Repository: systemml Updated Branches: refs/heads/gh-pages 14049d257 -> d38bf4ee9
[SYSTEMML-445] Removed batch_norm builtin functions - Removed batch_norm builtin functions to exploit codegen in CP. - Added rewrites for compiling efficient CuDNN operators. - Added rewrites for SGD update operations. - To simplify adding new GPU rewrites, added HopDagPatternMatcher that allows for pattern matching at the HOP-level. This can be extended for other rewrites as well. - Added GPU tests to validate the rewrites. - Updated the DML language documentation. Project: http://git-wip-us.apache.org/repos/asf/systemml/repo Commit: http://git-wip-us.apache.org/repos/asf/systemml/commit/d38bf4ee Tree: http://git-wip-us.apache.org/repos/asf/systemml/tree/d38bf4ee Diff: http://git-wip-us.apache.org/repos/asf/systemml/diff/d38bf4ee Branch: refs/heads/gh-pages Commit: d38bf4ee982946a7d06c855690c26f072d2ab17d Parents: 14049d2 Author: Niketan Pansare <[email protected]> Authored: Thu Aug 30 15:40:44 2018 -0700 Committer: Niketan Pansare <[email protected]> Committed: Thu Aug 30 15:40:44 2018 -0700 ---------------------------------------------------------------------- dml-language-reference.md | 2 -- 1 file changed, 2 deletions(-) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/systemml/blob/d38bf4ee/dml-language-reference.md ---------------------------------------------------------------------- diff --git a/dml-language-reference.md b/dml-language-reference.md index 924336a..cdcc529 100644 --- a/dml-language-reference.md +++ b/dml-language-reference.md @@ -1522,8 +1522,6 @@ Hence, the images are internally represented as a matrix with dimension (N, C * | bias_add | input, bias | [batch_size X num_channels* height_image* width_image] | [num_channels X 1] | [batch_size X num_channels* height_image* width_image] | | Adds the bias (row vector of size num_channels) to input with the given num_channels | | bias_multiply | input, bias | [batch_size X num_channels* height_image* width_image] | [num_channels X 1] | [batch_size X num_channels* height_image* width_image] | | Multiplies the bias (row vector of size num_channels) to input with the given num_channels | | lstm | X, W, bias, out0, c0 | [batch_size X seq_length*num_features] | [num_features+hidden_size X 4*hidden_size] | [batch_size X seq_length*hidden_size] if return_sequences else [batch_size X hidden_size] | return_sequences | Perform computation for single-layer unidirectional LSTM (outputs: out, carryOut) | -| batch_norm2d | input | [batch_size X num_channels* height_image* width_image] | | [batch_size X num_channels* height_image* width_image] | scale, shift, exponentialMovingAverage_Mean, exponentialMovingAverage_Variance, mode, epsilon, momentum | Performs batch normalization operation (outputs: updated exponential moving average mean and variance, cache of the batch mean and variance) | -| batch_norm2d_backward | input, dout | [batch_size X num_channels* height_image* width_image] | [batch_size X num_channels* height_image* width_image] | [batch_size X num_channels* height_image* width_image] | scale, epsilon, cache_mean (from forward), cache_inv_var (from forward) | Computed backpropagation error for batch normalization operation | Note: the builtin functions `batch_norm2d` and `batch_norm2d_backward` are deprecated and will be removed in the next release. The `lstm` builtin function is in experimental phase and is only supported for the GPU backend.
