ioangatop
diff --git a/‎README.md‎
Lines changed: 50 additions & 21 deletions b/‎README.md‎
Lines changed: 50 additions & 21 deletions
diff --git a/‎lab1/lab1.ipynb‎ renamed to ‎machine_learning_1/lab1/lab1.ipynb‎ b/‎lab1/lab1.ipynb‎ renamed to ‎machine_learning_1/lab1/lab1.ipynb‎
diff --git a/‎lab2/lab2.ipynb‎ renamed to ‎machine_learning_1/lab2/lab2.ipynb‎ b/‎lab2/lab2.ipynb‎ renamed to ‎machine_learning_1/lab2/lab2.ipynb‎
diff --git a/‎lab3/lab3.ipynb‎ renamed to ‎machine_learning_1/lab3/lab3.ipynb‎ b/‎lab3/lab3.ipynb‎ renamed to ‎machine_learning_1/lab3/lab3.ipynb‎
diff --git a/‎machine_learning_2/lab_1/12402559_12141666_lab1.ipynb‎
Lines changed: 2622 additions & 0 deletions b/‎machine_learning_2/lab_1/12402559_12141666_lab1.ipynb‎
Lines changed: 2622 additions & 0 deletions
diff --git a/‎machine_learning_2/lab_1/beet.wav‎
411 KB b/‎machine_learning_2/lab_1/beet.wav‎
411 KB
diff --git a/‎machine_learning_2/lab_1/beet9.wav‎
510 KB b/‎machine_learning_2/lab_1/beet9.wav‎
510 KB
diff --git a/‎machine_learning_2/lab_1/beet92.wav‎
436 KB b/‎machine_learning_2/lab_1/beet92.wav‎
436 KB
diff --git a/‎machine_learning_2/lab_1/environment.yml‎
Lines changed: 9 additions & 0 deletions b/‎machine_learning_2/lab_1/environment.yml‎
Lines changed: 9 additions & 0 deletions
diff --git a/‎machine_learning_2/lab_1/mike.wav‎
374 KB b/‎machine_learning_2/lab_1/mike.wav‎
374 KB
@@ -4,8 +4,11 @@
 
 ### Description
 Welcome to our __Machine Learning__ reposetory! Here you will find various projects from __polynomial regression__ to fully-connected __neural networks__ from scratch, __SVM__ and __Gaussian Processes__!
+On Part II, we analize __Independent Component Analysis__, __Graphical Models__, __EM__ and __VAEs__!
 
-## [__Lab 1: Linear Regression and Overfitting__](lab1/lab1.ipynb)
+# Machine Learning: Part I
+
+## [__Lab 1: Linear Regression and Overfitting__](machine_learning_1/lab1/lab1.ipynb)
 
 ### Part 1: Polynomial Regression
 
@@ -16,14 +19,6 @@ Welcome to our __Machine Learning__ reposetory! Here you will find various proje
     Polynomial regressions as prediction function, along with the data and the original sine function of various polynomial order. 
 </p>
 
-<!-- <p align="center">
-  <img src="readme_imgs/regularized_linear_regression.png" width="410" />
-  <img src="readme_imgs/best_cross-validated_fit.png" width="400" />
-</p> -->
-<!-- <p align="center">
-    <b>Left:</b> Polynomial regression with and without regularization. In regularized polynomial regression, the regularization term acts as a penalty term and has the desired effect of reducing the magnitude of the coefficients.
-    <b>Right:</b> Best cross-validated fit (M =  5, lambda = 1.0)
-</p> -->
 
 ### Part 2: Bayesian Linear (Polynomial) Regression
 <p align="center">
@@ -38,7 +33,7 @@ Welcome to our __Machine Learning__ reposetory! Here you will find various proje
 </p>
 
 
-## [__Lab 2: Classification__](lab2/lab2.ipynb)
+## [__Lab 2: Classification__](machine_learning_1/lab2/lab2.ipynb)
 
 ### Part 1: Multiclass logistic regression
 
@@ -62,8 +57,6 @@ Welcome to our __Machine Learning__ reposetory! Here you will find various proje
 </p>
 
 
-
-
 ### Part 2: Multilayer perceptron
 
 <p align="center">
@@ -75,13 +68,6 @@ Welcome to our __Machine Learning__ reposetory! Here you will find various proje
     Weights of the hidden layer at epoch 0, 4 and 9.
 </p>
 
-<!-- <p align="center">
-  <img src="readme_imgs/activation_functions.png" width="800" />
-</p>
-<p align="center">
-    Different activation functions.
-</p> -->
-
 
 ### Comparison
 
@@ -96,7 +82,7 @@ Welcome to our __Machine Learning__ reposetory! Here you will find various proje
 
 
 
-## [__Lab 3__: Gaussian Processes and Support Vector Machines](lab3/lab3.ipynb)
+## [__Lab 3__: Gaussian Processes and Support Vector Machines](machine_learning_1/lab3/lab3.ipynb)
 
 
 ### Part 1: Gaussian Processes
@@ -119,6 +105,49 @@ Welcome to our __Machine Learning__ reposetory! Here you will find various proje
 </p>
 
 
+
+# Machine Learning: Part II
+
+## [__Lab 1 - Independent Component Analysis__](machine_learning_2/lab_1/12402559_12141666_lab1.ipynb)
+
+In this assignment, we implement the __Independent Component Analysis__ algorithm,
+as described in chapter 34 of David MacKay's book "Information Theory, Inference, and Learning Algorithms".
+
+<p align="center">
+  <img src="readme_imgs/reconstruction.png" width="1000" />
+</p>
+<p align="center">
+    Results of signal reconstruction using different priors and W matrix initialization.
+</p>
+
+
+## [__Lab 2 - Inference in Graphical Models__](machine_learning_2/lab_2/12402559_12141666_lab2.ipynb)
+
+In this assignment, we implement the sum-product and max-sum algorithms for factor graphs over discrete variables.
+We implemented these algorithms to a medical graph, in order to infer the possible decease.
+
+<p align="center">
+  <img src="readme_imgs/bayesian_network.png" width="300" />
+</p>
+<p align="center">
+    Medical Directed Graph.
+</p>
+
+
+## [__Lab 3 - Expectation Maximization and Variational Autoencoder__](machine_learning_2/lab_3/12402559_12141666_lab3.ipynb)
+
+In this assignment, we implement the Expectation Maximization (EM) algorithm and Variational Autoencoder (VAE)
+on the MNIST dataset of written digits.
+
+<p align="center">
+  <img src="readme_imgs/vae_manifold.gif" width="400" />
+</p>
+<p align="center">
+    VAE's leanred manifold of the MNIST dataset of written digits.
+</p>
+
+
+
 #### _Acknowledgement - References_
 
-_The majority of the projects come from the lab assignments of the [Machine Learning 1](http://coursecatalogue.uva.nl/xmlpages/page/2018-2019-en/search-course/course/63074) course of the MSc in Artificial Intelligence at the University of Amsterdam._
+_The majority of the projects come from the lab assignments of the [Machine Learning 1](http://coursecatalogue.uva.nl/xmlpages/page/2018-2019-en/search-course/course/63074) and [Machine Learning 2](https://coursecatalogue.uva.nl/xmlpages/page/2019-2020-en/search-course/course/73105) courses of the MSc in Artificial Intelligence at the University of Amsterdam._
@@ -0,0 +1,9 @@
+name: ml2labs
+channels:
+- defaults
+dependencies:
+- jupyter=1.0.0
+- matplotlib=2.2.2
+- numpy=1.14.2
+- python=3.6.4
+- scipy=1.0.0