Issues
- 1
Impossible to compile with CUDA in an Anaconda env
#102 opened by TParcollet - 5
Some questions about RegisterShape
#71 opened by alongwithyou - 0
rnnt_loss compute bptt?
#100 opened by YooSungHyun - 1
"python setup.py install" error: command '/usr/bin/gcc' failed with exit code 1
#99 opened by Schwein3211 - 2
- 1
- 4
- 12
- 2
make fails on A40 with CUDA 11.3
#98 opened by jiay7 - 1
RNNT loss is constantly zero
#92 opened by Rajratnpranesh - 5
nvcc fatal : Unsupported gpu architecture 'compute_30' while execute make again
#97 opened by Stasiche - 1
the grad of cpu ,cuda are not same
#93 opened by lcylcy - 1
segfault with large tensor for pytorch
#95 opened by Peach-He - 0
Can someone explain how the memory is allocated for alpha, beta and log_probs2?
#94 opened by ronggong - 2
test_gpu fails on RTX 3090 with CUDA 11.1
#89 opened by mdabek - 1
building error
#91 opened by jinggaizi - 1
AttributeError: module 'warprnnt_pytorch.warp_rnnt' has no attribute 'gpu_rnnt'
#81 opened by li563042811 - 2
AttributeError: module 'tensorflow.python.framework.ops' has no attribute 'RegisterShape'
#77 opened by mayankpathaklumiq - 0
Getting segmentation fault on CentOS
#88 opened by stefan-falk - 0
- 0
cuda_curand_library not found at build time
#85 opened by gandroz - 1
- 2
- 1
- 1
- 4
error in tf2.1
#59 opened by nichongjia-2007 - 0
question about w
#74 opened by farmboy-dev - 1
building problem
#73 opened by rhoposit - 0
- 6
fatal error: tensorflow/core/framework/bounds_check.h: No such file or directory #include "tensorflow/core/framework/bounds_check.h"
#56 opened by yjiangling - 3
CPU gradient definitions
#69 opened by xkr1 - 0
- 2
import error when use warp_rnnt
#67 opened by shunfeichen - 3
cant find file "tensorflow/core/framework/op.h
#62 opened by ybNo1 - 9
segfault for pytorch
#60 opened by YimengZhu - 12
negative loss
#50 opened by huangnengCSU - 5
Unknown shaped and negative loss
#65 opened by tvatsal1996 - 1
TF 2.0 Support
#52 opened by dtchantao - 4
test.py error
#63 opened by BuaaAlban - 2
How can I build the pkg in windows?
#49 opened by Times125 - 2
- 4
cmake problem
#58 opened by yjiangling - 1
add_network
#57 opened by lijiangguo - 3
multi-gpu
#54 opened by lijiangguo - 1
Can blank assign with non-zero?
#53 opened by yanz0920 - 1
question about gradient to last blank transition
#55 opened by swpd - 2
- 2
tensorflow setup on CPU error
#47 opened by zhoudoufu - 4
Can log_probs be torch.float16?
#42 opened by Marcovaldong - 1