shigeki Ishida
feat: support vllm version filtering
f50141c
raw
history blame
3.91 kB
import json
import os
from datetime import datetime, timezone
from src.display.formatting import styled_error, styled_message, styled_warning
from src.display.utils import Version, VllmVersion
from src.envs import API, EVAL_REQUESTS_PATH, QUEUE_REPO, TOKEN
from src.submission.check_validity import already_submitted_models, check_model_card, is_model_on_hub
REQUESTED_MODELS = None
USERS_TO_SUBMISSION_DATES = None
def add_new_eval(
model: str,
revision: str,
precision: str,
model_type: str,
add_special_tokens: str,
):
global REQUESTED_MODELS
global USERS_TO_SUBMISSION_DATES
if not REQUESTED_MODELS:
REQUESTED_MODELS, USERS_TO_SUBMISSION_DATES = already_submitted_models(EVAL_REQUESTS_PATH)
current_version = Version.v1_4_1.value.name
current_vllm_version = VllmVersion.current.value.name
# γƒγƒΌγ‚Έγƒ§γƒ³ζƒ…ε ±γ‚’ε«γ‚γŸι‡θ€‡γƒγ‚§γƒƒγ‚―
submission_id = f"{model}_{precision}_{add_special_tokens}_{current_version}_{current_vllm_version}"
if submission_id in REQUESTED_MODELS:
return styled_warning(
f"This model has already been evaluated with llm-jp-eval version {current_version} "
f"and vllm version {current_vllm_version}"
)
user_name = ""
model_path = model
if "/" in model:
user_name = model.split("/")[0]
model_path = model.split("/")[1]
precision = precision.split(" ")[0]
current_time = datetime.now(timezone.utc).strftime("%Y-%m-%dT%H:%M:%SZ")
if model_type is None or model_type == "":
return styled_error("Please select a model type.")
# Does the model actually exist?
if revision == "":
revision = "main"
# Is the model on the hub?
model_on_hub, error, _ = is_model_on_hub(model_name=model, revision=revision, token=TOKEN, test_tokenizer=True)
if not model_on_hub:
return styled_error(f'Model "{model}" {error}')
# Is the model info correctly filled?
try:
model_info = API.model_info(repo_id=model, revision=revision)
except Exception:
return styled_error("Could not get your model information. Please fill it up properly.")
# Were the model card and license filled?
try:
_ = model_info.cardData["license"]
except Exception:
return styled_error("Please select a license for your model")
modelcard_OK, error_msg = check_model_card(model)
if not modelcard_OK:
return styled_error(error_msg)
# Seems good, creating the eval
print("Adding new eval")
eval_entry = {
"model_type": model_type,
"model": model,
"precision": precision,
"revision": revision,
"add_special_tokens": add_special_tokens,
"llm_jp_eval_version": current_version,
"vllm_version": current_vllm_version,
"status": "PENDING",
"submitted_time": current_time,
}
# Check for duplicate submission
if f"{model}_{precision}_{add_special_tokens}" in REQUESTED_MODELS:
return styled_warning("This model has been already submitted.")
print("Creating eval file")
OUT_DIR = f"{EVAL_REQUESTS_PATH}/{user_name}"
os.makedirs(OUT_DIR, exist_ok=True)
out_path = (
f"{OUT_DIR}/{model_path}_eval_request_False_{precision}_{add_special_tokens}_{current_vllm_version}.json"
)
with open(out_path, "w") as f:
f.write(json.dumps(eval_entry))
print("Uploading eval file")
API.upload_file(
path_or_fileobj=out_path,
path_in_repo=out_path.split("eval-queue/")[1],
repo_id=QUEUE_REPO,
repo_type="dataset",
commit_message=f"Add {model} to eval queue",
)
# Remove the local file
os.remove(out_path)
return styled_message(
"Your request has been submitted to the evaluation queue!\nPlease wait for up to an hour for the model to show in the PENDING list."
)