-
**Build Scans:**
- [elasticsearch-periodic #4579 / release-tests](https://gradle-enterprise.elastic.co/s/2zym3q2kui3mu)
- [elasticsearch-periodic #4574 / release-tests](https://gradle-enterprise.elast…
-
## Describe the bug
I am trying to use the Optimizer from clearml
When I launch the optimizer based on template task id, it is not picking up the hyperparameters arguments.
For clarification, my…
-
### Feature request
Hi thanks for the library! It would be great if the optimizers can be run on CPU. For example, I would like to try adamw_8bit to full-finetune a 8B model on a 24GB GPU card (RTX40…
-
### 🐛 Describe the bug
If I torch.compile `torch.amp.GradScaler`, it works. But if I copy paste grad_scaler.py and import GradScaler from there, I receive an error.
To reproduce (testcase taken …
-
This is a two-fold question.
Why are two optimizers used? One is used every iteration and the other is used after the 5,000 iterations. Secondly, does the AdamOptimizer need to be defined in the __in…
-
I am trying to run `BaggingRegressor` with a custom built optimizer. But it seems it's not currently supported. So, I am doing the following
```
import torch
import torch.nn as nn
import torch.o…
-
In the substitutions batch of our LogicalPlanOptimizer, there's[ 4 rules](https://github.com/elastic/elasticsearch/blob/b5ec38e75ddafcf0584609c36afa87145a4ab0dc/x-pack/plugin/esql/src/main/java/org/el…
-
Create documentation for the optimizer and rewriter, feel free to add any cases we think we need to cover
```[tasklist]
### Tasks
- [ ] An example to show how to use the rewrite tool using the repl…
-
Torchtune could provide a recipe to do HPO, where the user provides a config, the recipe, eval dataset, params to sweep and budget.
I just played with optimizer. Our default in lr 3e-4. I tried 3e-…
-
Currently every optimizer comes with a `config` specifically for that optimizer that manages the hyperparameters for the optimizer.
This is made because of the following reasons:
* A lot of hyp…