Skip to content

support no apllication of chat template for instructions #3

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

Draft
Ssukriti wants to merge 2 commits into
base: add_datasets_no_templates
Choose a base branch
from
Draft

support no apllication of chat template for instructions #3

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
b8aec18
support no apllication of chat template for instructions
Ssukriti Feb 26, 2024
6973b15
change variable name
Ssukriti Mar 5, 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
5 changes: 3 additions & 2 deletions trl/extras/dataset_formatting.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -58,7 +58,7 @@ def format_dataset(examples):


def get_formatting_func_from_dataset(
dataset: Union[Dataset, ConstantLengthDataset], tokenizer: AutoTokenizer
dataset: Union[Dataset, ConstantLengthDataset], tokenizer: AutoTokenizer, apply_chat_instruction_template=True
) -> Optional[Callable]:
r"""
Finds the correct formatting function based on the dataset structure. Currently supported datasets are:
Expand All @@ -83,6 +83,7 @@ def get_formatting_func_from_dataset(
return conversations_formatting_function(tokenizer, "conversations")
elif dataset.features == FORMAT_MAPPING["instruction"]:
logging.info("Formatting dataset with instruction format")
return instructions_formatting_function(tokenizer)
if apply_chat_instruction_template:
return instructions_formatting_function(tokenizer)

return None
22 changes: 11 additions & 11 deletions trl/trainer/sft_trainer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -145,6 +145,7 @@ def __init__(
dataset_num_proc: Optional[int] = None,
dataset_batch_size: int = 1000,
neftune_noise_alpha: Optional[float] = None,
apply_chat_instruction_template = True,
model_init_kwargs: Optional[Dict] = None,
dataset_kwargs: Optional[Dict] = None,
):
Expand Down Expand Up @@ -250,7 +251,7 @@ def make_inputs_require_grad(module, input, output):
if formatting_func is None and dataset_text_field is None:
# check if dataset has ChatML format or instruction format and is supported
# if not stays #None
formatting_func = get_formatting_func_from_dataset(train_dataset, tokenizer)
formatting_func = get_formatting_func_from_dataset(train_dataset, tokenizer, apply_chat_instruction_template)

requires_input_output_keys = False
Copy link
Collaborator

@alex-jw-brooks alex-jw-brooks Feb 26, 2024

Choose a reason for hiding this comment

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

This variable name should probably be changed - it doesn't make sense if the keys are prompt / completion

Copy link
Author

@Ssukriti Ssukriti Feb 26, 2024

Choose a reason for hiding this comment

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

ya I changed it below and left that one place, thanks for catching it

if not packing:
Expand Down Expand Up @@ -373,7 +374,7 @@ def _prepare_dataset(
num_of_sequences,
chars_per_token,
remove_unused_columns=True,
requires_input_output_keys=False,
requires_prompt_completion_keys=False,
append_concat_token=True,
add_special_tokens=True,
):
Expand All @@ -393,7 +394,7 @@ def _prepare_dataset(
formatting_func,
add_special_tokens,
remove_unused_columns,
requires_input_output_keys,
requires_prompt_completion_keys,
)

else:
Expand All @@ -418,14 +419,13 @@ def _prepare_non_packed_dataloader(
formatting_func=None,
add_special_tokens=True,
remove_unused_columns=True,
requires_input_output_keys=False,
requires_prompt_completion_keys=False,
):
use_formatting_func = formatting_func is not None and dataset_text_field is None
self._dataset_sanity_checked = False

# TODO : fix how EOS tokens are handled
# Inspired from https://github.com/allenai/open-instruct/blob/main/open_instruct/finetune.py#L266
def tokenize_input_output(element):
def tokenize_prompt_completion(element):

# It is difficult to add special tokens here, as separator / EOS tokens that may be added while tokenizing
# input texts may differ from concatenated text, making masking on input length incorrect.
Expand All @@ -437,7 +437,7 @@ def tokenize_input_output(element):
)

new_source = []
for (input_element, output_element) in zip(element['input'], element['output']):
for (input_element, output_element) in zip(element['prompt'], element['completion']):
if not input_element.endswith((' ', '\n', '\t')) and not output_element.startswith((' ', '\n', '\t')):
new_source.append(input_element + ' ' + output_element)
else:
Expand Down Expand Up @@ -491,15 +491,15 @@ def tokenize(element):
f"inspect dataset other columns (in this case {extra_columns}), you can subclass `DataCollatorForLanguageModeling` in case you used the default collator and create your own data collator in order to inspect the unused dataset columns."
)

if requires_input_output_keys:
if "input" in dataset.column_names and "output" in dataset.column_names:
if requires_prompt_completion_keys:
if "prompt" in dataset.column_names and "completion" in dataset.column_names:
# TODO: if we execute this input path, it is expected that we are using a seq2seq
# collator. If that is the case, the tokenizer should had a pad_token; this is set
# to eos automatically if it's unset and no tokenizer is provided, but we should
# properly handle if a tokenizer with no padding token is given.
tokenize_func = tokenize_input_output
tokenize_func = tokenize_prompt_completion
else:
raise KeyError("Missing input / output keys")
raise KeyError("Missing prompt / completion keys")
else:
tokenize_func = tokenize

Expand Down