Madhav Singhal
Madhav Singhal
### Problem We have our input images as a batched tensor `reals` of shape [B, C, H, W]. We feed this into mtcnn after permuting to change shape into [B,...
How can we train on custom Huggingface datasets with this? I could not find the code that takes the HF dataset outputted from pretokenize.py and converts to the sharded dataset....
Why === Currently if you pass in the `device` arg to `from_pretrained`, there is a logic bug which makes it get resolved to something else unnecessarily based on the device...
### Bug Description The example for running follow up tasks as given in `examples/features/follow_up_tasks.py` does not work for local Chrome browser. One gets the error: ``` WARNING [browser] Page load...