Sebastian Raschka

Just a little cosmetic change (hyper-parameter -> hyperparameter) for consistency

Full finetuning

1

This is an attempt to implement **full** finetuning (as opposed to efficient finetuning via adapter etc.) A script for full finetuning (updating all layers). Todos - [x] Create finetune/full.py script...

Deepspeed and bf16-true

1

In the finetuning scripts, we only allow ```python precision: Literal["bf16-true", "32-true"] = "bf16-true", ``` But we also use DeepSpeed when `devices > 1`. However, in this case, you'd get a...

Implement LoRA for efficient finetuning

4

Implements LoRA for efficient finetuning of parrot models - [x] add finetuning script - [x] add the howto guide - [x] add tests - [x] and generate script.

Support LoRA with multiple devices

6

When switching from DeepSpeed stage 2 over to DeepSpeed stage 3, there are currently issues with loading the model via the LoRA finetuning script: ``` ... 18176]) from checkpoint, the...

Thanks for sharing this awesome work (and the paper write-up)! I was wondering if you by chance have a plot similar to the one from the Pythia paper but for...

Warning is raised in PyTorch 2.2 and Lightning 2.2 when using the Tuner

1

### Bug description A [user reported](https://www.youtube.com/watch?v=WXK7JBf0pso) that there is now a user warning when using the Tuner ``` UserWarning: Detected call of `lr_scheduler.step()` before `optimizer.step()`. In PyTorch 1.1.0 and later,...

Use SciPy SVD instead of NumPy SVD.

1

As noted [here](https://x.com/yaroslavvb/status/1746279075210289331?s=20) NumPy's SVD can occasionally be incorrect, and it is better to use scipy.linalg.svd(cov, lapack_driver='gesvd') instead. SVD is used in the PCA class here: https://github.com/rasbt/mlxtend/blob/master/mlxtend/feature_extraction/principal_component_analysis.py

Adds `classes_` attributes to fix #1059

create_counterfactual enhancements (minimization method and fixed features)

1

### Discussed in https://github.com/rasbt/mlxtend/discussions/1028 Originally posted by **farbodr** April 17, 2023 I have a couple of questions about create_counterfactual? 1) On a small table with the shape of (461, 75)...