-
**Note** Reporting using the default pre-filled template, so definitely not a minimal example of code.
- [X] First, please verify that the bug is not already filled:
https://github.com/pylint-dev/…
-
- **~8 seconds** in Glum vs. **~4 seconds** in Scikit-learn
- Same amount of memory used
- **~0.015 RMSE**, using similar hyperparameters
**Results plot**
![image](https://github.com/Quantco/g…
-
In the classic editor, when I go to set up a WordPress image gallery, there are "Gallery Settings" for "Link To", number of "Columns", "Random Order", "Size", and "Type" ("Thumbnail Grid", "Slideshow"…
-
### Reminder
- [X] I have read the README and searched the existing issues.
### Reproduction
When I ran the Galore tuning method as `!bash galore_adamw.sh` in colab, I found an error by accident, t…
-
### Reminder
- [X] I have read the README and searched the existing issues.
### Reproduction
#!/usr/bin/env python3.10
CUDA_VISIBLE_DEVICES=0,1 accelerate launch \
--config_file ../accelera…
-
It looks like smac planner is ignoring costs when obstacles are too small.
Navfn on the other side doesn't show this behavior.
I posted the question already on [ros answers.](https://answers.ros.o…
-
**Describe the bug**
I am following [guide](https://docs.nvidia.com/nemo-framework/user-guide/latest/playbooks/llama2sft.html) to fine tune llama2-7B model on 2 nodes (H100).
my training hangs a…
-
The model being trained is set to evaluation mode at the start of the batched_forward_pass() function inside the PPOTrainer class ([reference](https://github.com/huggingface/trl/blob/main/trl/trainer/…
-
### 请提出你的问题
参考llm 的README 中llama静态图推理，报错信息如下
(paddleslim) :~/PaddleNLP/llm# python export_model.py --model_name_or_path meta-llama/Llama-2-7b-chat --output_path ./inference -
/root/miniconda3/envs/…
-
### Reminder
- [X] I have read the README and searched the existing issues.
### Reproduction
脚本配置参数如下：
CUDA_VISIBLE_DEVICES=0 python train_bash.py \
--stage sft \
--model_name_or_path ..…