hiro/ray
1
0
Fork 0
mirror of https://github.com/vale981/ray synced 2025-03-04 17:41:43 -05:00
Code Issues Projects Releases Packages Wiki Activity Actions

[AIR][CI] Speed up HF CI by ~20% (#28208)

Browse source 
Speeds up HuggingFaceTrainer/Predictor tests in CI by around ~20% by switching to a different GPT model. This is the same model Hugging Face team uses for their own CI.

Signed-off-by: Antoni Baum <antoni.baum@protonmail.com>
This commit is contained in:
Antoni Baum 2022-09-01 19:18:10 +02:00 • committed by GitHub
parent ac6d63e397
commit 48898aa03d
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
3 changed files with 6 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
python/ray/train/tests/_huggingface_data.py
View file

File diff suppressed because one or more lines are too long

4
python/ray/train/tests/test_huggingface_predictor.py
View file

@ -23,8 +23,8 @@ prompts = pd.DataFrame(
# We are only testing Casual Language Modeling here
model_checkpoint = "sshleifer/tiny-gpt2"
tokenizer_checkpoint = "sgugger/gpt2-like-tokenizer"
model_checkpoint = "hf-internal-testing/tiny-random-gpt2"
tokenizer_checkpoint = "hf-internal-testing/tiny-random-gpt2"
@pytest.fixture

4
python/ray/train/tests/test_huggingface_trainer.py
View file

@ -35,8 +35,8 @@ prompts = pd.DataFrame(
# We are only testing Casual Language Modelling here
model_checkpoint = "sshleifer/tiny-gpt2"
tokenizer_checkpoint = "sgugger/gpt2-like-tokenizer"
model_checkpoint = "hf-internal-testing/tiny-random-gpt2"
tokenizer_checkpoint = "hf-internal-testing/tiny-random-gpt2"
@pytest.fixture