Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Feb 26, 2016

• Regularized gradients

  gwgundersen

  committed
  6033aa1

  Copy full SHA for 6033aa1

• Regularized cost function

  gwgundersen

  committed
  7d90164

  Copy full SHA for 7d90164

• Gradients for Theta and X

  gwgundersen

  committed
  f4b4cee

  Copy full SHA for f4b4cee

Commits on Feb 25, 2016

• Cost function for collaborative filtering

  gwgundersen

  committed
  e5a72ae

  Copy full SHA for e5a72ae

• Comments describing algorithm

  gwgundersen

  committed
  04d4050

  Copy full SHA for 04d4050

• Selected best epsilon based on F1 score based on precision and recall of predictions

  gwgundersen

  committed
  db26074

  Copy full SHA for db26074

• Calculated Gaussian parameters

  gwgundersen

  committed
  b74e832

  Copy full SHA for b74e832

• Initial commit

  gwgundersen

  committed
  c3ca53c

  Copy full SHA for c3ca53c

Commits on Feb 21, 2016

• Data recovery

  gwgundersen

  committed
  8b19fb6

  Copy full SHA for 8b19fb6

• Projecting data onto eigenvectors

  gwgundersen

  committed
  2f7d233

  Copy full SHA for 2f7d233

• PCA

  gwgundersen

  committed
  4d807a8

  Copy full SHA for 4d807a8

• Random initialization of centroid locations

  gwgundersen

  committed
  862f151

  Copy full SHA for 862f151

• Centroid update step

  gwgundersen

  committed
  5dda9e9

  Copy full SHA for 5dda9e9

• Finding closest centroid

  gwgundersen

  committed
  ba6c7d7

  Copy full SHA for ba6c7d7

Commits on Feb 15, 2016

• Initial commit

  gwgundersen

  committed
  d8b94fd

  Copy full SHA for d8b94fd

• Features vectur creation

  gwgundersen

  committed
  c21ae76

  Copy full SHA for c21ae76

• Completed process email

  gwgundersen

  committed
  3b17919

  Copy full SHA for 3b17919

Commits on Feb 14, 2016

• Parameters optimization by testing C and sigma for multiple values

  gwgundersen

  committed
  4d13250

  Copy full SHA for 4d13250

Commits on Feb 1, 2016

• Initial commit.

  gwgundersen

  committed
  d3886a1

  Copy full SHA for d3886a1

Commits on Jan 30, 2016

• Use cross validation set to programmatically decide on best lambda

  gwgundersen

  committed
  706d517

  Copy full SHA for 706d517

• Added comment explaining lambda

  gwgundersen

  committed
  909a742

  Copy full SHA for 909a742

• Cleaned up comments

  gwgundersen

  committed
  f369620

  Copy full SHA for f369620

• Generate new polynomial features for more complex hypotheses

  gwgundersen

  committed
  5d69b2d

  Copy full SHA for 5d69b2d

• Implemented learning curves

  gwgundersen

  committed
  7a403db

  Copy full SHA for 7a403db

• Regularized linear regression

  gwgundersen

  committed
  2f79325

  Copy full SHA for 2f79325

Commits on Jan 23, 2016

• Week 6

  gwgundersen

  committed
  ebf03d8

  Copy full SHA for ebf03d8

• Regularized gradient

  gwgundersen

  committed
  13b820e

  Copy full SHA for 13b820e

• Backprop

  gwgundersen

  committed
  68ccfb5

  Copy full SHA for 68ccfb5

• Random initialization of weights

  gwgundersen

  committed
  a3e1324

  Copy full SHA for a3e1324

• Gradient of sigmoid function

  gwgundersen

  committed
  30b986b

  Copy full SHA for 30b986b

• Regularized term

  gwgundersen

  committed
  c916ae6

  Copy full SHA for c916ae6

• Cost function for forward propagation

  gwgundersen

  committed
  4430106

  Copy full SHA for 4430106

Commits on Jan 22, 2016

• Assignment 4

  gwgundersen

  committed
  3f7abdf

  Copy full SHA for 3f7abdf

Commits on Jan 17, 2016

• Predictions using a forward propagation with neural network

  gwgundersen

  committed
  c417f7e

  Copy full SHA for c417f7e

• Implement predictions based on thetas for each classifier in a multi-classification problem

  gwgundersen

  committed
  de8fcb6

  Copy full SHA for de8fcb6

Pagination

Previous

Next