forked from phoenix-oss/llama-stack-mirror
f02f7b28c1
# What does this PR do? adds an inline HF SFTTrainer provider. Alongside touchtune -- this is a super popular option for running training jobs. The config allows a user to specify some key fields such as a model, chat_template, device, etc the provider comes with one recipe `finetune_single_device` which works both with and without LoRA. any model that is a valid HF identifier can be given and the model will be pulled. this has been tested so far with CPU and MPS device types, but should be compatible with CUDA out of the box The provider processes the given dataset into the proper format, establishes the various steps per epoch, steps per save, steps per eval, sets a sane SFTConfig, and runs n_epochs of training if checkpoint_dir is none, no model is saved. If there is a checkpoint dir, a model is saved every `save_steps` and at the end of training. ## Test Plan re-enabled post_training integration test suite with a singular test that loads the simpleqa dataset: https://huggingface.co/datasets/llamastack/simpleqa and a tiny granite model: https://huggingface.co/ibm-granite/granite-3.3-2b-instruct. The test now uses the llama stack client and the proper post_training API runs one step with a batch_size of 1. This test runs on CPU on the Ubuntu runner so it needs to be a small batch and a single step. [//]: # (## Documentation) --------- Signed-off-by: Charlie Doern <cdoern@redhat.com>
107 lines
2.8 KiB
YAML
107 lines
2.8 KiB
YAML
version: '2'
|
|
image_name: experimental-post-training
|
|
container_image: null
|
|
conda_env: experimental-post-training
|
|
apis:
|
|
- agents
|
|
- datasetio
|
|
- eval
|
|
- inference
|
|
- vector_io
|
|
- safety
|
|
- scoring
|
|
- telemetry
|
|
- post_training
|
|
- tool_runtime
|
|
providers:
|
|
inference:
|
|
- provider_id: meta-reference-inference
|
|
provider_type: inline::meta-reference
|
|
config:
|
|
max_seq_len: 4096
|
|
checkpoint_dir: null
|
|
create_distributed_process_group: False
|
|
- provider_id: ollama
|
|
provider_type: remote::ollama
|
|
config:
|
|
url: ${env.OLLAMA_URL:http://localhost:11434}
|
|
eval:
|
|
- provider_id: meta-reference
|
|
provider_type: inline::meta-reference
|
|
config:
|
|
kvstore:
|
|
type: sqlite
|
|
namespace: null
|
|
db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/meta-reference-gpu}/meta_reference_eval.db
|
|
scoring:
|
|
- provider_id: basic
|
|
provider_type: inline::basic
|
|
config: {}
|
|
- provider_id: braintrust
|
|
provider_type: inline::braintrust
|
|
config:
|
|
openai_api_key: ${env.OPENAI_API_KEY:}
|
|
datasetio:
|
|
- provider_id: localfs
|
|
provider_type: inline::localfs
|
|
config:
|
|
kvstore:
|
|
type: sqlite
|
|
namespace: null
|
|
db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/experimental-post-training}/localfs_datasetio.db
|
|
- provider_id: huggingface
|
|
provider_type: remote::huggingface
|
|
config:
|
|
kvstore:
|
|
type: sqlite
|
|
namespace: null
|
|
db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/huggingface}/huggingface_datasetio.db
|
|
telemetry:
|
|
- provider_id: meta-reference
|
|
provider_type: inline::meta-reference
|
|
config: {}
|
|
post_training:
|
|
- provider_id: huggingface
|
|
provider_type: inline::huggingface
|
|
config:
|
|
checkpoint_format: huggingface
|
|
distributed_backend: null
|
|
device: cpu
|
|
agents:
|
|
- provider_id: meta-reference
|
|
provider_type: inline::meta-reference
|
|
config:
|
|
persistence_store:
|
|
type: sqlite
|
|
namespace: null
|
|
db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/experimental-post-training}/agents_store.db
|
|
safety:
|
|
- provider_id: llama-guard
|
|
provider_type: inline::llama-guard
|
|
config: {}
|
|
vector_io:
|
|
- provider_id: faiss
|
|
provider_type: inline::faiss
|
|
config:
|
|
kvstore:
|
|
type: sqlite
|
|
namespace: null
|
|
db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/experimental-post-training}/faiss_store.db
|
|
tool_runtime:
|
|
- provider_id: brave-search
|
|
provider_type: remote::brave-search
|
|
config:
|
|
api_key: ${env.BRAVE_SEARCH_API_KEY:}
|
|
max_results: 3
|
|
|
|
|
|
metadata_store:
|
|
namespace: null
|
|
type: sqlite
|
|
db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/experimental-post-training}/registry.db
|
|
models: []
|
|
shields: []
|
|
vector_dbs: []
|
|
datasets: []
|
|
scoring_fns: []
|
|
benchmarks: []
|