phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-07-29 15:23:51 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

add yaml to manifest

Browse source
This commit is contained in:
Hardik Shah 2024-07-22 13:34:08 -07:00
parent 6f0d348b1c
commit 74442e88b1
3 changed files with 7 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

1
MANIFEST.in Normal file
View file

@ -0,0 +1 @@
include llama_toolchain/data/*.yaml

9
llama_toolchain/cli/inference/configure.py
View file

@ -1,5 +1,6 @@
import argparse import argparse
import os import os
import pkg_resources
import textwrap import textwrap
from pathlib import Path from pathlib import Path
@ -37,6 +38,7 @@ class InferenceConfigure(Subcommand):
def read_user_inputs(self): def read_user_inputs(self):
checkpoint_dir = input("Enter the checkpoint directory for the model (e.g., ~/.llama/checkpoints/Meta-Llama-3-8B/): ") checkpoint_dir = input("Enter the checkpoint directory for the model (e.g., ~/.llama/checkpoints/Meta-Llama-3-8B/): ")
model_parallel_size = input("Enter model parallel size (e.g., 1 for 8B / 8 for 70B and 405B): ") model_parallel_size = input("Enter model parallel size (e.g., 1 for 8B / 8 for 70B and 405B): ")
assert model_parallel_size.isdigit() and int(model_parallel_size) in {1, 8}, "model parallel size must be 1 or 8"
return checkpoint_dir, model_parallel_size return checkpoint_dir, model_parallel_size
@ -46,9 +48,10 @@ class InferenceConfigure(Subcommand):
model_parallel_size, model_parallel_size,
yaml_output_path yaml_output_path
): ):
current_dir = os.path.dirname(os.path.abspath(__file__)) default_conf_path = pkg_resources.resource_filename(
default_conf_path = os.path.join(current_dir, "default_configuration.yaml") 'llama_toolchain',
'data/default_inference_config.yaml'
)
with open(default_conf_path, "r") as f: with open(default_conf_path, "r") as f:
yaml_content = f.read() yaml_content = f.read()

0
llama_toolchain/cli/inference/default_configuration.yaml → llama_toolchain/data/default_inference_config.yaml
View file