-
Issue:
I used 70mil parameters AM as a Base model for training with the set of synthetic data and getting output as "call" or "can e can al" when I run the decoder with the trained AM model even thou…
-
Is there a way that we can export the trained model weights? For example, export the weights to numpy /pytorch dictionary format.
-
### Question
I am trying to run wav2vec 2.0 fine-tuning on a custom dataset. However, it requires wav2letter python bindings to be installed. So in order to install wav2letter python bindings, do I n…
-
There is a way for CRAN packages to declare system dependencies and then meet them at configure time (e.g gputools has a SystemDependency on Nvidia's CUDA toolkit which is ultimately bound using a con…
-
Description
===========
I successfully installed the prerequisites:
```
libfftw3-dev is already the newest version (3.3.8-2ubuntu1).
liblapacke-dev is already the newest version (3.9.0-1bui…
-
I create a MFC application with Visual Studio wizards! After setting Preprocessor and link library then build it! Error occurs!
c:\program files\arrayfire\v3\include\af\algorithm.h(86): error C2059: …
-
As v3.7.2 supports CUDA11 I tried to update my docker container accordingly, however, the build fails with: 'nvcc fatal : Unsupported gpu architecture 'compute_30''
Description
===========
I …
pthon updated
4 years ago
-
I have installed HDF5 and ArrayFire but still I'm unable to build my project it is just MNIST example from your site
![Screenshot from 2020-07-22 21-54-58](https://user-images.githubusercontent.com/…
-
-
incidentally I found that
```
AF_OPENCL_DEVICE_TYPE=CPU time cargo run --example xor --release
```
with original xor.rs, also produced strange results - it ran almost instantaneously and gave non…