 | Avg Losses >> G/D 1.0375/1.1777 [D-Real: 0.5905 D-Fake: 0.4071]\n", - "Epoch 006 | Avg Losses >> G/D 0.9422/1.2174 [D-Real: 0.5754 D-Fake: 0.4265]\n", - "Epoch 007 | Avg Losses >> G/D 0.8936/1.2574 [D-Real: 0.5577 D-Fake: 0.4380]\n", - "Epoch 008 | Avg Losses >> G/D 0.8840/1.2712 [D-Real: 0.5521 D-Fake: 0.4427]\n", - "Epoch 009 | Avg Losses >> G/D 0.9981/1.1884 [D-Real: 0.5890 D-Fake: 0.4132]\n", - "Epoch 010 | Avg Losses >> G/D 1.0122/1.1749 [D-Real: 0.5968 D-Fake: 0.4100]\n", - "Epoch 011 | Avg Losses >> G/D 0.9869/1.1894 [D-Real: 0.5897 D-Fake: 0.4123]\n", - "Epoch 012 | Avg Losses >> G/D 0.8997/1.2510 [D-Real: 0.5631 D-Fake: 0.4351]\n", - "Epoch 013 | Avg Losses >> G/D 0.8892/1.2610 [D-Real: 0.5590 D-Fake: 0.4407]\n", - "Epoch 014 | Avg Losses >> G/D 0.8136/1.3101 [D-Real: 0.5358 D-Fake: 0.4603]\n", - "Epoch 015 | Avg Losses >> G/D 0.8842/1.2685 [D-Real: 0.5548 D-Fake: 0.4434]\n", - "Epoch 016 | Avg Losses >> G/D 0.8571/1.2733 [D-Real: 0.5551 D-Fake: 0.4495]\n", - "Epoch 017 | Avg Losses >> G/D 0.8477/1.2829 [D-Real: 0.5484 D-Fake: 0.4511]\n", - "Epoch 018 | Avg Losses >> G/D 0.8547/1.2757 [D-Real: 0.5539 D-Fake: 0.4497]\n", - "Epoch 019 | Avg Losses >> G/D 0.8674/1.2816 [D-Real: 0.5503 D-Fake: 0.4469]\n", - "Epoch 020 | Avg Losses >> G/D 0.8361/1.2930 [D-Real: 0.5453 D-Fake: 0.4552]\n", - "Epoch 021 | Avg Losses >> G/D 0.8203/1.3056 [D-Real: 0.5394 D-Fake: 0.4612]\n", - "Epoch 022 | Avg Losses >> G/D 0.8113/1.3134 [D-Real: 0.5353 D-Fake: 0.4629]\n", - "Epoch 023 | Avg Losses >> G/D 0.7900/1.3273 [D-Real: 0.5289 D-Fake: 0.4687]\n", - "Epoch 024 | Avg Losses >> G/D 0.7649/1.3443 [D-Real: 0.5201 D-Fake: 0.4761]\n", - "Epoch 025 | Avg Losses >> G/D 0.7903/1.3263 [D-Real: 0.5312 D-Fake: 0.4706]\n", - "Epoch 026 | Avg Losses >> G/D 0.7929/1.3269 [D-Real: 0.5294 D-Fake: 0.4694]\n", - "Epoch 027 | Avg Losses >> G/D 0.8207/1.3031 [D-Real: 0.5403 D-Fake: 0.4608]\n", - "Epoch 028 | Avg Losses >> G/D 0.8320/1.2921 [D-Real: 0.5470 D-Fake: 0.4578]\n", - "Epoch 029 | Avg Losses >> G/D 0.8349/1.2895 [D-Real: 0.5473 D-Fake: 0.4558]\n", - "Epoch 030 | Avg Losses >> G/D 0.8149/1.3041 [D-Real: 0.5400 D-Fake: 0.4598]\n", - "Epoch 031 | Avg Losses >> G/D 0.7898/1.3246 [D-Real: 0.5307 D-Fake: 0.4695]\n", - "Epoch 032 | Avg Losses >> G/D 0.7815/1.3337 [D-Real: 0.5259 D-Fake: 0.4711]\n", - "Epoch 033 | Avg Losses >> G/D 0.7767/1.3355 [D-Real: 0.5257 D-Fake: 0.4737]\n", - "Epoch 034 | Avg Losses >> G/D 0.7713/1.3398 [D-Real: 0.5235 D-Fake: 0.4759]\n", - "Epoch 035 | Avg Losses >> G/D 0.7910/1.3283 [D-Real: 0.5289 D-Fake: 0.4695]\n", - "Epoch 036 | Avg Losses >> G/D 0.7704/1.3392 [D-Real: 0.5248 D-Fake: 0.4761]\n", - "Epoch 037 | Avg Losses >> G/D 0.7568/1.3478 [D-Real: 0.5202 D-Fake: 0.4795]\n", - "Epoch 038 | Avg Losses >> G/D 0.7568/1.3468 [D-Real: 0.5202 D-Fake: 0.4795]\n", - "Epoch 039 | Avg Losses >> G/D 0.7662/1.3439 [D-Real: 0.5221 D-Fake: 0.4774]\n", - "Epoch 040 | Avg Losses >> G/D 0.7622/1.3452 [D-Real: 0.5205 D-Fake: 0.4781]\n", - "Epoch 041 | Avg Losses >> G/D 0.7661/1.3474 [D-Real: 0.5212 D-Fake: 0.4785]\n", - "Epoch 042 | Avg Losses >> G/D 0.7590/1.3475 [D-Real: 0.5200 D-Fake: 0.4793]\n", - "Epoch 043 | Avg Losses >> G/D 0.7678/1.3365 [D-Real: 0.5254 D-Fake: 0.4769]\n", - "Epoch 044 | Avg Losses >> G/D 0.7648/1.3441 [D-Real: 0.5216 D-Fake: 0.4781]\n", - "Epoch 045 | Avg Losses >> G/D 0.7657/1.3423 [D-Real: 0.5234 D-Fake: 0.4771]\n", - "Epoch 046 | Avg Losses >> G/D 0.7510/1.3517 [D-Real: 0.5171 D-Fake: 0.4811]\n", - "Epoch 047 | Avg Losses >> G/D 0.7572/1.3481 [D-Real: 0.5206 D-Fake: 0.4806]\n", - "Epoch 048 | Avg Losses >> G/D 0.7379/1.3625 [D-Real: 0.5130 D-Fake: 0.4856]\n", - "Epoch 049 | Avg Losses >> G/D 0.7403/1.3598 [D-Real: 0.5138 D-Fake: 0.4857]\n", - "Epoch 050 | Avg Losses >> G/D 0.7661/1.3426 [D-Real: 0.5233 D-Fake: 0.4784]\n", - "Epoch 051 | Avg Losses >> G/D 0.7614/1.3433 [D-Real: 0.5225 D-Fake: 0.4785]\n", - "Epoch 052 | Avg Losses >> G/D 0.7506/1.3582 [D-Real: 0.5146 D-Fake: 0.4827]\n", - "Epoch 053 | Avg Losses >> G/D 0.7430/1.3542 [D-Real: 0.5167 D-Fake: 0.4840]\n", - "Epoch 054 | Avg Losses >> G/D 0.7521/1.3467 [D-Real: 0.5198 D-Fake: 0.4806]\n", - "Epoch 055 | Avg Losses >> G/D 0.7447/1.3562 [D-Real: 0.5164 D-Fake: 0.4837]\n", - "Epoch 056 | Avg Losses >> G/D 0.7504/1.3561 [D-Real: 0.5164 D-Fake: 0.4824]\n", - "Epoch 057 | Avg Losses >> G/D 0.7604/1.3445 [D-Real: 0.5219 D-Fake: 0.4797]\n", - "Epoch 058 | Avg Losses >> G/D 0.7549/1.3510 [D-Real: 0.5190 D-Fake: 0.4812]\n", - "Epoch 059 | Avg Losses >> G/D 0.7474/1.3538 [D-Real: 0.5178 D-Fake: 0.4834]\n", - "Epoch 060 | Avg Losses >> G/D 0.7614/1.3452 [D-Real: 0.5218 D-Fake: 0.4793]\n", - "Epoch 061 | Avg Losses >> G/D 0.7549/1.3487 [D-Real: 0.5202 D-Fake: 0.4806]\n", - "Epoch 062 | Avg Losses >> G/D 0.7641/1.3471 [D-Real: 0.5204 D-Fake: 0.4785]\n", - "Epoch 063 | Avg Losses >> G/D 0.7610/1.3454 [D-Real: 0.5218 D-Fake: 0.4789]\n", - "Epoch 064 | Avg Losses >> G/D 0.7643/1.3462 [D-Real: 0.5211 D-Fake: 0.4784]\n", - "Epoch 065 | Avg Losses >> G/D 0.7704/1.3408 [D-Real: 0.5243 D-Fake: 0.4775]\n", - "Epoch 066 | Avg Losses >> G/D 0.7719/1.3356 [D-Real: 0.5256 D-Fake: 0.4752]\n", - "Epoch 067 | Avg Losses >> G/D 0.7720/1.3429 [D-Real: 0.5232 D-Fake: 0.4767]\n", - "Epoch 068 | Avg Losses >> G/D 0.7591/1.3485 [D-Real: 0.5205 D-Fake: 0.4803]\n", - "Epoch 069 | Avg Losses >> G/D 0.7603/1.3439 [D-Real: 0.5219 D-Fake: 0.4790]\n", - "Epoch 070 | Avg Losses >> G/D 0.7570/1.3469 [D-Real: 0.5211 D-Fake: 0.4801]\n", - "Epoch 071 | Avg Losses >> G/D 0.7606/1.3471 [D-Real: 0.5208 D-Fake: 0.4793]\n", - "Epoch 072 | Avg Losses >> G/D 0.7669/1.3405 [D-Real: 0.5232 D-Fake: 0.4769]\n", - "Epoch 073 | Avg Losses >> G/D 0.7632/1.3420 [D-Real: 0.5225 D-Fake: 0.4773]\n", - "Epoch 074 | Avg Losses >> G/D 0.7676/1.3439 [D-Real: 0.5222 D-Fake: 0.4769]\n", - "Epoch 075 | Avg Losses >> G/D 0.7617/1.3460 [D-Real: 0.5208 D-Fake: 0.4786]\n", - "Epoch 076 | Avg Losses >> G/D 0.7716/1.3401 [D-Real: 0.5246 D-Fake: 0.4766]\n", - "Epoch 077 | Avg Losses >> G/D 0.7707/1.3366 [D-Real: 0.5254 D-Fake: 0.4752]\n", - "Epoch 078 | Avg Losses >> G/D 0.7714/1.3368 [D-Real: 0.5253 D-Fake: 0.4753]\n", - "Epoch 079 | Avg Losses >> G/D 0.7715/1.3364 [D-Real: 0.5258 D-Fake: 0.4754]\n", - "Epoch 080 | Avg Losses >> G/D 0.7768/1.3364 [D-Real: 0.5257 D-Fake: 0.4747]\n", - "Epoch 081 | Avg Losses >> G/D 0.7791/1.3315 [D-Real: 0.5283 D-Fake: 0.4734]\n", - "Epoch 082 | Avg Losses >> G/D 0.7772/1.3333 [D-Real: 0.5268 D-Fake: 0.4737]\n", - "Epoch 083 | Avg Losses >> G/D 0.7769/1.3409 [D-Real: 0.5235 D-Fake: 0.4742]\n", - "Epoch 084 | Avg Losses >> G/D 0.7694/1.3397 [D-Real: 0.5237 D-Fake: 0.4764]\n", - "Epoch 085 | Avg Losses >> G/D 0.7730/1.3423 [D-Real: 0.5232 D-Fake: 0.4757]\n", - "Epoch 086 | Avg Losses >> G/D 0.7706/1.3411 [D-Real: 0.5239 D-Fake: 0.4772]\n", - "Epoch 087 | Avg Losses >> G/D 0.7557/1.3486 [D-Real: 0.5202 D-Fake: 0.4812]\n", - "Epoch 088 | Avg Losses >> G/D 0.7545/1.3531 [D-Real: 0.5169 D-Fake: 0.4804]\n", - "Epoch 089 | Avg Losses >> G/D 0.7639/1.3423 [D-Real: 0.5228 D-Fake: 0.4776]\n", - "Epoch 090 | Avg Losses >> G/D 0.7589/1.3506 [D-Real: 0.5194 D-Fake: 0.4805]\n", - "Epoch 091 | Avg Losses >> G/D 0.7683/1.3395 [D-Real: 0.5240 D-Fake: 0.4769]\n", - "Epoch 092 | Avg Losses >> G/D 0.7710/1.3368 [D-Real: 0.5254 D-Fake: 0.4752]\n", - "Epoch 093 | Avg Losses >> G/D 0.7585/1.3461 [D-Real: 0.5206 D-Fake: 0.4792]\n", - "Epoch 094 | Avg Losses >> G/D 0.7725/1.3386 [D-Real: 0.5241 D-Fake: 0.4751]\n", - "Epoch 095 | Avg Losses >> G/D 0.7635/1.3487 [D-Real: 0.5206 D-Fake: 0.4796]\n", - "Epoch 096 | Avg Losses >> G/D 0.7696/1.3405 [D-Real: 0.5238 D-Fake: 0.4763]\n", - "Epoch 097 | Avg Losses >> G/D 0.7582/1.3470 [D-Real: 0.5208 D-Fake: 0.4803]\n", - "Epoch 098 | Avg Losses >> G/D 0.7640/1.3435 [D-Real: 0.5219 D-Fake: 0.4779]\n", - "Epoch 099 | Avg Losses >> G/D 0.7636/1.3476 [D-Real: 0.5214 D-Fake: 0.4791]\n", - "Epoch 100 | Avg Losses >> G/D 0.7547/1.3481 [D-Real: 0.5197 D-Fake: 0.4801]\n" - ] - } - ], + "id": "639686f4", + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], "source": [ "fixed_z = create_noise(batch_size, z_size, mode_z).to(device)\n", "\n", @@ -1880,8 +1964,10 @@ }, { "cell_type": "markdown", - "id": "89c2ce43", - "metadata": {}, + "id": "4a7df60c", + "metadata": { + "editable": true + }, "source": [ "## Visualizing" ] @@ -1889,20 +1975,12 @@ { "cell_type": "code", "execution_count": 10, - "id": "d2f5e678", - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], + "id": "c575fd26", + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], "source": [ "import itertools\n", "\n", @@ -1934,20 +2012,12 @@ { "cell_type": "code", "execution_count": 11, - "id": "753c7a92", - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], + "id": "15299086", + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], "source": [ "selected_epochs = [1, 2, 4, 10, 50, 100]\n", "fig = plt.figure(figsize=(10, 14))\n", @@ -1973,8 +2043,10 @@ }, { "cell_type": "markdown", - "id": "5a43d263", - "metadata": {}, + "id": "f4b675b8", + "metadata": { + "editable": true + }, "source": [ "## Calculating scores" ] @@ -1982,8 +2054,11 @@ { "cell_type": "code", "execution_count": 12, - "id": "4c0ee2ef", - "metadata": {}, + "id": "4f8dd76d", + "metadata": { + "collapsed": false, + "editable": true + }, "outputs": [], "source": [ "import math\n", @@ -2034,8 +2109,10 @@ }, { "cell_type": "markdown", - "id": "369cd25f", - "metadata": {}, + "id": "604b1a6b", + "metadata": { + "editable": true + }, "source": [ "## Diffusion models, basics\n", "\n", @@ -2049,8 +2126,10 @@ }, { "cell_type": "markdown", - "id": "bbd17b58", - "metadata": {}, + "id": "69adfd20", + "metadata": { + "editable": true + }, "source": [ "## Problems with probabilistic models\n", "\n", @@ -2072,8 +2151,10 @@ }, { "cell_type": "markdown", - "id": "105eb19e", - "metadata": {}, + "id": "db15b3f4", + "metadata": { + "editable": true + }, "source": [ "## Diffusion models\n", "Diffusion models have several interesting features\n", @@ -2088,8 +2169,10 @@ }, { "cell_type": "markdown", - "id": "07aaadd2", - "metadata": {}, + "id": "8feb45ca", + "metadata": { + "editable": true + }, "source": [ "## Original idea\n", "\n", @@ -2107,8 +2190,10 @@ }, { "cell_type": "markdown", - "id": "b233ca01", - "metadata": {}, + "id": "72d44bc3", + "metadata": { + "editable": true + }, "source": [ "## Diffusion learning\n", "\n", @@ -2122,25 +2207,7 @@ ] } ], - "metadata": { - "kernelspec": { - "display_name": "Python 3 (ipykernel)", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.9.18" - } - }, + "metadata": {}, "nbformat": 4, "nbformat_minor": 5 } diff --git a/doc/pub/week15/pdf/week15.pdf b/doc/pub/week15/pdf/week15.pdf index 013cee95..4aa89051 100644 Binary files a/doc/pub/week15/pdf/week15.pdf and b/doc/pub/week15/pdf/week15.pdf differ diff --git a/doc/src/week15/week15.do.txt b/doc/src/week15/week15.do.txt index 4ebce1ad..83739c6e 100644 --- a/doc/src/week15/week15.do.txt +++ b/doc/src/week15/week15.do.txt @@ -8,7 +8,7 @@ DATE: April 30, 2024 !bblock Deep generative models o Summary of Variational Autoencoders -o Generative Adversarial Networks (GANs) +o Generative Adversarial Networks (GANs), see URL:"https://lilianweng.github.io/posts/2017-08-20-gan/" for nice overview o Start discussion of diffusion models o "Video of lecture":"https://youtu.be/Cg8n9aWwHuU" o "Whiteboard notes":"https://github.com/CompPhysics/AdvancedMachineLearning/blob/main/doc/HandwrittenNotes/2024/NotesApril30.pdf"