-
Hi, I am running experiment in Google Colab.
```
gcc version 7.5.0
g++ version 7.5.0
```
**Error:**
```
CMake Error at /usr/local/lib/python2.7/dist-packages/cmake/data/share/cmake-3.12/Module…
-
안녕하세요, 좋은 자료 공유해주셔서 감사드립니다.
```
git clone https://github.com/HawkAaron/warp-transducer
cd warp-transducer
mkdir build; cd build
cmake ..
make
```
위의 RNNT Loss를 설치하는 과정에서, cmake .. 명령어까지 한 …
-
The formular for gradient is below in` warprnnt_numba` and `warp_transducer cpu`:
```
T, U, _ = log_probs.shape
grads = np.full(log_probs.shape, -float("inf"))
log_like = betas[0, 0] …
-
I am try to use this implementation with apex half precision training, but it can't.
showing that it need float rather that half:
______________
File "/data/asr_v3/src/model/transformer_transduc…
-
Hi!
when I finished install warp-transducer ,and pytorch_binding, I run python test.py ,it happen the problem
"ImportError :dynamic module does not define module export function(PyInit_warp_rnnt)"…
-
why are my loss always equal to zero?I checked the dimension and the code, when caculating the rnnloss, it became zero, is it the problem about my packege warp-transducer?
-
Thank you for the warp-transducer, I got the error (show blow) when I make the makefile.
Sorry, but I do not know English in order to translate this correctly.
Предлаг…
-
Same as this: https://github.com/espnet/espnet/issues/4054#issuecomment-1109352507
https://github.com/b-flo/warp-transducer/blob/master/pytorch_binding/src/binding.cpp#L151 The error is raised here…
-
RNNT model requires a third party library [warp-transducer](https://github.com/HawkAaron/warp-transducer), which includes a deprecated pytorch header file, [THC.h](https://github.com/HawkAaron/warp-tr…