2022-10-18 07:19:03 -06:00
|
|
|
import os
|
2023-01-05 04:01:23 -07:00
|
|
|
import sys
|
2022-10-17 06:59:00 -06:00
|
|
|
import typer
|
|
|
|
|
|
|
|
from pathlib import Path
|
2023-01-05 04:01:23 -07:00
|
|
|
from loguru import logger
|
2023-01-31 10:53:56 -07:00
|
|
|
from typing import Optional
|
2022-10-17 06:59:00 -06:00
|
|
|
|
|
|
|
|
|
|
|
app = typer.Typer()
|
|
|
|
|
|
|
|
|
|
|
|
@app.command()
|
2022-10-18 07:19:03 -06:00
|
|
|
def serve(
|
2023-02-03 04:43:37 -07:00
|
|
|
model_id: str,
|
2023-01-31 10:53:56 -07:00
|
|
|
revision: Optional[str] = None,
|
2022-10-18 07:19:03 -06:00
|
|
|
sharded: bool = False,
|
2022-10-27 06:25:29 -06:00
|
|
|
quantize: bool = False,
|
2023-03-30 07:26:27 -06:00
|
|
|
uds_path: Path = "/tmp/text-generation-server",
|
2023-01-05 04:01:23 -07:00
|
|
|
logger_level: str = "INFO",
|
|
|
|
json_output: bool = False,
|
2023-02-13 05:02:45 -07:00
|
|
|
otlp_endpoint: Optional[str] = None,
|
2022-10-17 06:59:00 -06:00
|
|
|
):
|
2022-10-18 07:19:03 -06:00
|
|
|
if sharded:
|
|
|
|
assert (
|
|
|
|
os.getenv("RANK", None) is not None
|
|
|
|
), "RANK must be set when sharded is True"
|
|
|
|
assert (
|
|
|
|
os.getenv("WORLD_SIZE", None) is not None
|
|
|
|
), "WORLD_SIZE must be set when sharded is True"
|
|
|
|
assert (
|
|
|
|
os.getenv("MASTER_ADDR", None) is not None
|
|
|
|
), "MASTER_ADDR must be set when sharded is True"
|
|
|
|
assert (
|
|
|
|
os.getenv("MASTER_PORT", None) is not None
|
|
|
|
), "MASTER_PORT must be set when sharded is True"
|
|
|
|
|
2023-02-13 05:02:45 -07:00
|
|
|
# Remove default handler
|
|
|
|
logger.remove()
|
|
|
|
logger.add(
|
|
|
|
sys.stdout,
|
|
|
|
format="{message}",
|
2023-03-07 10:52:22 -07:00
|
|
|
filter="text_generation_server",
|
2023-02-13 05:02:45 -07:00
|
|
|
level=logger_level,
|
|
|
|
serialize=json_output,
|
|
|
|
backtrace=True,
|
|
|
|
diagnose=False,
|
|
|
|
)
|
2023-04-16 16:26:47 -06:00
|
|
|
|
|
|
|
# Import here after the logger is added to log potential import exceptions
|
|
|
|
from text_generation_server import server
|
|
|
|
from text_generation_server.tracing import setup_tracing
|
|
|
|
|
2023-02-13 05:02:45 -07:00
|
|
|
# Setup OpenTelemetry distributed tracing
|
|
|
|
if otlp_endpoint is not None:
|
|
|
|
setup_tracing(shard=os.getenv("RANK", 0), otlp_endpoint=otlp_endpoint)
|
|
|
|
|
2023-02-03 04:43:37 -07:00
|
|
|
server.serve(model_id, revision, sharded, quantize, uds_path)
|
2022-10-17 06:59:00 -06:00
|
|
|
|
|
|
|
|
|
|
|
@app.command()
|
2022-10-22 12:00:15 -06:00
|
|
|
def download_weights(
|
2023-02-03 04:43:37 -07:00
|
|
|
model_id: str,
|
2023-01-31 10:53:56 -07:00
|
|
|
revision: Optional[str] = None,
|
2022-10-28 11:24:00 -06:00
|
|
|
extension: str = ".safetensors",
|
2023-02-14 05:02:16 -07:00
|
|
|
logger_level: str = "INFO",
|
|
|
|
json_output: bool = False,
|
2022-10-17 06:59:00 -06:00
|
|
|
):
|
2023-02-14 05:02:16 -07:00
|
|
|
# Remove default handler
|
|
|
|
logger.remove()
|
|
|
|
logger.add(
|
|
|
|
sys.stdout,
|
|
|
|
format="{message}",
|
2023-03-07 10:52:22 -07:00
|
|
|
filter="text_generation_server",
|
2023-02-14 05:02:16 -07:00
|
|
|
level=logger_level,
|
|
|
|
serialize=json_output,
|
|
|
|
backtrace=True,
|
|
|
|
diagnose=False,
|
|
|
|
)
|
|
|
|
|
2023-04-16 16:26:47 -06:00
|
|
|
# Import here after the logger is added to log potential import exceptions
|
|
|
|
from text_generation_server import utils
|
|
|
|
|
2023-02-14 05:02:16 -07:00
|
|
|
# Test if files were already download
|
|
|
|
try:
|
|
|
|
utils.weight_files(model_id, revision, extension)
|
|
|
|
logger.info(
|
|
|
|
"Files are already present in the local cache. " "Skipping download."
|
|
|
|
)
|
|
|
|
return
|
|
|
|
# Local files not found
|
|
|
|
except utils.LocalEntryNotFoundError:
|
|
|
|
pass
|
|
|
|
|
|
|
|
# Download weights directly
|
|
|
|
try:
|
|
|
|
filenames = utils.weight_hub_files(model_id, revision, extension)
|
|
|
|
utils.download_weights(filenames, model_id, revision)
|
|
|
|
except utils.EntryNotFoundError as e:
|
|
|
|
if not extension == ".safetensors":
|
|
|
|
raise e
|
|
|
|
|
|
|
|
logger.warning(
|
|
|
|
f"No safetensors weights found for model {model_id} at revision {revision}. "
|
|
|
|
f"Converting PyTorch weights instead."
|
|
|
|
)
|
|
|
|
|
|
|
|
# Try to see if there are pytorch weights
|
|
|
|
pt_filenames = utils.weight_hub_files(model_id, revision, ".bin")
|
|
|
|
# Download pytorch weights
|
|
|
|
local_pt_files = utils.download_weights(pt_filenames, model_id, revision)
|
|
|
|
local_st_files = [
|
|
|
|
p.parent / f"{p.stem.lstrip('pytorch_')}.safetensors"
|
|
|
|
for p in local_pt_files
|
|
|
|
]
|
|
|
|
# Convert pytorch weights to safetensors
|
|
|
|
utils.convert_files(local_pt_files, local_st_files)
|
2022-10-17 06:59:00 -06:00
|
|
|
|
|
|
|
|
|
|
|
if __name__ == "__main__":
|
|
|
|
app()
|