Paddle

Commit Graph

Author	SHA1	Message	Date
dzhwinter	e983cc90fc	"fix decode bug" (#7711 ) * "fix decode bug" * "follow commnet" * "fix error" * "fix hook bug" * fix based comment * fix copyright * fix based on comment	8 years ago
whs	448fee3db4	Merge pull request #7414 from wanghaoshuang/warpctc Adapt warpctc grad op for gradient checking	8 years ago
dzhwinter	b9b75377a2	Feature/hooks (#7513 ) * add copyright hook * add copyright hook * refine copyright hook * "test copyright hook" * fix check style * fix ci	8 years ago
wanghaoshuang	8f37c3c2a7	Fix sequence scale functor cuda kernel 1. Fix kernel 2. Add more test case	8 years ago
wanghaoshuang	137f0dfc21	1. Fix warpctc grad tensor initial bug. 2. Remove num_seq arguments. 3. Refine CUDA kernel of ScaleLoDTensorFunctor. 4. Change max_relative_error of gradient unitest to 0.007	8 years ago
wanghaoshuang	fd24e195a3	Uncomment check output in unitest	8 years ago
wanghaoshuang	b1af5e435f	1. Fix warpctc grad op 2. Add check grad test	8 years ago
Yiqun Liu	b5fda2723f	Port WarpCTC Operator (#5107 ) * Add Seq2BatchFunctor, which will be used in WarpCTCOp. * Implement WrapCTCFunctor and WrapCTCKernel. * Add unittest of warpctc_op. * Modify the check_output inferface in python unittest framework to allow check a subset of outputs. * Use absolute offset lod in warpctc_op and related functors. * Refine the comments of warpctc_op. * The new python unittest supports checking a subset of the outputs, so revoke the previous change. * Rename the transform from LoDTensor to Tensor with shape [max_sequence_length, num_sequences, sequence_width] to PaddingSequenceFunctor. * Update to the newest codes. * Rename the PaddingSequenceFunctor to PaddingLoDTensorFunctor and remove the computation of dimensions out of the functos.	8 years ago

8 Commits (353d26ce7ee76c651c7856b90fc7cffffee58973)