Daniel Falbel

See: https://github.com/mlverse/torch/issues/634#issue-958626746

See https://github.com/mlverse/torch/issues/642#issuecomment-894612858

``` r library(torch) model [1] FALSE test_model [1] TRUE ``` I think this is related to #620

`Optional[T]`

https://pytorch.org/docs/stable/generated/torch.nn.DataParallel.html See https://community.rstudio.com/t/how-to-use-torch-nn-dataparallel-in-r/105754

The current error message is not helpful at all: ``` Error in torch_relu(input) : argument "input" is missing, with no default ``` ``` r library(torch) nn_sequential( nnf_relu() ) #> Error...

Investigate issues with nn_multihead_attemntion

2

See #496

missing tensor methods

1

- [x] bool - [x] element_size - [ ] get_device - [ ] imag - [ ] is_shared - [ ] map_ - [ ] new_ones - [ ] new_tensor...

Add some CI infrastructure to test the examples. Shouldn't be at every commit, but perhaps dailies or weekly tests... Examples in vignettes/examples