We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Fixed misprint in filename. Thanks to Ethan Lame.
Minor updates to markdown pages
Deprecates old version of LCA-CIFAR tutorial
Updates to the LCA CIFAR tutorial. Moves the Parameter files section in Part 1 to its own part of the tutorial, currently numbered Part 4. Some other edits and improvements have been made as well.
LCA-CIFAR tutorial mentions useGpu lua variable
Updated Building and Executing PetaVision (markdown)
Updates to python-oriented LCA-CIFAR Tutorial Part3 (learning weights) adds a comparison of the reconstruction from initial weights and reconstruction from the learned weights. There are additional corrections as well.
Updates to LCA CIFAR tutorial - Adds Part1, Part2, etc. to the filenames of the tutorial pages. - Begins a Part3 page for learning weights - Other miscellaneous changes
Updates to LCA-CIFAR tutorial
Fixing markdown error
Start of LCA-CIFAR tutorial page for parallel runs Some additional edits and corrections of other pages are included in this commit as well.
Updates to the one-image the tutorial page
Updates the one-image part of the tutorial
Trying to fix presentation of figures in page
Removes italics on "V1" and "I" layers
Expands the description of LCA
Updates the explanation of sparse coding and LCA
Adds missing images
Updates the .lua file used by LCA-CIFAR-Tutorial The LCA-CIFAR-Tutorial.md page is the old version of the tutorial. It has been updated to use LCA-CIFAR-full.lua, which has nbatch=16 and numImages=50000, as expected by the old version of the tutorial. LCA-CIFAR.lua has nbatch=1 and numImages=1. It is being used by the new version of the tutorial, currently under development. Under the plan for the new version, the person going through the tutorial will start with this simplest problem and then step up to learning weights on the full CIFAR-10 dataset.
Start of a python-based tutorial for OpenPV/LCA This commit adds a page, LCA-CIFAR-Tutorial-One-Image, which provides the first part of a tutorial that uses PetaVision to run the LCA on one image, without learning weights; and then uses python tools to analyze the output.
Fixes sparse values array type from COO to CSR Thanks to Ragib Arnab for identifying this error.
Updates build instructions and tutorial
Updated Python Bindings (WIP) (markdown)
Created Python Bindings (WIP) (markdown)