-
### Solution to issue cannot be found in the documentation.
- [X] I checked the documentation.
### Issue
Hi authors,
Thanks for contributing the package. After installed via `conda install -c co…
-
I'm getting weird errors with `BatchNorm`. One example is the code below, where ODE parameters are optimized and the drift is a neural network with some `BatchNorm` layers. The error thrown is `TypeEr…
-
1. Write a custom [`DistributedSampler`](https://pytorch.org/docs/stable/data.html#torch.utils.data.distributed.DistributedSampler) that also takes as input the overall set of indices to pull data fro…
-
This looks neat! I'm just curious about supporting [Equinox](http://github.com/patrick-kidger/equinox) as a possible backend neural network library.
This is typically called as:
```python
model =…
-
### Describe the bug
In Mac OS, starcoder does not even load, probably because it has no Nvidia GPU.
In Windows, the main issue is the dependency on the `bitsandbytes` library. Since the makers …
-
Did you intentionally not include the loss associated with 'initial inference' (prior to unrolling for k steps)? Possible that I am misunderstanding paper.
-
### Bug description
When training with a low precision type (fp16, bf16, etc) logging loss/etc values via `self.log(..., on_epoch=True)` will yield really inaccurate reductions (whether mean, sum, …
-
### Feature request
1. Adds gradient accumulation trick to https://github.com/huggingface/transformers/blob/main/examples/flax/summarization/run_summarization_flax.py
2. Adds [Activation Checkpoin…
-
For various applications and downstream method development, we require at least a basic transfer learning implementation.
The implementation should implement:
- a path to a pre-trained model check…
-
Dear @tianjuxue,
Once again, thank you for this package. It's great! Recently, I was trying to implement an optimisation loop using jaxopt and I encountered an error which I think might be related t…