phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-07-21 20:18:52 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

feat: NVIDIA beginner e2e notebook

Browse source
This commit is contained in:
Jash Gulabrai 2025-04-15 23:26:38 -04:00
parent 7cdd2a0410
commit 6927cdf5ce
31 changed files with 888 additions and 1621 deletions
Download patch file Download diff file Expand all files Collapse all files

14
llama_stack/providers/remote/post_training/nvidia/README.md
View file

@ -125,6 +125,20 @@ client.post_training.job.cancel(job_uuid="your-job-id")
### Inference with the fine-tuned model
#### 1. Register the model
```python
model = Model(
identifier="test-example-model@v1",
provider_id="nvidia",
provider_model_id="test-example-model@v1",
model_type=ModelType.llm,
)
client.register_model(model)
```
#### 2. Inference with the fine-tuned model
```python
response = client.inference.completion(
content="Complete the sentence using one word: Roses are red, violets are ",

20
llama_stack/providers/remote/post_training/nvidia/post_training.py
View file

@ -27,11 +27,12 @@ from .models import _MODEL_ENTRIES
# Map API status to JobStatus enum
STATUS_MAPPING = {
"running": "in_progress",
"completed": "completed",
"failed": "failed",
"cancelled": "cancelled",
"pending": "scheduled",
"running": JobStatus.in_progress.value,
"completed": JobStatus.completed.value,
"failed": JobStatus.failed.value,
"cancelled": JobStatus.cancelled.value,
"pending": JobStatus.scheduled.value,
"unkown": JobStatus.scheduled.value,
}
@ -391,15 +392,14 @@ class NvidiaPostTrainingAdapter(ModelRegistryHelper):
# Handle LoRA-specific configuration
if algorithm_config:
algortihm_config_dict = algorithm_config.model_dump()
if algortihm_config_dict.get("type") == "LoRA":
if algorithm_config.get("type") == "LoRA":
warn_unsupported_params(algorithm_config, supported_params["lora_config"], "LoRA config")
job_config["hyperparameters"]["lora"] = {
k: v
for k, v in {
"adapter_dim": algortihm_config_dict.get("adapter_dim"),
"alpha": algortihm_config_dict.get("alpha"),
"adapter_dropout": algortihm_config_dict.get("adapter_dropout"),
"adapter_dim": algorithm_config.get("adapter_dim"),
"alpha": algorithm_config.get("alpha"),
"adapter_dropout": algorithm_config.get("adapter_dropout"),
}.items()
if v is not None
}