Conversation
Signed-off-by: Alexandros Koumparoulis <akoumparouli@nvidia.com>
Signed-off-by: Alexandros Koumparoulis <akoumparouli@nvidia.com>
|
/ok to test 0346ed6 |
akoumpa
requested review from
HuiyingLi,
adil-a,
hemildesai and
jgerh
as code owners
@akoumpa, there was an error processing your request: See the following link for more information: https://docs.gha-runners.nvidia.com/cpr/e/2/ |
| INFO:root:step 30 | epoch 0 | val_loss 1.1230 | val_acc1 0.7820 | val_mrr 0.8450 | ||
| ``` | ||
|
|
||
| ## Using the Fine-Tuned Model |
There was a problem hiding this comment.
better to change this to transformers usage script. as in here:
https://huggingface.co/nvidia/llama-nemotron-embed-1b-v2#transformers-usage
| dataset: | ||
| _target_: nemo_automodel.components.datasets.llm.make_retrieval_dataset | ||
| data_dir_list: | ||
| - /path/to/train_data.json |
There was a problem hiding this comment.
here better to use the predefined path that we added lately to automatically download the example dataset from HF hub.
see the latest yaml file pls: https://github.com/NVIDIA-NeMo/Automodel/blob/main/examples/biencoder/llama3_2_1b_biencoder.yaml#L55
|
|
||
| ```yaml | ||
| model: | ||
| _target_: nemo_automodel.NeMoAutoModelBiencoder.from_pretrained |
There was a problem hiding this comment.
let's not merge this MR yet, we can do once MR 1449 is finalized, we might need to update class names etc.
Pull request was converted to draft
Co-authored-by: jgerh <163925524+jgerh@users.noreply.github.com>
3 participants
What does this PR do ?
Add a one line overview of what this PR aims to accomplish.
Changelog
Before your PR is "Ready for review"
Pre checks:
If you haven't finished some of the above items you can still open "Draft" PR.
Additional Information