Open Hub
Open Hub

 Settings |  Report Duplicate

0
I Use This!
Very Low Activity

Commits : Listings

Analyzed about 7 hours ago. based on code collected 1 day ago.
Nov 05, 2025 — Nov 05, 2026
Commit Message Contributor Files Modified Lines Added Lines Removed Code Location Date
Theory - Low Rank Video EventNet Tensors
Srinivasan Kannan
More... about 8 years ago
Theory - Movie Rankings
Srinivasan Kannan
More... about 8 years ago
Theory - Ranking Videos and Movies by ImageGraph Tensor Products Algorithm
Srinivasan Kannan
More... about 8 years ago
Theory - ImageGraph Tensor Products - Merit of Large Scale Visuals
Srinivasan Kannan
More... about 8 years ago
----------------------------------------------------------------------------------------------------------- Scheduler Analytics - FlameGraph - Convolution Neural Network ----------------------------------------------------------------------------------------------------------- 1. DeepLearning_SchedulerAnalytics.py has been changed to analyze FlameGraph performance profiler JPEG image perf.flamegraph.jpeg which picturises System performance (stack traces etc.,) as flame graph colored proportional to activity 2. FlameGraph SVG and JPEG for this have been created by flamegraph.sh shell script from (out.perf.dd.svg has been saved as perf.flamegraph.jpeg): - https://www.netways.de/fileadmin/images/Events_Trainings/Events/OSDC/2015/Slides_2015/Georg_Schoenberger_Linux_Performance_Profiling_and_Monitoring.pdf (page 57) 3. FlameGraph perl scripts are cloned from https://github.com/brendangregg/FlameGraph 4. ConvNet Deep Learning logs for this JPEG has been captured in testlogs/DeepLearning_SchedulerAnalytics.log.11August2018 5. Flame graph description: http://agentzh.org/misc/slides/yapc-na-2013-flame-graphs.pdf 6. Flamegraph JPEG is presently based on "perf record" and "perf script"
Srinivasan Kannan
More... about 8 years ago
Theory - EventNet Video ImageGraph - Tensor Product
Srinivasan Kannan
More... about 8 years ago
Correction - EventNet ImageGraphs
Srinivasan Kannan
More... about 8 years ago
Theory - EventNet ImageGraphs Tensor Product
Srinivasan Kannan
More... about 8 years ago
Theory - ImageGraph Algorithm for Intrinsic Merit of Large Scale Visuals
Srinivasan Kannan
More... about 8 years ago
Theory - ImageGraph Algorithm for Intrinsic Merit of Large Scale Visuals
Srinivasan Kannan
More... about 8 years ago
Feature - SAT Solver - CVXOPT gels() Feature - Latent Semantic Analysis - Low Rank Approximation
Srinivasan Kannan
More... about 8 years ago
--------------------------------------------------------------------------------------------------------------- SAT Solver - CVXOPT LAPACK gels() least squares --------------------------------------------------------------------------------------------------------------- 1. SAT Solver has been changed to invoke LAPACK gels() least squares from CVXOPT 2. For this NumPy array is cast to CVXOPT real matrix object 3. CVXOPT gels() works only if the matrix is full-rank (number of linearly independent rows is equal to number of rows and similarly for columns) and throws ArithmeticError exception otherwise 4. Approximately 1040 random 3SAT instances for 10 variables - 10 clauses have been executed and MAXSAT approximation ratio converges to ~97%. For larger values of variables, finding full rank matrix needs lot of iterations but still for 20 variables - 20 clauses similar convergence to ~96% was observed. 5. Support for CVXOPT is crucial because of wide range of functionalities (including Convex programs, SDPs) present in CVXOPT 6. logs for this have been committed to testlogs/CNFSATSolver.10variablesAlpha1.0.CVXOPT.log.9August2018
Srinivasan Kannan
More... about 8 years ago
Updated Requirements.txt
Srinivasan Kannan
More... over 8 years ago
Feature - Latent Semantic Analysis
Srinivasan Kannan
More... over 8 years ago
------------------------------------------------------------------------------------------------------------ Latent Semantic Analysis/Indexing (LSA/LSI) Implementation ------------------------------------------------------------------------------------------------------------ 1. Latent Semantic Analysis/Indexing has been implemented as python class in LatentSemanticAnalysis.py 2. This reads list of filenames from LatentSemanticAnalysis.txt and converts into term-document matrix by invoking Pandas DataFrame and ScikitLearn Vector libraries 3. Pandas DataFrame is converted to NumPy array and Singular Value Decomposition (SVD) is done on term-document matrix to arrive at UEV^T decomposition. 4. Documents are represented by U and terms by V. E is matrix of singular values. 5. Cosine similarity of pairs of documents in U is computed for all pairs of documents and printed in log: testlogs/LatentSemanticAnalysis.log.7August2018 6. This class is multipurpose and can be used for variety of applications including Program Analysis. 7. Corpus for this execution is list of python files and similarity between python code is evidenced by cosine similarity
Srinivasan Kannan
More... over 8 years ago
Feature - Empath lexical analyzer and sentiment analysis
Srinivasan Kannan
More... over 8 years ago
----------------------------------------------------------------------------------------------------------- *) Sentiment Analyzer - Empath word embeddings - multipolar sentiment analysis *) Factorization - JSON ------------------------------------------------------------------------------------------------------------ 1.SentimentAnalyzer.py has been changed to invoke Empath lexical analyzer for various signals (sentiments) in Markov Random Fields Belief Propagation Sentiment Analysis function. 2.Empath invocation computes objective maximum score from the dictionary returned by lexical analyzer for each vertex in a clique 3.RecursiveLambdaFunctionGrowth.py invokes Markov Random Fields Belief Propagation Sentiment Analyzer function for the text graph and returns the sentiment score in intrinsic_merit_dict["empath_sentiment"] 4.Logs for this have been committed to testlogs/RecursiveLambdaFunctionGrowth.log.Empath_BeliefPropagation_RGO_MRF_Sentiment.6August2018 ----- 5.JSON for a sample factorization execution has also been committed
Srinivasan Kannan
More... over 8 years ago
Autononmous Vehicle Cloud and Relativity - Reference
Srinivasan Kannan
More... over 8 years ago
Theory - EventNet Logical Clock, Relativity, Wireless Cloud of Autonomous Vehicles
Srinivasan Kannan
More... over 8 years ago
Theory - Lorentz Transform, Cloud of Autonomous Vehicles, Relativity, Lamport Clock, EventNet Clock
Srinivasan Kannan
More... over 8 years ago
Updated Requirements.txt
Srinivasan Kannan
More... over 8 years ago
Theory - Rice Theorem, Complement Diophantines, Program Analysis, Debug Analytics
Srinivasan Kannan
More... over 8 years ago
Theory - Fame Versus Merit - CPI versus GDP in Economies
Srinivasan Kannan
More... over 8 years ago
Theory - Fame Versus Merit - CPI versus GDP in Economies
Srinivasan Kannan
More... over 8 years ago
Theory - Computational Geometric Parallel Wavelet Tree Factorization - NC-PRAM equivalence - References
Srinivasan Kannan
More... over 8 years ago
Theory - k-SAT Sharp Threshold, Friedgut Theorem
Srinivasan Kannan
More... over 8 years ago
Feature and Theory - SAT Solver
Srinivasan Kannan
More... over 8 years ago
------------------------------------------------------------------------------------------------------------- SAT Solver Update - some refactoring, and replace=True ------------------------------------------------------------------------------------------------------------- 1.CNFSATSolver.py has been refactored for invoking Uniform and Non-uniform choice based on Variables==Clauses or Variables != Clauses and lsmr() invocation has been changed to remove regularized LSQR 2.replace=False has been set to replace=True in np.random.choice() for Uniform choice of random 3SAT variables (and everywhere else). For very long time replace was set to False 3.logs for this have been committed to testlogs/CNFSATSolver.2000VariablesAlpha1.0.log.20July2018 and MAXSAT approximation ratio for few iterations ranges from ~95% to ~99% (logs for 98.25% have not been committed)
Srinivasan Kannan
More... over 8 years ago
Correction: thresold -> threshold
Srinivasan Kannan
More... over 8 years ago
Theory - Fame Versus Merit - References
Srinivasan Kannan
More... over 8 years ago