Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly
Sign up
Appearance settings

Commit ae5afd5

Browse files
ml2 added
1 parent c0926b4 commit ae5afd5

File tree

18 files changed

+7290
-21
lines changed

18 files changed

+7290
-21
lines changed

‎README.md‎

Lines changed: 50 additions & 21 deletions
Original file line numberDiff line numberDiff line change
@@ -4,8 +4,11 @@
44

55
### Description
66
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__!
7+
On Part II, we analize __Independent Component Analysis__, __Graphical Models__, __EM__ and __VAEs__!
78

8-
## [__Lab 1: Linear Regression and Overfitting__](lab1/lab1.ipynb)
9+
# Machine Learning: Part I
10+
11+
## [__Lab 1: Linear Regression and Overfitting__](machine_learning_1/lab1/lab1.ipynb)
912

1013
### Part 1: Polynomial Regression
1114

@@ -16,14 +19,6 @@ Welcome to our __Machine Learning__ reposetory! Here you will find various proje
1619
Polynomial regressions as prediction function, along with the data and the original sine function of various polynomial order.
1720
</p>
1821

19-
<!-- <p align="center">
20-
<img src="readme_imgs/regularized_linear_regression.png" width="410" />
21-
<img src="readme_imgs/best_cross-validated_fit.png" width="400" />
22-
</p> -->
23-
<!-- <p align="center">
24-
<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.
25-
<b>Right:</b> Best cross-validated fit (M = 5, lambda = 1.0)
26-
</p> -->
2722

2823
### Part 2: Bayesian Linear (Polynomial) Regression
2924
<p align="center">
@@ -38,7 +33,7 @@ Welcome to our __Machine Learning__ reposetory! Here you will find various proje
3833
</p>
3934

4035

41-
## [__Lab 2: Classification__](lab2/lab2.ipynb)
36+
## [__Lab 2: Classification__](machine_learning_1/lab2/lab2.ipynb)
4237

4338
### Part 1: Multiclass logistic regression
4439

@@ -62,8 +57,6 @@ Welcome to our __Machine Learning__ reposetory! Here you will find various proje
6257
</p>
6358

6459

65-
66-
6760
### Part 2: Multilayer perceptron
6861

6962
<p align="center">
@@ -75,13 +68,6 @@ Welcome to our __Machine Learning__ reposetory! Here you will find various proje
7568
Weights of the hidden layer at epoch 0, 4 and 9.
7669
</p>
7770

78-
<!-- <p align="center">
79-
<img src="readme_imgs/activation_functions.png" width="800" />
80-
</p>
81-
<p align="center">
82-
Different activation functions.
83-
</p> -->
84-
8571

8672
### Comparison
8773

@@ -96,7 +82,7 @@ Welcome to our __Machine Learning__ reposetory! Here you will find various proje
9682

9783

9884

99-
## [__Lab 3__: Gaussian Processes and Support Vector Machines](lab3/lab3.ipynb)
85+
## [__Lab 3__: Gaussian Processes and Support Vector Machines](machine_learning_1/lab3/lab3.ipynb)
10086

10187

10288
### Part 1: Gaussian Processes
@@ -119,6 +105,49 @@ Welcome to our __Machine Learning__ reposetory! Here you will find various proje
119105
</p>
120106

121107

108+
109+
# Machine Learning: Part II
110+
111+
## [__Lab 1 - Independent Component Analysis__](machine_learning_2/lab_1/12402559_12141666_lab1.ipynb)
112+
113+
In this assignment, we implement the __Independent Component Analysis__ algorithm,
114+
as described in chapter 34 of David MacKay's book "Information Theory, Inference, and Learning Algorithms".
115+
116+
<p align="center">
117+
<img src="readme_imgs/reconstruction.png" width="1000" />
118+
</p>
119+
<p align="center">
120+
Results of signal reconstruction using different priors and W matrix initialization.
121+
</p>
122+
123+
124+
## [__Lab 2 - Inference in Graphical Models__](machine_learning_2/lab_2/12402559_12141666_lab2.ipynb)
125+
126+
In this assignment, we implement the sum-product and max-sum algorithms for factor graphs over discrete variables.
127+
We implemented these algorithms to a medical graph, in order to infer the possible decease.
128+
129+
<p align="center">
130+
<img src="readme_imgs/bayesian_network.png" width="300" />
131+
</p>
132+
<p align="center">
133+
Medical Directed Graph.
134+
</p>
135+
136+
137+
## [__Lab 3 - Expectation Maximization and Variational Autoencoder__](machine_learning_2/lab_3/12402559_12141666_lab3.ipynb)
138+
139+
In this assignment, we implement the Expectation Maximization (EM) algorithm and Variational Autoencoder (VAE)
140+
on the MNIST dataset of written digits.
141+
142+
<p align="center">
143+
<img src="readme_imgs/vae_manifold.gif" width="400" />
144+
</p>
145+
<p align="center">
146+
VAE's leanred manifold of the MNIST dataset of written digits.
147+
</p>
148+
149+
150+
122151
#### _Acknowledgement - References_
123152

124-
_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._
153+
_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._
File renamed without changes.
File renamed without changes.
File renamed without changes.

‎machine_learning_2/lab_1/12402559_12141666_lab1.ipynb‎

Lines changed: 2622 additions & 0 deletions
Large diffs are not rendered by default.

‎machine_learning_2/lab_1/beet.wav‎

411 KB
Binary file not shown.

‎machine_learning_2/lab_1/beet9.wav‎

510 KB
Binary file not shown.

‎machine_learning_2/lab_1/beet92.wav‎

436 KB
Binary file not shown.
Lines changed: 9 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,9 @@
1+
name: ml2labs
2+
channels:
3+
- defaults
4+
dependencies:
5+
- jupyter=1.0.0
6+
- matplotlib=2.2.2
7+
- numpy=1.14.2
8+
- python=3.6.4
9+
- scipy=1.0.0

‎machine_learning_2/lab_1/mike.wav‎

374 KB
Binary file not shown.

0 commit comments

Comments
(0)

AltStyle によって変換されたページ (->オリジナル) /