Janusz Lisiecki

> In the mean time, any thoughts on doing indexing directly in the decoder to save memory space (and presumably speed things up)? Did I miss an option? Let me...

Related to https://github.com/conda/conda-build/issues/1462.

Hi @LinYuOu, Thank you for reaching out. I believe `2024-11-14 11:13:17 [ERROR] Connection to server is broken.` may be emitted by something other than DALI. If you are sure the...

Hi @ziw-liu, Thank you for reaching out. > However, the API seems to require that one file only contains one sample, so each sample will have to be in a...

Hi @ziw-liu, Thank you for providing the details of your use case. > I used to have a I/O bottleneck when reading with python code and the data is on...

> I was also looking at kvikio. I guess if I use GDS via that I would use an [external source with the CuPy interface](https://docs.nvidia.com/deeplearning/dali/user-guide/docs/examples/general/data_loading/external_input.html#Interacting-with-the-GPU-Input) to feed that into a...

Hi @treasan, Thank you for reaching out. Yes, that sounds like a good feature to add. Let us add this to our ToDo list. Could you also tell me how...

Thank you for the clarification. In this case, I think it would be best to return this data directly from the video decoder (at least timesteps for each frame), and...

Hi @MoonBunnyZZZ, Thank you for reaching out. In your case, the most flexible approach would be to use [the `external source` operator](https://docs.nvidia.com/deeplearning/dali/user-guide/docs/examples/general/data_loading/parallel_external_source.html), which allows you to return any number of...

Hi @acecchini, Thank you for reaching out. > I really think this a major flaw in the design of the numpy reader, would be great if you could solve this...