-
I used orpo colab example for mistral model and I am getting this error. I am using below configs
from trl import ORPOConfig, ORPOTrainer
from unsloth import is_bfloat16_supported
orpo_trainer …
-
gmloose improved the test coverage of `slurm.py` by using snapshots of outputs of the results of the Slurm command line tools to improve tests coverage without requiring that Slurm be installed
https…
-
Hi,author:
I noticed that the 'chembl_preprocess.py' file contains the code to preprocess the CHEMBL dataset and generate the 'chembl_processed_chembl32.csv' file, but I did not see the code to gener…
-
### Description
In https://docs.ray.io/en/latest/cluster/kubernetes/k8s-ecosystem/volcano.html#kuberay-volcano
The flag `--enable-batch-scheduler` has been deprecated. We should use `--batch-sch…
-
### Bug description
I have a lightning module which logs the metrics `val_loss`, and a scheduler that monitors it
```python
def get_plateau_scheduler(self, optimizer):
plateau_scheduler = torc…
-
ubuntu:/data1/fengchuang/LAR-IQA-main$ python scripts/main.py --csv_files /data1/fengchuang/LAR-IQA-main/dataset/training.csv --root_dirs /data1/fengchuang/UHD-IQA-database/UHD --val_csv_file /data…
-
Does the module implement the batch_size scheduler?
-
## 🚀 Model / language coverage
Support the https://huggingface.co/microsoft/Phi-3-mini-128k-instruct model. This is a tracking issue.
Dynamo is splitting this into 13 subgraphs. The good news is…
-
Tokio file IO has [significant overhead](https://docs.rs/tokio/latest/tokio/fs/index.html#tuning-your-file-io), since each syscall like `write` needs to spawn a task and perform a thread context switc…
-
I’m encountering challenges with fine-tuning XTTS for Moroccan Darija despite having a high-quality extensive dataset.
Dataset: 1,000 hours of Moroccan Darija audio, segmented into 1–5 second clips, …