Janusz Lisiecki

Hi @farzanehnakhaee70, Could you tell us more about what do you mean by: > Currently dali_backend just works with post and preprocessing for the functions which do not require initialization...

Hi @farzanehnakhaee70, DALI is initialized once when the model is loaded. Then when the inference request is processed the pipeline is just run, However I'm not sure if DALI is...

Hi @danvas, > I have it working with Triton when I'm not using DALI but I get errors when trying to do it with DALI. Can you share the code...

@eric-haibin-lin - thank you for your contribution. Can I ask you to make the PR title and the description more verbose so that someone in the future who would inspect...

Hi @thomas-bouvier, I created a toy example to see if there is something wrong with shuffling or sharding but it seems to work as expected: ``` import nvidia.dali.fn as fn...

Hi @thomas-bouvier, 1) and 2) should be functionally equivalent. When the `file_root` is provided DALI traverse the root directory, assigns a unique label to each subdirectory, and includes all images...

Hi @thomas-bouvier, Please check [the `get_property` operator](https://docs.nvidia.com/deeplearning/dali/user-guide/docs/operations/nvidia.dali.fn.get_property.html) to get the source info of the returned tensor. The metadata is not propagated through the operator so you need to call it...

Hi @thomas-bouvier, > A question by the way, is it correct to use len() on a DALIGenericIterator? Yes, it should work. It returns the number of iterations for given iterator...

!build

!build