lmnt-com haste issues - Githubissues

lmnt-com / haste

Haste: a fast, simple, and open RNN library

Apache License 2.0

325 stars 27 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Note year and company in license.

#51 shaper closed 1 year ago
0
Nan loss when replace pytorch LSTM with your LSTM or LayerNormLSTM

#50 andkot opened 2 years ago
2
Activations functions

#49 Adel-Moumen closed 2 years ago
0
Supporting RWKV (a RNN that can match transformer LM & zero-shot performance at 1B+ params)

#48 BlinkDL opened 2 years ago
0
Support fp16 in LSTM

#47 zhou13 opened 2 years ago
0
Support for PyTorch packed sequences

#46 joeylitalien closed 2 years ago
2
haste_tf compilation fails with "‘bfloat16’ in namespace ‘Eigen’ does not name a type"

#45 alethes opened 2 years ago
0
Fix `RNN`s backward with `zoneout`

#44 manipopopo closed 2 years ago
0
`RNN`s with `zoneout > 0.0` have wrong gradients

#43 manipopopo closed 2 years ago
1
Feature request for cell classes for pytorch

#42 hadaev8 closed 2 years ago
7
haste_pytorch does not install properly with conda cudatoolkit?

#41 Harimus closed 3 years ago
3
Activation function in IndRNN

#40 Qashqay closed 2 years ago
1
Update setup.pytorch.py

#39 nammingi closed 3 years ago
0
Can Bidirectional Rnn and multi-layer Rnn be supported?

#38 zhang662817 opened 3 years ago
1
layer_norm_gru_cell

#37 PatReis closed 2 years ago
1
return_state_sequence for tf version

#36 thegodone opened 3 years ago
0
Zoneout remains during eval()

#35 DaStapo closed 3 years ago
2
Biases in final IndRNN layer are 0

#34 DaStapo closed 3 years ago
1
Can't run haste layers in Keras

#33 mahenning opened 3 years ago
12
How to expose LayerNormGRUCell to python ?

#32 thegodone closed 3 years ago
2
haste_pytorch: Gradient for kernel/recurrent_kernel becomes zero when trained on gpu

#31 tyterry closed 3 years ago
4
CUDA error: an illegal memory access was encountered

#30 jaak-s closed 3 years ago
6
Support zoneout on lstm cell state and add recurrent dropout

#29 azraelkuan opened 3 years ago
2
Segmentation fault on Cuda 10.0

#28 bratao closed 4 years ago
2
Install on pip on systems without cuda

#27 bratao opened 4 years ago
7
Back propagation in Haste

#26 dgwarlug47 closed 4 years ago
2
Multiple Layers LSTM

#25 dgwarlug47 closed 4 years ago
3
Numerical precision problem encounterd.

#24 w1d2s closed 4 years ago
4
haste_tf\libhaste_tf.so not found

#23 MichaelJanz opened 4 years ago
2
Install on Windows

#22 HumamHelfawi closed 4 years ago
3
Pypi release

#21 bratao closed 4 years ago
2
lib/blas.h:25:50: error: field initializer is not constant

#20 huangmeng123 closed 4 years ago
6
Problem installing on Ubuntu 20.04, tensorflow 2.2

#19 doubleapple123 closed 4 years ago
4
TF: Dropout still applies on inference if training = False

#18 amurashov closed 4 years ago
4
HASTE produces wrong gradients on K80 device

#17 amurashov closed 4 years ago
2
Install fails on AWS DLAMI v. 29.0

#16 amurashov closed 4 years ago
5
Make haste does not work (C++ compile error - field initializer is not constant)

#15 amurashov closed 4 years ago
20
Failing to install / work

#14 cathoderaymission closed 4 years ago
3
Building on google colab gives lots of warnings

#13 HarikrishnanBalagopal closed 4 years ago
3
Bidirectional masking?

#12 PetrochukM closed 4 years ago
4
PIP Install

#11 PetrochukM closed 4 years ago
4
Testing?

#10 PetrochukM closed 4 years ago
3
README PyTorch example

#9 nimz closed 4 years ago
1
Stateful in Pytorch

#8 OverLordGoldDragon closed 4 years ago
8
IndRNNs

#7 OverLordGoldDragon opened 4 years ago
10
[Pytorch] Pass in hidden and cell state in forward pass?

#6 sosuperic closed 4 years ago
2
[Feature] Windows build support

#5 OverLordGoldDragon closed 4 years ago
17
Reproducibility

#4 OverLordGoldDragon closed 4 years ago
2
Are Haste weights compatible with default PyTorch LSTM weights?

#3 shamoons closed 4 years ago
3
Will Haste work without CUDA?

#2 shamoons closed 4 years ago
2