Tags

Tags give the ability to mark specific points in history as being important

v1.2.0

c9da1885 · Correct set() to scalar function for tensors, added appropriate test to... · Apr 13, 2015

Correct set() to scalar function for tensors, added appropriate test to scalar.cxx, and also fixed a new bug in tensor print()

v1.1.0

8ea1a6e3 · Fixed bugs poined out by Evgeny · Mar 12, 2014

Version 1.1: GPU support added, non-redistributed mappings explicitly considered, performance models improved, miscellanous bugs fixed

v1.0.0

edd97bde · fix minor memory leak for a corner-case of slice · Dec 23, 2013

First release of Cyclops Tensor Framework (CTF). Tensors up to 6D (CCSDT) tested on icc, gcc, and xlc. CTF instantiated for double and complex<double>.

v1.3.1

19750c24 · adjustment to last commit, don't print timings instead of trying to... · Jun 29, 2016

v1.35

b8d08a65 · incremented version to 1.35 · May 17, 2016

Corrected bug in internal memory accounting, fixes crashes after extended runs that leverage certain kernels.

v1.34

93e9b9b2 · Made intermediates formed in contraction expressions be smaller when possible,... · May 11, 2016
```
lower order intermediates for tensor contraction chains
```

v1.33

bae3a59a · modifications and corrections updating gpu capabilities. adjustment to... · Apr 26, 2016

v1.33 minor changes with respect to previous, improvements to sequential performance by avoiding extra copy

v1.32

a262055b · Added call to set_abs_default() that should have been in Set::Set, as pointed out by Devin · Dec 16, 2015

v1.31

80828cdb · Updated version macro in include header file, always forget to do this · Nov 09, 2015

Version 1.31

New features/updates:

1. Renewed support for CUDA-based offloading
2. New macro -DTUNE, activates model tuning, which adjusts performance model parameters. A benchmark aimed to execute a characteristic training set is included in bench/model_trainer and can be used to tune models for any architecture. Output at the end of the benchmark should be pasted into src/shared/init_models.cxx. It is not advisable to always run with -DTUNE on. Current parameters are based on a 16 node Edison runs with 4 processes per node and should be reasonable in most settings. Cubic polynomial model also available, but not used due to inferior observed modelling quality.
3. Added a mapping search that exhaustively tries all possible mappings. As no effective benefit was observed from this expanded search space and the search itself slowed time down for low-cost contractions (e.g. in CCSDT), this mapping search is only done when the time estimated by the old scheme is longer than 1 second.
4. Fixed up configure file and generated config.mk file to be more effective and better documented.

v1.30

75df1bc1 · Merge branch 'sparsity' · Sep 14, 2015
v1.23

5815ec7f · label update v1.22 -> v1.23 · Jul 27, 2015

v1.22

029cb8a6 · Made include/ctf.hpp define CTF_VERSION 1.22 · Jul 09, 2015

Latest stable version, with performance improvements. Slight bug fixes and changes to interface made and as a result tag was moved up from previous commit, performance should not change.

v1.21

835718c5 · 1. Added capability (via constructor) to repack tensor data into a different... · May 20, 2015
v1.20

c9da1885 · Correct set() to scalar function for tensors, added appropriate test to... · Apr 13, 2015

v1.1

8ea1a6e3 · Fixed bugs poined out by Evgeny · Mar 12, 2014

Version 1.1: GPU support added, non-redistributed mappings explicitly considered, performance models improved, miscellanous bugs fixed

v1.0

edd97bde · fix minor memory leak for a corner-case of slice · Dec 23, 2013

First release of Cyclops Tensor Framework (CTF). Tensors up to 6D (CCSDT) tested on icc, gcc, and xlc. CTF instantiated for double and complex<double>.