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

additional logging to get maximum token length of a sequence in the dataset #1066

Merged
merged 2 commits into from
Jan 10, 2024
Merged

additional logging to get maximum token length of a sequence in the dataset #1066

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
b88534b
additional logging to get maximum token length of a sequence in the d…
winglian Jan 8, 2024
86882cd
fix ordering to properly determine the max_len of tokens before dropp…
winglian Jan 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
16 changes: 10 additions & 6 deletions src/axolotl/utils/trainer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -109,12 +109,6 @@ def disable_datasets_caching():
def process_datasets_for_packing(cfg, train_dataset, eval_dataset, tokenizer):
drop_long = partial(drop_long_seq, sequence_len=cfg.sequence_len)
with zero_first(is_main_process()):
train_dataset = train_dataset.filter(drop_long, num_proc=cfg.dataset_processes)
if eval_dataset:
eval_dataset = eval_dataset.filter(
drop_long, num_proc=cfg.dataset_processes
)

if cfg.group_by_length:
train_dataset = train_dataset.map(
add_length, num_proc=cfg.dataset_processes
Expand All @@ -130,6 +124,16 @@ def process_datasets_for_packing(cfg, train_dataset, eval_dataset, tokenizer):
add_position_ids, num_proc=cfg.dataset_processes
)

if cfg.group_by_length or cfg.sample_packing:
NanoCode012 marked this conversation as resolved.
Show resolved Hide resolved
max_input_len = np.max(get_dataset_lengths(train_dataset))
LOG.debug(f"max_input_len: {max_input_len}", main_process_only=True)

train_dataset = train_dataset.filter(drop_long, num_proc=cfg.dataset_processes)
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is it possible to count how many samples are dropped?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

yeah, let's slate that for another feature/PR. This was just shuffling around the order.

if eval_dataset:
eval_dataset = eval_dataset.filter(
drop_long, num_proc=cfg.dataset_processes
)

# Phi doesn't want the attention_mask feature when training
if (
"CodeGenTokenizer" in tokenizer.__class__.__name__
Expand Down