58
0%

Ran 25 Sep 2018 05:52AM UTC

Jobs 1

Files 0

Run time –

Badge

Embed ▾

pending completion

Build # 58

Build Type

push

travis-ci

Committed by

web-flow

Commit Message

Scalability improvements (#9)

* Adding Relaxed empty region graph implementations
* Build script for testing C++ interface directly
* Consolidating and simplifying the API to the underlying C/CUDA code
* Updating test scripts from python and C++ sides
* Adding utils file, updating __init__, and changing Graph to be an iterable.
* Fixing bug in SKL search index
* Fixing bug in FAISS index. It turns out you need a handle to the res.
* Fixing the "chunking" bug
* Adding function to query the available device memory
* Implementing intelligent query_size estimation based on available memory and size of data
* Splitting chunking and non-chunking solutions into separate functions
* Adding size of edges to the signatures of C++/CUDA calls.
* Mapping indices on the GPU using a dumb method that requires an over-allocation of memory just to ensure the lookups are constant time on the GPU.
* Reducing API surface area to remove some redundancies in code
* Cleaning up code.
* Removing FAISSSearchIndex from the main import, so the code does not depend on faiss being installed.
* Adding a dynamic version number to the setup.py file.
* Adding re import to setup.py
* Fixing bug in mapping to edge list
* Updating import statement for sklearn
* Incrementing version number due to significant improvements

Run Details

Jobs

ID	Job ID	Ran	Files	Coverage
1	58.1	25 Sep 2018 05:52AM UTC	0		Travis Job 58.1

Source Files on build 58

Detailed source file information is not available for this build.

maljovec / nglpy_cuda / 58
0%

README BADGES
x

Markdown

Textile

RDoc

HTML

Rst

Jobs

Source Files on build 58

maljovec / nglpy_cuda / 58 0%

README BADGES x

Markdown

Textile

RDoc

HTML

Rst

Jobs

Source Files on build 58

maljovec / nglpy_cuda / 58
0%

README BADGES
x