From 84a4e5df0c490eaecaf5e9fb286f1f86b5d159c1 Mon Sep 17 00:00:00 2001 From: Morten Hjorth-Jensen Date: Thu, 18 Sep 2025 06:35:28 +0200 Subject: [PATCH] added video on bias-variance --- doc/pub/week38/html/._week38-bs001.html | 1 + doc/pub/week38/html/week38-reveal.html | 1 + doc/pub/week38/html/week38-solarized.html | 1 + doc/pub/week38/html/week38.html | 1 + doc/pub/week38/ipynb/ipynb-week38-src.tar.gz | Bin 1022756 -> 1022756 bytes doc/pub/week38/ipynb/week38.ipynb | 401 +++++++------------ doc/src/week38/week38.do.txt | 6 + 7 files changed, 147 insertions(+), 264 deletions(-) diff --git a/doc/pub/week38/html/._week38-bs001.html b/doc/pub/week38/html/._week38-bs001.html index a04cfdd61..e252e97e3 100644 --- a/doc/pub/week38/html/._week38-bs001.html +++ b/doc/pub/week38/html/._week38-bs001.html @@ -261,6 +261,7 @@ MathJax.Hub.Config({
  • Resampling techniques, Bootstrap and cross validation and bias-variance tradeoff
  • The material we did not cover last week, that is on more advanced methods for updating the learning rate, are covered by its own video. We will briefly discuss these topics at the beginning of the lecture and during the lab sessions. See video on ADAgrad, RMSprop and ADAM (material from last week not covered during lecture) at https://youtu.be/J_41Hld6tTU
  • Video of Lecture
    • +
  • Video from lab sessions on the bias-variance tradeoff
  • Whiteboard notes
    • diff --git a/doc/pub/week38/html/week38-reveal.html b/doc/pub/week38/html/week38-reveal.html index 0271e775a..6fbf2ca21 100644 --- a/doc/pub/week38/html/week38-reveal.html +++ b/doc/pub/week38/html/week38-reveal.html @@ -202,6 +202,7 @@ MathJax.Hub.Config({

  • Resampling techniques, Bootstrap and cross validation and bias-variance tradeoff

  • The material we did not cover last week, that is on more advanced methods for updating the learning rate, are covered by its own video. We will briefly discuss these topics at the beginning of the lecture and during the lab sessions. See video on ADAgrad, RMSprop and ADAM (material from last week not covered during lecture) at https://youtu.be/J_41Hld6tTU

  • Video of Lecture
    • +

  • Video from lab sessions on the bias-variance tradeoff

  • Whiteboard notes
    • diff --git a/doc/pub/week38/html/week38-solarized.html b/doc/pub/week38/html/week38-solarized.html index 43f7fbd86..391bf4621 100644 --- a/doc/pub/week38/html/week38-solarized.html +++ b/doc/pub/week38/html/week38-solarized.html @@ -240,6 +240,7 @@ MathJax.Hub.Config({
  • Resampling techniques, Bootstrap and cross validation and bias-variance tradeoff
  • The material we did not cover last week, that is on more advanced methods for updating the learning rate, are covered by its own video. We will briefly discuss these topics at the beginning of the lecture and during the lab sessions. See video on ADAgrad, RMSprop and ADAM (material from last week not covered during lecture) at https://youtu.be/J_41Hld6tTU
  • Video of Lecture
    • +
  • Video from lab sessions on the bias-variance tradeoff
  • Whiteboard notes
    • diff --git a/doc/pub/week38/html/week38.html b/doc/pub/week38/html/week38.html index 57e0831cb..8a07c4a1c 100644 --- a/doc/pub/week38/html/week38.html +++ b/doc/pub/week38/html/week38.html @@ -317,6 +317,7 @@ MathJax.Hub.Config({
  • Resampling techniques, Bootstrap and cross validation and bias-variance tradeoff
  • The material we did not cover last week, that is on more advanced methods for updating the learning rate, are covered by its own video. We will briefly discuss these topics at the beginning of the lecture and during the lab sessions. See video on ADAgrad, RMSprop and ADAM (material from last week not covered during lecture) at https://youtu.be/J_41Hld6tTU
  • Video of Lecture
    • +
  • Video from lab sessions on the bias-variance tradeoff
  • Whiteboard notes
    • diff --git a/doc/pub/week38/ipynb/ipynb-week38-src.tar.gz b/doc/pub/week38/ipynb/ipynb-week38-src.tar.gz index 99906213f2606093f500fa8fa4282242a746ea9d..1e2fa70f2c50172f0444175f1d0d1f957a4829cf 100644 GIT binary patch delta 63 zcmWN_IT3&`002S$3%>#eC-FvcRV2{H0+>JoW?;+Bcg%NWPoa3zb0DQuQcEMPbn<7A OUIrOul3A9Ebvr*&RS-7- delta 63 zcmWN_ISqh7002S$J$?m*1R>{*Vhe%<+Oq%}kOfmtdSZGa>lw0Rdv_$4LQ1KmmPY;z P(n=@23^K}eGcVT%EG!Rs diff --git a/doc/pub/week38/ipynb/week38.ipynb b/doc/pub/week38/ipynb/week38.ipynb index aea50a762..cd2b6ab03 100644 --- a/doc/pub/week38/ipynb/week38.ipynb +++ b/doc/pub/week38/ipynb/week38.ipynb @@ -2,7 +2,7 @@ "cells": [ { "cell_type": "markdown", - "id": "8f27372d", + "id": "cd058661", "metadata": { "editable": true }, @@ -14,7 +14,7 @@ }, { "cell_type": "markdown", - "id": "fff8ca30", + "id": "bb0e0285", "metadata": { "editable": true }, @@ -27,7 +27,7 @@ }, { "cell_type": "markdown", - "id": "7ee7e714", + "id": "5d0bf374", "metadata": { "editable": true }, @@ -44,12 +44,14 @@ "\n", "4. [Video of Lecture](https://youtu.be/4Fo7ITVA7V4)\n", "\n", - "5. [Whiteboard notes](https://github.com/CompPhysics/MachineLearning/blob/master/doc/HandWrittenNotes/2025/FYSSTKweek38.pdf)" + "5. [Video from lab sessions on the bias-variance tradeoff](https://youtu.be/GBWc1abChKo)\n", + "\n", + "6. [Whiteboard notes](https://github.com/CompPhysics/MachineLearning/blob/master/doc/HandWrittenNotes/2025/FYSSTKweek38.pdf)" ] }, { "cell_type": "markdown", - "id": "3b5ac440", + "id": "38a10c06", "metadata": { "editable": true }, @@ -70,7 +72,7 @@ }, { "cell_type": "markdown", - "id": "6d5dba52", + "id": "2beeb82a", "metadata": { "editable": true }, @@ -98,7 +100,7 @@ }, { "cell_type": "markdown", - "id": "bfc2983a", + "id": "84021a7f", "metadata": { "editable": true }, @@ -114,7 +116,7 @@ }, { "cell_type": "markdown", - "id": "2b5f5980", + "id": "1291c926", "metadata": { "editable": true }, @@ -133,7 +135,7 @@ }, { "cell_type": "markdown", - "id": "3464c7e8", + "id": "bf15a73d", "metadata": { "editable": true }, @@ -147,7 +149,7 @@ }, { "cell_type": "markdown", - "id": "ed0fd2df", + "id": "ed7830e9", "metadata": { "editable": true }, @@ -159,7 +161,7 @@ }, { "cell_type": "markdown", - "id": "feb9d4c2", + "id": "b1d75235", "metadata": { "editable": true }, @@ -170,7 +172,7 @@ }, { "cell_type": "markdown", - "id": "eb6d71f8", + "id": "0255cd11", "metadata": { "editable": true }, @@ -182,7 +184,7 @@ }, { "cell_type": "markdown", - "id": "566399f6", + "id": "04897143", "metadata": { "editable": true }, @@ -194,7 +196,7 @@ }, { "cell_type": "markdown", - "id": "6b33f497", + "id": "2a6cea60", "metadata": { "editable": true }, @@ -210,7 +212,7 @@ }, { "cell_type": "markdown", - "id": "5f2f79f2", + "id": "08eb2262", "metadata": { "editable": true }, @@ -221,7 +223,7 @@ }, { "cell_type": "markdown", - "id": "199121b0", + "id": "0f36d3c2", "metadata": { "editable": true }, @@ -244,7 +246,7 @@ }, { "cell_type": "markdown", - "id": "9a1cc529", + "id": "ea74022f", "metadata": { "editable": true }, @@ -255,7 +257,7 @@ }, { "cell_type": "markdown", - "id": "149e63be", + "id": "d6eba03b", "metadata": { "editable": true }, @@ -267,7 +269,7 @@ }, { "cell_type": "markdown", - "id": "6a6fb04a", + "id": "b8a7314f", "metadata": { "editable": true }, @@ -279,7 +281,7 @@ }, { "cell_type": "markdown", - "id": "79420d06", + "id": "ed668c22", "metadata": { "editable": true }, @@ -293,7 +295,7 @@ }, { "cell_type": "markdown", - "id": "0e3de992", + "id": "6f4ab09a", "metadata": { "editable": true }, @@ -324,7 +326,7 @@ }, { "cell_type": "markdown", - "id": "d3ea2897", + "id": "7b7808c7", "metadata": { "editable": true }, @@ -346,7 +348,7 @@ }, { "cell_type": "markdown", - "id": "da5e3927", + "id": "456afe19", "metadata": { "editable": true }, @@ -358,7 +360,7 @@ }, { "cell_type": "markdown", - "id": "7ab5488b", + "id": "0a38fc64", "metadata": { "editable": true }, @@ -371,7 +373,7 @@ }, { "cell_type": "markdown", - "id": "f904a739", + "id": "851bebe1", "metadata": { "editable": true }, @@ -383,7 +385,7 @@ }, { "cell_type": "markdown", - "id": "10fd648b", + "id": "fe64e9b5", "metadata": { "editable": true }, @@ -395,7 +397,7 @@ }, { "cell_type": "markdown", - "id": "4812c2a4", + "id": "496492d5", "metadata": { "editable": true }, @@ -407,7 +409,7 @@ }, { "cell_type": "markdown", - "id": "199d8531", + "id": "503eb7b2", "metadata": { "editable": true }, @@ -419,7 +421,7 @@ }, { "cell_type": "markdown", - "id": "96c16676", + "id": "1a33763c", "metadata": { "editable": true }, @@ -442,7 +444,7 @@ }, { "cell_type": "markdown", - "id": "a2a1a004", + "id": "70a645e3", "metadata": { "editable": true }, @@ -454,7 +456,7 @@ }, { "cell_type": "markdown", - "id": "5aad445b", + "id": "7fced9cb", "metadata": { "editable": true }, @@ -467,7 +469,7 @@ }, { "cell_type": "markdown", - "id": "d197c8bb", + "id": "313c05af", "metadata": { "editable": true }, @@ -479,7 +481,7 @@ }, { "cell_type": "markdown", - "id": "e2e7462f", + "id": "66eeeef9", "metadata": { "editable": true }, @@ -491,7 +493,7 @@ }, { "cell_type": "markdown", - "id": "eb635d3d", + "id": "cda5e4d2", "metadata": { "editable": true }, @@ -503,7 +505,7 @@ }, { "cell_type": "markdown", - "id": "445ed13e", + "id": "2e6ed5cd", "metadata": { "editable": true }, @@ -514,7 +516,7 @@ }, { "cell_type": "markdown", - "id": "319bfc6c", + "id": "ba81d29e", "metadata": { "editable": true }, @@ -526,7 +528,7 @@ }, { "cell_type": "markdown", - "id": "90abf35a", + "id": "26e2d548", "metadata": { "editable": true }, @@ -537,7 +539,7 @@ }, { "cell_type": "markdown", - "id": "04b66fbd", + "id": "0d5ef8ad", "metadata": { "editable": true }, @@ -549,7 +551,7 @@ }, { "cell_type": "markdown", - "id": "4a27b5a7", + "id": "b6c5763c", "metadata": { "editable": true }, @@ -559,7 +561,7 @@ }, { "cell_type": "markdown", - "id": "8d12543f", + "id": "e4afd86f", "metadata": { "editable": true }, @@ -590,7 +592,7 @@ }, { "cell_type": "markdown", - "id": "2e5cd118", + "id": "03d912b0", "metadata": { "editable": true }, @@ -602,7 +604,7 @@ }, { "cell_type": "markdown", - "id": "c71a5edf", + "id": "fef4cb78", "metadata": { "editable": true }, @@ -614,7 +616,7 @@ }, { "cell_type": "markdown", - "id": "e663bf2e", + "id": "000125c6", "metadata": { "editable": true }, @@ -624,7 +626,7 @@ }, { "cell_type": "markdown", - "id": "c4bc4873", + "id": "4f665607", "metadata": { "editable": true }, @@ -636,7 +638,7 @@ }, { "cell_type": "markdown", - "id": "f5bc59b8", + "id": "5f5877fa", "metadata": { "editable": true }, @@ -646,7 +648,7 @@ }, { "cell_type": "markdown", - "id": "4f6ddf4a", + "id": "1c342299", "metadata": { "editable": true }, @@ -658,7 +660,7 @@ }, { "cell_type": "markdown", - "id": "afda0a6b", + "id": "4b155a17", "metadata": { "editable": true }, @@ -668,7 +670,7 @@ }, { "cell_type": "markdown", - "id": "b5335dc0", + "id": "c23eaf84", "metadata": { "editable": true }, @@ -680,7 +682,7 @@ }, { "cell_type": "markdown", - "id": "4f86a52d", + "id": "7699c6f7", "metadata": { "editable": true }, @@ -690,7 +692,7 @@ }, { "cell_type": "markdown", - "id": "5cdb1767", + "id": "84c9b69d", "metadata": { "editable": true }, @@ -709,7 +711,7 @@ }, { "cell_type": "markdown", - "id": "69435d77", + "id": "59e6b611", "metadata": { "editable": true }, @@ -737,7 +739,7 @@ }, { "cell_type": "markdown", - "id": "cefbb559", + "id": "3ea44242", "metadata": { "editable": true }, @@ -763,7 +765,7 @@ }, { "cell_type": "markdown", - "id": "2659401a", + "id": "a98de365", "metadata": { "editable": true }, @@ -780,7 +782,7 @@ }, { "cell_type": "markdown", - "id": "4d5d7748", + "id": "2fd2ca6a", "metadata": { "editable": true }, @@ -800,7 +802,7 @@ }, { "cell_type": "markdown", - "id": "54df92b3", + "id": "87ab1f2b", "metadata": { "editable": true }, @@ -829,7 +831,7 @@ }, { "cell_type": "markdown", - "id": "5b1a1390", + "id": "88ffab6d", "metadata": { "editable": true }, @@ -854,7 +856,7 @@ }, { "cell_type": "markdown", - "id": "39f233e4", + "id": "96fabf7e", "metadata": { "editable": true }, @@ -874,7 +876,7 @@ }, { "cell_type": "markdown", - "id": "361320d8", + "id": "6e876164", "metadata": { "editable": true }, @@ -886,7 +888,7 @@ }, { "cell_type": "markdown", - "id": "a363db1e", + "id": "2b00fa3c", "metadata": { "editable": true }, @@ -896,7 +898,7 @@ }, { "cell_type": "markdown", - "id": "92967efc", + "id": "75d6acad", "metadata": { "editable": true }, @@ -911,7 +913,7 @@ }, { "cell_type": "markdown", - "id": "1bffca97", + "id": "8b412a9e", "metadata": { "editable": true }, @@ -924,7 +926,7 @@ }, { "cell_type": "markdown", - "id": "0dacb6fc", + "id": "3bdb59e7", "metadata": { "editable": true }, @@ -937,7 +939,7 @@ }, { "cell_type": "markdown", - "id": "baeedf81", + "id": "d709f4c1", "metadata": { "editable": true }, @@ -949,7 +951,7 @@ }, { "cell_type": "markdown", - "id": "20cc7770", + "id": "bf40508f", "metadata": { "editable": true }, @@ -962,7 +964,7 @@ }, { "cell_type": "markdown", - "id": "f67d3b94", + "id": "8b2b63fe", "metadata": { "editable": true }, @@ -973,7 +975,7 @@ }, { "cell_type": "markdown", - "id": "17f59fb6", + "id": "4c1720db", "metadata": { "editable": true }, @@ -987,7 +989,7 @@ }, { "cell_type": "markdown", - "id": "5f899fbe", + "id": "5aba4a1e", "metadata": { "editable": true }, @@ -997,7 +999,7 @@ }, { "cell_type": "markdown", - "id": "19a1f5bb", + "id": "00a5fc23", "metadata": { "editable": true }, @@ -1011,7 +1013,7 @@ }, { "cell_type": "markdown", - "id": "1db8fcf2", + "id": "354b2ab3", "metadata": { "editable": true }, @@ -1024,7 +1026,7 @@ }, { "cell_type": "markdown", - "id": "bfadf7e5", + "id": "2ee3d80b", "metadata": { "editable": true }, @@ -1037,7 +1039,7 @@ }, { "cell_type": "markdown", - "id": "7c65ce24", + "id": "5c6f424d", "metadata": { "editable": true }, @@ -1047,7 +1049,7 @@ }, { "cell_type": "markdown", - "id": "8cd5650a", + "id": "5b6a9003", "metadata": { "editable": true }, @@ -1060,7 +1062,7 @@ }, { "cell_type": "markdown", - "id": "11fdc936", + "id": "88b7b6c2", "metadata": { "editable": true }, @@ -1070,7 +1072,7 @@ }, { "cell_type": "markdown", - "id": "ed88642e", + "id": "bb8051d4", "metadata": { "editable": true }, @@ -1083,7 +1085,7 @@ }, { "cell_type": "markdown", - "id": "82c61b81", + "id": "4950aac9", "metadata": { "editable": true }, @@ -1095,7 +1097,7 @@ }, { "cell_type": "markdown", - "id": "bc43db46", + "id": "6d705546", "metadata": { "editable": true }, @@ -1114,7 +1116,7 @@ }, { "cell_type": "markdown", - "id": "25418113", + "id": "749b506b", "metadata": { "editable": true }, @@ -1127,7 +1129,7 @@ }, { "cell_type": "markdown", - "id": "e5d3c3eb", + "id": "02d5afea", "metadata": { "editable": true }, @@ -1139,7 +1141,7 @@ }, { "cell_type": "markdown", - "id": "c504cba4", + "id": "2664f854", "metadata": { "editable": true }, @@ -1152,7 +1154,7 @@ }, { "cell_type": "markdown", - "id": "079ded2a", + "id": "a986ee46", "metadata": { "editable": true }, @@ -1172,7 +1174,7 @@ }, { "cell_type": "markdown", - "id": "e8534a50", + "id": "f21341e3", "metadata": { "editable": true }, @@ -1195,7 +1197,7 @@ }, { "cell_type": "markdown", - "id": "2fc73431", + "id": "b22eb043", "metadata": { "editable": true }, @@ -1210,7 +1212,7 @@ }, { "cell_type": "markdown", - "id": "0f8b0845", + "id": "a2e2c4c5", "metadata": { "editable": true }, @@ -1222,7 +1224,7 @@ }, { "cell_type": "markdown", - "id": "25105753", + "id": "be028ae6", "metadata": { "editable": true }, @@ -1242,7 +1244,7 @@ }, { "cell_type": "markdown", - "id": "89be6eea", + "id": "e746545b", "metadata": { "editable": true }, @@ -1262,7 +1264,7 @@ }, { "cell_type": "markdown", - "id": "6c240b38", + "id": "dea3037c", "metadata": { "editable": true }, @@ -1286,7 +1288,7 @@ }, { "cell_type": "markdown", - "id": "fbd95a5c", + "id": "fd576cb1", "metadata": { "editable": true }, @@ -1307,7 +1309,7 @@ }, { "cell_type": "markdown", - "id": "dc50d43a", + "id": "8629a2e8", "metadata": { "editable": true }, @@ -1337,7 +1339,7 @@ }, { "cell_type": "markdown", - "id": "283068cc", + "id": "ab8c1d8a", "metadata": { "editable": true }, @@ -1361,13 +1363,10 @@ { "cell_type": "code", "execution_count": 1, - "id": "ff4790ba", + "id": "d7b87cf8", "metadata": { "collapsed": false, - "editable": true, - "jupyter": { - "outputs_hidden": false - } + "editable": true }, "outputs": [], "source": [ @@ -1403,7 +1402,7 @@ }, { "cell_type": "markdown", - "id": "3e6adc2f", + "id": "d57a0c6c", "metadata": { "editable": true }, @@ -1413,7 +1412,7 @@ }, { "cell_type": "markdown", - "id": "6ec8223c", + "id": "bd8574db", "metadata": { "editable": true }, @@ -1424,13 +1423,10 @@ { "cell_type": "code", "execution_count": 2, - "id": "3cf4144d", + "id": "5715940c", "metadata": { "collapsed": false, - "editable": true, - "jupyter": { - "outputs_hidden": false - } + "editable": true }, "outputs": [], "source": [ @@ -1447,7 +1443,7 @@ }, { "cell_type": "markdown", - "id": "db5a8f91", + "id": "9584858b", "metadata": { "editable": true }, @@ -1465,7 +1461,7 @@ }, { "cell_type": "markdown", - "id": "327bce6a", + "id": "6f3cee73", "metadata": { "editable": true }, @@ -1477,7 +1473,7 @@ }, { "cell_type": "markdown", - "id": "1c671d4e", + "id": "fecd4f4b", "metadata": { "editable": true }, @@ -1494,7 +1490,7 @@ }, { "cell_type": "markdown", - "id": "6e05fc43", + "id": "1bf50201", "metadata": { "editable": true }, @@ -1506,7 +1502,7 @@ }, { "cell_type": "markdown", - "id": "c45e0752", + "id": "aa1ee75a", "metadata": { "editable": true }, @@ -1516,7 +1512,7 @@ }, { "cell_type": "markdown", - "id": "bafa4ab6", + "id": "0b88cfa1", "metadata": { "editable": true }, @@ -1528,7 +1524,7 @@ }, { "cell_type": "markdown", - "id": "ea0bc471", + "id": "51802535", "metadata": { "editable": true }, @@ -1545,7 +1541,7 @@ }, { "cell_type": "markdown", - "id": "08b603f3", + "id": "c1fab3ca", "metadata": { "editable": true }, @@ -1557,7 +1553,7 @@ }, { "cell_type": "markdown", - "id": "4114d10e", + "id": "bf1b97b3", "metadata": { "editable": true }, @@ -1567,7 +1563,7 @@ }, { "cell_type": "markdown", - "id": "8890c666", + "id": "4e6a9591", "metadata": { "editable": true }, @@ -1579,7 +1575,7 @@ }, { "cell_type": "markdown", - "id": "7d5b7ce4", + "id": "3bec9e3c", "metadata": { "editable": true }, @@ -1589,7 +1585,7 @@ }, { "cell_type": "markdown", - "id": "3913c5b9", + "id": "a65f2f18", "metadata": { "editable": true }, @@ -1601,7 +1597,7 @@ }, { "cell_type": "markdown", - "id": "5e0067b1", + "id": "d73eda6c", "metadata": { "editable": true }, @@ -1611,7 +1607,7 @@ }, { "cell_type": "markdown", - "id": "326bc8f1", + "id": "ecc681f6", "metadata": { "editable": true }, @@ -1627,7 +1623,7 @@ }, { "cell_type": "markdown", - "id": "d3713eca", + "id": "0b1fdbf0", "metadata": { "editable": true }, @@ -1638,13 +1634,10 @@ { "cell_type": "code", "execution_count": 3, - "id": "01c3b507", + "id": "e1bb5682", "metadata": { "collapsed": false, - "editable": true, - "jupyter": { - "outputs_hidden": false - } + "editable": true }, "outputs": [], "source": [ @@ -1706,7 +1699,7 @@ }, { "cell_type": "markdown", - "id": "949e3a5e", + "id": "256590ad", "metadata": { "editable": true }, @@ -1717,102 +1710,12 @@ { "cell_type": "code", "execution_count": 4, - "id": "7e7f4926", + "id": "a3b16f08", "metadata": { "collapsed": false, - "editable": true, - "jupyter": { - "outputs_hidden": false - } + "editable": true }, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Polynomial degree: 0\n", - "Error: 0.2937910450030775\n", - "Bias^2: 0.2929212799917661\n", - "Var: 0.0008697650113114114\n", - "0.2937910450030775 >= 0.2929212799917661 + 0.0008697650113114114 = 0.2937910450030775\n", - "Polynomial degree: 1\n", - "Error: 0.06894146856540673\n", - "Bias^2: 0.06832043024896824\n", - "Var: 0.0006210383164384981\n", - "0.06894146856540673 >= 0.06832043024896824 + 0.0006210383164384981 = 0.06894146856540674\n", - "Polynomial degree: 2\n", - "Error: 0.06106765054837855\n", - "Bias^2: 0.060547654220995305\n", - "Var: 0.0005199963273832366\n", - "0.06106765054837855 >= 0.060547654220995305 + 0.0005199963273832366 = 0.06106765054837854\n", - "Polynomial degree: 3\n", - "Error: 0.03346202229536658\n", - "Bias^2: 0.033140956468054594\n", - "Var: 0.0003210658273119926\n", - "0.03346202229536658 >= 0.033140956468054594 + 0.0003210658273119926 = 0.033462022295366586\n", - "Polynomial degree: 4\n", - "Error: 0.0335277871704832\n", - "Bias^2: 0.03311607538577367\n", - "Var: 0.0004117117847095275\n", - "0.0335277871704832 >= 0.03311607538577367 + 0.0004117117847095275 = 0.0335277871704832\n", - "Polynomial degree: 5\n", - "Error: 0.025517151530854775\n", - "Bias^2: 0.024968890209256446\n", - "Var: 0.0005482613215983264\n", - "0.025517151530854775 >= 0.024968890209256446 + 0.0005482613215983264 = 0.02551715153085477\n", - "Polynomial degree: 6\n", - "Error: 0.019946076068427913\n", - "Bias^2: 0.019502076889868644\n", - "Var: 0.0004439991785592758\n", - "0.019946076068427913 >= 0.019502076889868644 + 0.0004439991785592758 = 0.01994607606842792\n", - "Polynomial degree: 7\n", - "Error: 0.01869592865541787\n", - "Bias^2: 0.017979840090002457\n", - "Var: 0.0007160885654154145\n", - "0.01869592865541787 >= 0.017979840090002457 + 0.0007160885654154145 = 0.018695928655417873\n", - "Polynomial degree: 8\n", - "Error: 0.010736105188369677\n", - "Bias^2: 0.010376602508045077\n", - "Var: 0.00035950268032460084\n", - "0.010736105188369677 >= 0.010376602508045077 + 0.00035950268032460084 = 0.010736105188369678\n", - "Polynomial degree: 9\n", - "Error: 0.011013290652731595\n", - "Bias^2: 0.010539027867198285\n", - "Var: 0.00047426278553330483\n", - "0.011013290652731595 >= 0.010539027867198285 + 0.00047426278553330483 = 0.01101329065273159\n", - "Polynomial degree: 10\n", - "Error: 0.010972468815261458\n", - "Bias^2: 0.010593565969983081\n", - "Var: 0.00037890284527837315\n", - "0.010972468815261458 >= 0.010593565969983081 + 0.00037890284527837315 = 0.010972468815261455\n", - "Polynomial degree: 11\n", - "Error: 0.010840555937745872\n", - "Bias^2: 0.010348475861969925\n", - "Var: 0.0004920800757759405\n", - "0.010840555937745872 >= 0.010348475861969925 + 0.0004920800757759405 = 0.010840555937745865\n", - "Polynomial degree: 12\n", - "Error: 0.010192472149432197\n", - "Bias^2: 0.009610568640079007\n", - "Var: 0.0005819035093531925\n", - "0.010192472149432197 >= 0.009610568640079007 + 0.0005819035093531925 = 0.0101924721494322\n", - "Polynomial degree: 13\n", - "Error: 0.010312285920757117\n", - "Bias^2: 0.009802534263931692\n", - "Var: 0.0005097516568254201\n", - "0.010312285920757117 >= 0.009802534263931692 + 0.0005097516568254201 = 0.010312285920757112\n" - ] - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
    " - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], + "outputs": [], "source": [ "import matplotlib.pyplot as plt\n", "import numpy as np\n", @@ -1824,7 +1727,7 @@ "\n", "np.random.seed(2018)\n", "\n", - "n = 400\n", + "n = 40\n", "n_boostraps = 100\n", "maxdegree = 14\n", "\n", @@ -1864,7 +1767,7 @@ }, { "cell_type": "markdown", - "id": "33c5cae5", + "id": "8c4d3e7f", "metadata": { "editable": true }, @@ -1902,7 +1805,7 @@ }, { "cell_type": "markdown", - "id": "f931f0f2", + "id": "6ba8872d", "metadata": { "editable": true }, @@ -1929,13 +1832,10 @@ { "cell_type": "code", "execution_count": 5, - "id": "58daa28d", + "id": "624a6bc3", "metadata": { "collapsed": false, - "editable": true, - "jupyter": { - "outputs_hidden": false - } + "editable": true }, "outputs": [], "source": [ @@ -1994,7 +1894,7 @@ }, { "cell_type": "markdown", - "id": "3bbcf741", + "id": "7dcfbdc3", "metadata": { "editable": true }, @@ -2019,7 +1919,7 @@ }, { "cell_type": "markdown", - "id": "4b0ffe06", + "id": "583f2b85", "metadata": { "editable": true }, @@ -2047,7 +1947,7 @@ }, { "cell_type": "markdown", - "id": "b11baed6", + "id": "2b422220", "metadata": { "editable": true }, @@ -2060,13 +1960,10 @@ { "cell_type": "code", "execution_count": 6, - "id": "39e76d49", + "id": "ac654a70", "metadata": { "collapsed": false, - "editable": true, - "jupyter": { - "outputs_hidden": false - } + "editable": true }, "outputs": [], "source": [ @@ -2163,7 +2060,7 @@ }, { "cell_type": "markdown", - "id": "e7d12ef0", + "id": "84ccde87", "metadata": { "editable": true }, @@ -2174,13 +2071,10 @@ { "cell_type": "code", "execution_count": 7, - "id": "47f6ae18", + "id": "631a50c9", "metadata": { "collapsed": false, - "editable": true, - "jupyter": { - "outputs_hidden": false - } + "editable": true }, "outputs": [], "source": [ @@ -2266,7 +2160,7 @@ }, { "cell_type": "markdown", - "id": "9c1d4754", + "id": "45c7bf7f", "metadata": { "editable": true }, @@ -2276,7 +2170,7 @@ }, { "cell_type": "markdown", - "id": "b698ac66", + "id": "5d58c073", "metadata": { "editable": true }, @@ -2289,13 +2183,10 @@ { "cell_type": "code", "execution_count": 8, - "id": "0a2409b0", + "id": "6e8fb6ba", "metadata": { "collapsed": false, - "editable": true, - "jupyter": { - "outputs_hidden": false - } + "editable": true }, "outputs": [], "source": [ @@ -2370,7 +2261,7 @@ }, { "cell_type": "markdown", - "id": "56f130b5", + "id": "0c13445c", "metadata": { "editable": true }, @@ -2388,25 +2279,7 @@ ] } ], - "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": "3.9.15" - } - }, + "metadata": {}, "nbformat": 4, "nbformat_minor": 5 } diff --git a/doc/src/week38/week38.do.txt b/doc/src/week38/week38.do.txt index 18367d6b9..f0fa1c3ca 100644 --- a/doc/src/week38/week38.do.txt +++ b/doc/src/week38/week38.do.txt @@ -12,10 +12,16 @@ o Statistical interpretation of OLS and various expectation values o Resampling techniques, Bootstrap and cross validation and bias-variance tradeoff o The material we did not cover last week, that is on more advanced methods for updating the learning rate, are covered by its own video. We will briefly discuss these topics at the beginning of the lecture and during the lab sessions. See video on ADAgrad, RMSprop and ADAM (material from last week not covered during lecture) at URL:"https://youtu.be/J_41Hld6tTU" o "Video of Lecture":"https://youtu.be/4Fo7ITVA7V4" +o "Video from lab sessions on the bias-variance tradeoff":"https://youtu.be/GBWc1abChKo" o "Whiteboard notes":"https://github.com/CompPhysics/MachineLearning/blob/master/doc/HandWrittenNotes/2025/FYSSTKweek38.pdf" !eblock + + + + + !split ===== Readings and Videos ===== !bblock