Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Add poem instruction ds #2813

Merged
merged 8 commits into from
Apr 21, 2023
Merged

Add poem instruction ds #2813

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
d91a8b5
update poem instruction path
CloseChoice Apr 20, 2023
bb651e3
update entities
CloseChoice Apr 21, 2023
16649e7
Merge branch 'fix/add-entities' into feature/add-poem-instruction-ds
CloseChoice Apr 21, 2023
ef1ec01
update custom datasets init
CloseChoice Apr 21, 2023
7212d96
move poem instruction dataset
CloseChoice Apr 21, 2023
612d0d3
remove entities
CloseChoice Apr 21, 2023
64fc8c0
update qa datasets
CloseChoice Apr 21, 2023
fd2924f
Merge branch 'main' into feature/add-poem-instruction-ds
andreaskoepf Apr 21, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions model/model_training/custom_datasets/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -162,9 +162,9 @@ def get_one_dataset(
elif dataset_name == "hellaswag":
train, eval = load_hellaswag()
elif dataset_name == "dolly15k":
dataset = DatabricksDolly15k(cache_dir=data_path)
dataset = DatabricksDolly15k(cache_dir=data_path, mode=mode, **kwargs)
elif dataset_name == "alpaca_gpt4":
dataset = AlpacaGpt4(cache_dir=data_path, **kwargs)
dataset = AlpacaGpt4(cache_dir=data_path, mode=mode, **kwargs)
else:
raise ValueError(f"Unknown dataset {dataset_name}")

Expand Down
1 change: 1 addition & 0 deletions model/model_training/custom_datasets/instruction.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,6 +18,7 @@
"zhihu-kol": "wangrui6/zhihu-kol",
"minimath": "kentsui/minimath",
"oa_wiki_qa_bart_10000row": "michaelthwan/oa_wiki_qa_bart_10000row",
"poem_instructions": "checkai/instruction-poems",
}


Expand Down
3 changes: 3 additions & 0 deletions model/model_training/custom_datasets/qa_datasets.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,6 +21,8 @@
# @agoryuno contributed this
re_reference_remove = re.compile(r"\[\d+(?:,\s*\d+)*?\]")
re_single_reference_remove = re.compile(r"\[\s?\d+\s?\]")

# check if the whole string is just a combination of (multiple) whitespaces and newlines
re_whitespace_newline_match = re.compile(r"^[\s\n]*$")


Expand Down Expand Up @@ -450,6 +452,7 @@ def process_split(
dataset: Subset, reverse_augmentation: bool = False, keep_unreversed: bool = True
) -> list[tuple[str, str]]:
data = []

for row in dataset:
question = row["instruction"]
if len(row["input"]) > 0:
Expand Down