Open Hub
Open Hub

 Settings |  Report Duplicate

0
I Use This!
Very Low Activity

Commits : Listings

Analyzed about 11 hours ago. based on code collected 1 day ago.
Nov 06, 2025 — Nov 06, 2026
Commit Message Contributor Files Modified Lines Added Lines Removed Code Location Date
Updated AsFer Design Document
Srinivasan Kannan
More... about 11 years ago
- Increased number of iterations in BackPropagation to 100000; logs for this with weights printed are added in testlogs/; shows a beautiful convergence and error tapering (~10^-12) - hurricane datasets are also shown as diffs due to perm 644
Srinivasan Kannan
More... about 11 years ago
Miscellaneous code changes: --------------------------- - Weight updates in each iteration are printed in DeepLearning_BackPropagation.py - Changed maitreya_textclient path for Maitreya's Dreams 7.0 text client; Updated to read HURDAT2 NOAA Hurricane Dataset (years 1851-2012) in MaitreyaToEncHoro.py - Maximum sequence length set to 7 for mining HURDAT2 asfer.enchoros.seqmining - New files asfer.enchoros.ascrelative.hurricanes, asfer.enchoros.zodiacal.hurricanes have been added which are created by MaitreyaToEncHoro.py - an example chartsummary for Maitreya's Dreams 7.0 has been updated - 2 logs for SequenceMining for HURDAT2 encoded datasets and Text Class Association Rules learnt by SequenceMining.py Apriori GSP have been added to testlogs/
Srinivasan Kannan
More... about 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... about 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... about 11 years ago
1. Removed some hardcoded input data like weights and expected outputs from the python class and made them as parameters 2. Iterations have been increased to 50000. print_layers() is invoked in each iteration - convergence logs for this has been added to testlogs.
Srinivasan Kannan
More... about 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... about 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... about 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... about 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... about 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... about 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... about 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... about 11 years ago
- changed the hidden-to-output weight updates in the last 4 of the iteration loop with input-to-hidden weight updates - logs for the above have been added to testlogs/ - Updated AsFer Design Document
Srinivasan Kannan
More... about 11 years ago
Deep Learning - Backpropagation partial differential equations implementation: ------------------------------------------------------------------------------ - This is for a minimal 3 layered perceptron backpropagation - The sum of squared errors is computed for each iteration - The weights are recomputed for each iteration and updated - similar to Gradient Descent but with PDE chain rules - For 20000 iterations, error gradually decreases and there is a convergence probably due to floating point - The reference for this implementation is the PDE derivation in: http://mattmazur.com/2015/03/17/a-step-by-step-backpropagation-example/ - logs for this has been added in testlogs/
Srinivasan Kannan
More... about 11 years ago
Updated reference in comments
Srinivasan Kannan
More... about 11 years ago
Updated ProductOwnerProfile
Srinivasan Kannan
More... about 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... about 11 years ago
logs for Deep Learning Convolution Perceptron python implementation
Srinivasan Kannan
More... about 11 years ago
Deep Learning Convolution Perceptron Python implementation
Srinivasan Kannan
More... about 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... about 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... about 11 years ago
Updated AsFer Design Document - corrections to Judge Boolean Function and EXP-completeness
Srinivasan Kannan
More... about 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... about 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... about 11 years ago
Initial code for LinkedIn crawl-scrape. Uses Python-linkedin from https://github.com/ozgur/python-linkedin with some additional parsing for url rewriting and wait for authurl input: - prints a linkedin url which is manually accessed through curl or browser - created redirect_url with code is supplied to raw_input "authurl:" - parses the authcode and logs-in to linkedin to retrieve profile data; get_connections() creates a forbidden error
Srinivasan Kannan
More... about 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... about 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... about 11 years ago
- Updated AsFer Design Document - some changes due to chmod
Srinivasan Kannan
More... about 11 years ago
- Updated AsFer Design Document - some additional commits due to chmod forcing git to show change
Srinivasan Kannan
More... about 11 years ago