p04798526
/
LLaMA-Factory-Mirror
1
0
Fork 1
Code Issues Pull Requests Packages Projects Releases Wiki Activity

fix paligemma data preprocess

This commit is contained in:
hiyouga 2024-05-20 23:51:32 +08:00
parent 542229abb3
commit e55c85ac72
1 changed files with 27 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

36
src/llamafactory/data/preprocess.py
View File

@ -2,7 +2,7 @@ from functools import partial
from itertools import chain from itertools import chain
from typing import TYPE_CHECKING, Any, Callable, Dict, List, Literal, Optional, Sequence, Tuple from typing import TYPE_CHECKING, Any, Callable, Dict, List, Literal, Optional, Sequence, Tuple
from ..extras.constants import IGNORE_INDEX from ..extras.constants import IGNORE_INDEX, IMAGE_TOKEN
from ..extras.logging import get_logger from ..extras.logging import get_logger
from ..extras.packages import is_pillow_available from ..extras.packages import is_pillow_available
from .utils import Role from .utils import Role
@ -80,11 +80,17 @@ def preprocess_supervised_dataset(
logger.warning("Dropped invalid example: {}".format(examples["prompt"][i] + examples["response"][i])) logger.warning("Dropped invalid example: {}".format(examples["prompt"][i] + examples["response"][i]))
continue continue
if processor is not None: if processor is not None and not hasattr(processor, "image_seq_length"): # llava case
examples["prompt"][i][0]["content"] = "<image>" + examples["prompt"][i][0]["content"] examples["prompt"][i][0]["content"] = IMAGE_TOKEN + examples["prompt"][i][0]["content"]
messages = examples["prompt"][i] + examples["response"][i] messages = examples["prompt"][i] + examples["response"][i]
input_ids, labels = [], [] input_ids, labels = [], []
if processor is not None and hasattr(processor, "image_seq_length"): # paligemma case
image_token_id = tokenizer.convert_tokens_to_ids(IMAGE_TOKEN)
input_ids += [image_token_id] * getattr(processor, "image_seq_length")
labels += [image_token_id] * getattr(processor, "image_seq_length")
for turn_idx, (source_ids, target_ids) in enumerate( for turn_idx, (source_ids, target_ids) in enumerate(
template.encode_multiturn( template.encode_multiturn(
tokenizer, tokenizer,
@ -183,8 +189,8 @@ def preprocess_unsupervised_dataset(
logger.warning("Dropped invalid example: {}".format(examples["prompt"][i] + examples["response"][i])) logger.warning("Dropped invalid example: {}".format(examples["prompt"][i] + examples["response"][i]))
continue continue
if processor is not None: if processor is not None and not hasattr(processor, "image_seq_length"): # llava case
examples["prompt"][i][0]["content"] = "<image>" + examples["prompt"][i][0]["content"] examples["prompt"][i][0]["content"] = IMAGE_TOKEN + examples["prompt"][i][0]["content"]
if len(examples["response"][i]) == 1: if len(examples["response"][i]) == 1:
messages = examples["prompt"][i] + examples["response"][i] messages = examples["prompt"][i] + examples["response"][i]
@ -203,6 +209,10 @@ def preprocess_unsupervised_dataset(
if template.efficient_eos: if template.efficient_eos:
labels += [tokenizer.eos_token_id] labels += [tokenizer.eos_token_id]
if processor is not None and hasattr(processor, "image_seq_length"): # paligemma case
image_token_id = tokenizer.convert_tokens_to_ids(IMAGE_TOKEN)
input_ids = [image_token_id] * getattr(processor, "image_seq_length") + input_ids
model_inputs["input_ids"].append(input_ids) model_inputs["input_ids"].append(input_ids)
model_inputs["attention_mask"].append([1] * len(input_ids)) model_inputs["attention_mask"].append([1] * len(input_ids))
model_inputs["labels"].append(labels) model_inputs["labels"].append(labels)
@ -230,8 +240,8 @@ def preprocess_pairwise_dataset(
logger.warning("Dropped invalid example: {}".format(examples["prompt"][i] + examples["response"][i])) logger.warning("Dropped invalid example: {}".format(examples["prompt"][i] + examples["response"][i]))
continue continue
if processor is not None: if processor is not None and not hasattr(processor, "image_seq_length"): # llava case
examples["prompt"][i][0]["content"] = "<image>" + examples["prompt"][i][0]["content"] examples["prompt"][i][0]["content"] = IMAGE_TOKEN + examples["prompt"][i][0]["content"]
chosen_messages = examples["prompt"][i] + [examples["response"][i][0]] chosen_messages = examples["prompt"][i] + [examples["response"][i][0]]
rejected_messages = examples["prompt"][i] + [examples["response"][i][1]] rejected_messages = examples["prompt"][i] + [examples["response"][i][1]]
@ -256,6 +266,10 @@ def preprocess_pairwise_dataset(
chosen_ids += [tokenizer.eos_token_id] chosen_ids += [tokenizer.eos_token_id]
rejected_ids += [tokenizer.eos_token_id] rejected_ids += [tokenizer.eos_token_id]
if processor is not None and hasattr(processor, "image_seq_length"): # paligemma case
image_token_id = tokenizer.convert_tokens_to_ids(IMAGE_TOKEN)
prompt_ids = [image_token_id] * getattr(processor, "image_seq_length") + prompt_ids
model_inputs["prompt_ids"].append(prompt_ids) model_inputs["prompt_ids"].append(prompt_ids)
model_inputs["chosen_ids"].append(chosen_ids) model_inputs["chosen_ids"].append(chosen_ids)
model_inputs["rejected_ids"].append(rejected_ids) model_inputs["rejected_ids"].append(rejected_ids)
@ -292,8 +306,8 @@ def preprocess_kto_dataset(
logger.warning("Dropped invalid example: {}".format(examples["prompt"][i] + examples["response"][i])) logger.warning("Dropped invalid example: {}".format(examples["prompt"][i] + examples["response"][i]))
continue continue
if processor is not None: if processor is not None and not hasattr(processor, "image_seq_length"): # llava case
examples["prompt"][i][0]["content"] = "<image>" + examples["prompt"][i][0]["content"] examples["prompt"][i][0]["content"] = IMAGE_TOKEN + examples["prompt"][i][0]["content"]
if examples["response"][i][0]["content"]: # desired example if examples["response"][i][0]["content"]: # desired example
kto_tag = True kto_tag = True
@ -328,6 +342,10 @@ def preprocess_kto_dataset(
response_ids += [tokenizer.eos_token_id] response_ids += [tokenizer.eos_token_id]
kl_response_ids += [tokenizer.eos_token_id] kl_response_ids += [tokenizer.eos_token_id]
if processor is not None and hasattr(processor, "image_seq_length"): # paligemma case
image_token_id = tokenizer.convert_tokens_to_ids(IMAGE_TOKEN)
prompt_ids = [image_token_id] * getattr(processor, "image_seq_length") + prompt_ids
input_ids = prompt_ids + response_ids input_ids = prompt_ids + response_ids
labels = [IGNORE_INDEX] * len(prompt_ids) + response_ids labels = [IGNORE_INDEX] * len(prompt_ids) + response_ids
kl_input_ids = prompt_ids + kl_response_ids kl_input_ids = prompt_ids + kl_response_ids