Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | change RNN/LSTM/LSTMP parameter order, which is compatible to old version | Qi Liu | 2016-05-24 |
| | |||
* | Merge branch 'master' into 'master' beta-1.2 | Qi Liu | 2016-05-12 |
|\ | | | | | | | | | | | | | | | | | | | multiple change 1. Merge affine & lstm_gate & projection layer 2. Change clip behavior 3. Seq_buffer support sequence level shuffle 4. LSTM & LSTMP layer support multiple input See merge request !4 | ||
| * | lstm & lstmp layer support multiple input | Qi Liu | 2016-05-12 |
| | | |||
| * | add doc | Qi Liu | 2016-05-11 |
| | | |||
| * | seq buffer support shuffle | Qi Liu | 2016-05-11 |
| | | |||
| * | clip on gradient now | Qi Liu | 2016-05-09 |
| | | |||
| * | fix bug in affine.lua | Qi Liu | 2016-05-09 |
| | | |||
| * | merge affine & lstm_gate & projection layer | Qi Liu | 2016-05-09 |
|/ | |||
* | use `trainer.lua` in acoustic model examplesbeta-1.1 | Determinant | 2016-05-08 |
| | |||
* | prepare for the replacement of `asr_trainer.lua` with `trainer.lua` | Determinant | 2016-05-08 |
| | |||
* | Merge branch 'master' of ssh://speechlab.sjtu.edu.cn:8022/nerv-dev/nervbeta-1 | Determinant | 2016-05-05 |
|\ | |||
| * | Merge branch 'master' into 'master' | mfy43 | 2016-05-05 |
| |\ | | | | | | | | | | | | | | | | | | | fix param_repo:add bug See merge request !3 | ||
| | * | fix param_repo:add bug | Qi Liu | 2016-05-05 |
| |/ | |||
* / | make `bind_params` and `init` consistent in `affine.lua`, | Determinant | 2016-05-05 |
|/ | | | | `lstm_gate.lua`, `projection.lua` | ||
* | add `no_update` attribute to `nerv.MatrixParam`; add corresponding | Determinant | 2016-05-03 |
| | | | | settings to `nerv.AffineLayer` and `nerv.LSTMGateLayer` | ||
* | add "Main Loop" section to overview doc | Determinant | 2016-05-01 |
| | |||
* | add abstraction section in overview | Determinant | 2016-05-01 |
| | |||
* | Merge branch 'master' of ssh://speechlab.sjtu.edu.cn:8022/nerv-dev/nerv | Determinant | 2016-05-01 |
|\ | |||
| * | Merge branch 'master' into 'master' | mfy43 | 2016-04-30 |
| |\ | | | | | | | | | | | | | | | | | | | add relu layer See merge request !2 | ||
| | * | add relu layer | Qi Liu | 2016-04-29 |
| |/ | |||
* / | add overview doc | Determinant | 2016-05-01 |
|/ | |||
* | back-propagate error to the second head of nerv.MSELayer | Determinant | 2016-04-28 |
| | |||
* | add doc for repo | Determinant | 2016-04-28 |
| | |||
* | ... | Determinant | 2016-04-28 |
| | |||
* | add doc | Determinant | 2016-04-28 |
| | |||
* | remove redundant parameters in nerv.ParamRepo:add | Determinant | 2016-04-21 |
| | |||
* | add doc for io | Determinant | 2016-04-20 |
| | |||
* | ... | Determinant | 2016-04-19 |
| | |||
* | ... | Determinant | 2016-04-19 |
| | |||
* | add gitlab ci conf | Determinant | 2016-04-19 |
| | |||
* | add doc for nerv.Network; remove the generated HTML doc | Determinant | 2016-04-19 |
| | |||
* | Merge branch 'master' into 'master' | mfy43 | 2016-04-19 |
|\ | | | | | | | | | | | | | add projection layer and lstmp layer See merge request !1 | ||
| * | add projection layer and lstmp layer | tan tian | 2016-04-18 |
|/ | |||
* | support SeqBuffer decoding; fix bugs in finding paramsalpha-4 | Determinant | 2016-04-18 |
| | |||
* | Merge branch 'master' of github.com:Nerv-SJTU/nervalpha-3.4 | Determinant | 2016-04-04 |
|\ | |||
| * | Merge pull request #37 from liuq901/master | Ted Yin | 2016-03-31 |
| |\ | | | | | | | add general trainer | ||
| | * | add randomize into default setting | Qi Liu | 2016-03-31 |
| | | | |||
| | * | change ptb example | Qi Liu | 2016-03-31 |
| | | | |||
| | * | update general trainer | Qi Liu | 2016-03-31 |
| | | | |||
| | * | update general trainer | Qi Liu | 2016-03-30 |
| | | | |||
* | | | minor changes in asr_trainer.lua | Determinant | 2016-04-04 |
|/ / | |||
* / | fix bug in passing err_input to network; gen zero vectors for biasalpha-3.3 | Determinant | 2016-03-30 |
|/ | |||
* | Merge pull request #35 from liuq901/masteralpha-3.2 | Ted Yin | 2016-03-29 |
|\ | | | | | fix bug of momentum & update mse layer | ||
| * | fix bug of momentum & update mse layer | Qi Liu | 2016-03-29 |
|/ | |||
* | fix "not implemented" and lstm rebinding bugsalpha-3.1 | Determinant | 2016-03-26 |
| | |||
* | add explicit KALDI_BASE setting to READMEalpha-3 | Determinant | 2016-03-17 |
| | |||
* | do not chdir while training | Determinant | 2016-03-17 |
| | |||
* | adjust the parameter order of methods from ParamRepo | Determinant | 2016-03-17 |
| | |||
* | Merge branch 'liuq901-master' | Determinant | 2016-03-17 |
|\ | |||
| * | Merge branch 'master' of github.com:liuq901/nerv into liuq901-master | Determinant | 2016-03-17 |
|/| | | | | | | | | | Conflicts: nerv/layer/rnn.lua |