v0.10.0

• https://github.com/hughperkins/TensorFlow-Examples/blob/enforce-gpu/examples/2_BasicModels/linear_regression.py runs ok now :-) (a bit slow, but not monstrously slow, maybe 3-4 times slower than on CUDA)
• kernels cached between kernel launches
• bunch of behind-the-scenes ops added, like Cast
• memory and device name reported correctly now
• reduce_min working now
• softmax added

v0.9.0

• added reduce_sum, reduce_prod, reduce_max, reduce_mean, in beta, test_reductions.py

v0.8.0

Bug fix:

• OpenCL compile error on AMD variable with automatic storage duration cannot be stored in the named address space should be fixed

v0.7.0

Training works :-)

v0.6.0

• BLAS (using CLBlast ) working now :-), test script test_blas.py

v0.5.0

• Variables can be placed on GPU now test_gradients.py

v0.4.0

• added following per-element unary operators:, which really are running on gpu now :-), test_tf4.py: tanh, abs, acos, asin, atan, ceil, cos, exp, floor, inverse, isfinite, isinf, isnan, log, neg, sign, sin, sqrt, square,tan`

v0.3.0

• (re-)added maximum, minimum, pow per-element operatoins

v0.2.0

• added per-element sub, div, mul

v0.1.0

put Cedrics name in front of 'CLBlast' in'design' section