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

Updating the streaming version in setup.py #1103

Merged
merged 4 commits into from
Apr 9, 2024
Merged

Updating the streaming version in setup.py #1103

Show file tree
Hide file tree
Changes from 3 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
c1d2e7a
updating the streaming version in setup.py
ShashankMosaicML Apr 9, 2024
cc6bdf1
updating constructor call to StreamingDataset
ShashankMosaicML Apr 9, 2024
4238d0d
making allow_unsafe_types and replication configurable through datase…
ShashankMosaicML Apr 9, 2024
1e6a044
adding docstring
ShashankMosaicML Apr 9, 2024
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
2 changes: 2 additions & 0 deletions llmfoundry/data/finetuning/dataloader.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -170,6 +170,8 @@ def build_finetuning_dataloader(cfg: DictConfig,
sampling_granularity=cfg.dataset.get('sampling_granularity', 1),
batching_method=cfg.dataset.get('batching_method', 'random'),
max_seq_len=cfg.dataset.max_seq_len,
allow_unsafe_types=cfg.dataset.get('allow_unsafe_types', False),
replication=cfg.dataset.get('replication', None),
)

else:
Expand Down
4 changes: 4 additions & 0 deletions llmfoundry/data/finetuning/tasks.py
View file
Open in desktop
ShashankMosaicML marked this conversation as resolved.
Show resolved Hide resolved
Original file line number Diff line number Diff line change
Expand Up @@ -516,6 +516,8 @@ def __init__(self,
sampling_granularity: int = 1,
batching_method: str = 'random',
max_seq_len: int = 2048,
allow_unsafe_types: bool = False,
replication: Optional[int] = None,
**kwargs: Any):

if len(kwargs) > 0:
Expand Down Expand Up @@ -552,6 +554,8 @@ def __init__(self,
sampling_method=sampling_method,
sampling_granularity=sampling_granularity,
batching_method=batching_method,
allow_unsafe_types=allow_unsafe_types,
replication=replication,
)

self.tokenizer = tokenizer
Expand Down
4 changes: 4 additions & 0 deletions llmfoundry/data/text_data.py
View file
Open in desktop
ShashankMosaicML marked this conversation as resolved.
Show resolved Hide resolved
Original file line number Diff line number Diff line change
Expand Up @@ -109,6 +109,8 @@ def __init__(self,
sampling_method: str = 'balanced',
sampling_granularity: int = 1,
batching_method: str = 'random',
allow_unsafe_types: bool = False,
replication: Optional[int] = None,
**kwargs: Any):

if len(kwargs) > 0:
Expand Down Expand Up @@ -151,6 +153,8 @@ def __init__(self,
sampling_method=sampling_method,
sampling_granularity=sampling_granularity,
batching_method=batching_method,
allow_unsafe_types=allow_unsafe_types,
replication=replication,
)
self.tokenizer = tokenizer
self.max_seq_len = max_seq_len
Expand Down
2 changes: 1 addition & 1 deletion setup.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -55,7 +55,7 @@
'mlflow>=2.10,<3',
'accelerate>=0.25,<0.26', # for HF inference `device_map`
'transformers>=4.39.3,<4.40',
'mosaicml-streaming>=0.7.4,<0.8',
'mosaicml-streaming>=0.7.5,<0.8',
'torch>=2.2.1,<2.3',
'datasets>=2.16,<2.17',
'fsspec==2023.6.0', # newer version results in a bug in datasets that duplicates data
Expand Down
Loading