Open Hub
Open Hub

 Settings |  Report Duplicate

0
I Use This!
Very Low Activity

Commits : Listings

Analyzed about 23 hours ago. based on code collected 2 days ago.
Nov 05, 2025 — Nov 05, 2026
Commit Message Contributor Files Modified Lines Added Lines Removed Code Location Date
- Interview Algorithm crawl-visual script changed for treewidth invocation - Spidered text changed - Rule Search script corrected to use datetime objects - Rule Search script corrected to use timedelta for incrementing date and time - logs and screenshots for above
Srinivasan Kannan
More... almost 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... almost 11 years ago
- New Mined Class Association Rule Search script has been added. This script searches the astronomical data with parsed Maitreya Text client data with date, time and longitude-latitude queries. - Where this is useful is after mining the astronomical data with SequenceMining, there is a necessity to search when a particular mined rule occurs. This is automated prediction. - Logs for this has been added in testlogs/ and chartsummary.rulesearch
Srinivasan Kannan
More... almost 11 years ago
- Interview Algorithm crawl-visual script has been changed to compute tree width for only subgraphs of size <= 4 because for larger sizes python MemoryError occurs in 32-bit python. Might require 64-bit python and architecture. - TreeWidth script has been changed to replace the zip() which was wrong with two nested for-loops for pairwise intersection of all subgraphs - Web Spidered text has been changed - AsFer Design Document has been updated with connections between Neural networks, Circuit Complexity and WordNet (kind of Valiant-like Circuit of Mind) - screenshot for above has been added
Srinivasan Kannan
More... almost 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... almost 11 years ago
- TreeWidth implementation has been corrected to take as input set of edges of the RGO graph - TreeWidth for set of subgraphs less than an input parameter size is computed as TreeWidth computation is exponential in graph size and there are MemoryErrors in python for huge set of all subgraphs. For example even a small graph with 10 nodes gives rise to 1024 possible subgraphs - Spidered text has been updated to create a small RGO graph - Logs and screenshots have been added - Each subgraph is hashed to create a unique string for each subgraph - TreeWidth is printed by finding maximum set in junction tree
Srinivasan Kannan
More... almost 11 years ago
- Updated AsFer Design Document - Interview Algorithm Crawl-Visual script has been updated with a DOT graph file writing which creates a .dot file for the RGO graph constructed - Also variety of connectivity and graph complexity measures have been added - Importantly a new Tree Width computing script has been added. NetworkX does not have API for tree width, hence a naive script that iterates through all subgraphs and finds intersecting subgraphs to connect them and form a junction tree, has been written. This is a costly measure compared to intrinsic merit which depends only on graph edges,vertices and depth of recursive gloss overlap. Tree decomposition of graph is NP-hard.
Srinivasan Kannan
More... almost 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... almost 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... almost 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... almost 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... almost 11 years ago
- Updated AsFer Design Document - New Interview Algorithm script with NetworkX+Matplotlib rendering and takes as input a randomly crawled HTML webpage(uses beautiful soup to rip off script and style tags and write only text in the HTML page) has been added - Above script also computes the graph theoretic connectivity of the RGO wordnet subgraph based on Menger's theorem - prints number of nodes required to disconnect the graph or equivalently number of node indpendent paths - logs and screenshots for the above have been added - WebSpider.py has been updated to crawl based on a crawling target parameter - Either a streaming website(twitter, streamed news etc.,) or a HTML webpage
Srinivasan Kannan
More... almost 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... almost 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... almost 11 years ago
- Abstract_DBBackend changes have been reflected in Scrapy Web Spider - backend added as argument in execute_query() - Abstract_DBBackend.py has a subtle problem: - Multiple @inject(s) are not supported in Python Injector - only the innermost @inject works and outer @inject throws a __init__ argument errors in webspider/ - Conditional @inject depending on backend is required but at present switching the order of @inject(s) circumvents this - most recent scrapy crawl logs for this have been added to webspider/testlogs
Srinivasan Kannan
More... almost 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... almost 11 years ago
Dependency Injection code commits for MongoDB backend - With this MongoDB is also a storage backend for AsFer algorithms similar to MySQL: - Abstract_DBBackend.py has been updated for both MySQL and MongoDB injections - MongoDB configuration and backend connect/query code has been added. Backend is either populated by Robomongo or pymongo reading from the Streaming Abstract Generator iterable framework. - With this AsFer supports both SQL(MySQL) and NoSQL(file,hive,hbase,cassandra backends in Streaming Abstract Generator). - log with a simple NoSQL table with StreamingData.txt and USBWWAN data has been added to testlogs/. - MongoDB configuration has a database(asfer-database) and a collection(asfer-collection). - MongoDB_DBBackend @provides pymongo.collection.Collection which is @inject-ed to Abstract_DBBackend
Srinivasan Kannan
More... almost 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... almost 11 years ago
- Updated AsFer Design Document - New python implementation for CountMeanMinSketch Streaming Data Frequency Estimation has been added which is an improved version of CountMinSketch that computes median of average of estimator rows in sketch - Corrections to CountMinSketch hashing algorithms and Sketch width-depth as per the error bounds has been made - Spark MapReducer prints the number of elements in the stream data set - Logs for the above have been added to testlogs
Srinivasan Kannan
More... almost 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... almost 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... almost 11 years ago
- Updated the Streaming LogLogCounter and HyperLogLogCounter scripts to accept StreamData.txt dataset from Abstract Generator and added a Spark MapReducer script for Streaming Data sets for comparison of exact data with Streaming_<algorithm>.py estimations. - Added logs for the Counters and Spark MapReducer script on the StreamData.txt
Srinivasan Kannan
More... almost 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... almost 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... almost 11 years ago
Cardinality estimation with LogLog and HyperLogLog counters for USB stream datasets - HyperLogLog estimation: ~110 elements - LogLog estimation: ~140 elements
Srinivasan Kannan
More... almost 11 years ago
- USB stream file storage name updated in Streaming_AbstractGenerator - Corrections to CountMinSketch - hashing updated to include rows (now the element frquencies are estimated almost exactly) - logs for above update to CountMinSketch
Srinivasan Kannan
More... almost 11 years ago
Updated AsFer Design Document - 53.11 and 53.12
Srinivasan Kannan
More... almost 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... almost 11 years ago
Updated AsFer Design Document
Srinivasan Kannan
More... almost 11 years ago
All the Streaming_<>.py Streaming Algorithm implementations have been updated with: - hashlib ripemd160 hash MD algorithm for hash functions and return hexdigest() - USBWWAN byte stream data from USBmd print_buffer() logs has been added as a Data Storage and Data Source - logs for the above have been added to testlogs/ - Streaming Abstract Generator has been updated with USB stream data iterable and parametrized for data source and storage - Some corrections to the scripts
Srinivasan Kannan
More... almost 11 years ago