Last: 6808(?)
MXNet v0.11 Release Candidate
Major Features
Apple Core ML model converter
Support for Keras v1.2
Gluon: A new high-level interface for mxnet. 0.11 includes a developer preview of gluon
API Changes
Added `CachedOp`. You can now cache the operators that’s called frequently with the same set of arguments to reduce overhead.
Added sample_multinomial for sampling from multinomial distributions.
Added `trunc` operator for rounding towards zero.
Added linalg_gemm, linalg_potrf, ... operators for lapack support.
Added verbose option to Initializer for printing out initialization details.
Added DeformableConvolution to contrib from the Deformable Convolutional Networks paper.
Added float64 support for dot and batch_dot operator.
`allow_extra` is added to Module.set_params to ignore extra parameters.
Added `mod` operator for modulo.
Added `multi_precision` option to SGD optimizer to improve training with float16. Resnet50 now achieves the same accuracy when trained with float16 and gives 50% speedup on Titan XP.
Performance Improvements
ImageRecordIter now stores data in pinned memory to improve GPU memcopy speed.
Bugfixes
Cython interface is fixed. `make cython` and `python setup.py install --with-cython` should install the cython interface and reduce overhead in applications that use imperative/bucketing.
Fixed various bugs in Faster-RCNN example: https://github.com/dmlc/mxnet/pull/6486
Fixed various bugs in SSD example.
Fixed `out` argument not working for `zeros`, `ones`, `full`, etc.
`expand_dims` now supports backward shape inference.
Fixed a bug in rnn.BucketingSentenceIter that causes incorrect layout handling on multi-GPU.
Fixed context mismatch when loading optimizer states.
Fixed a bug in ReLU activation when using MKL.
Fixed a few race conditions that causes crashes on shutdown.
Refactors
Refactored TShape/TBlob to use int64 dimensions and DLTensor as internal storage. Getting ready for migration to DLPack. As a result TBlob::dev_mask_ and TBlob::stride_ are removed.