423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2025-12-14 10:56:56 +08:00
Code Issues Packages Projects Releases Wiki Activity

support llama pro #2338 , add rslora

Browse Source 
Former-commit-id: 7924ffc55d
This commit is contained in:
hiyouga
2024-02-15 02:27:36 +08:00
parent 7b6af4e379
commit 96265ec154
24 changed files with 438 additions and 203 deletions
Download Patch File Download Diff File Expand all files Collapse all files

32
src/llmtuner/api/app.py
View File

@@ -74,6 +74,13 @@ def create_app(chat_model: "ChatModel") -> "FastAPI":
)
semaphore = asyncio.Semaphore(int(os.environ.get("MAX_CONCURRENT", 1)))
role_mapping = {
Role.USER: DataRole.USER,
Role.ASSISTANT: DataRole.ASSISTANT,
Role.SYSTEM: DataRole.SYSTEM,
Role.FUNCTION: DataRole.FUNCTION,
Role.TOOL: DataRole.OBSERVATION,
}
@app.get("/v1/models", response_model=ModelList)
async def list_models():
@@ -85,28 +92,27 @@ def create_app(chat_model: "ChatModel") -> "FastAPI":
if not chat_model.can_generate:
raise HTTPException(status_code=status.HTTP_405_METHOD_NOT_ALLOWED, detail="Not allowed")
if len(request.messages) == 0 or request.messages[-1].role not in [Role.USER, Role.TOOL]:
if len(request.messages) == 0:
raise HTTPException(status_code=status.HTTP_400_BAD_REQUEST, detail="Invalid length")
messages = [dictify(message) for message in request.messages]
if len(messages) and messages[0]["role"] == Role.SYSTEM:
system = messages.pop(0)["content"]
if role_mapping[request.messages[0].role] == DataRole.SYSTEM:
system = request.messages.pop(0).content
else:
system = None
system = ""
if len(messages) % 2 == 0:
if len(request.messages) % 2 == 0:
raise HTTPException(status_code=status.HTTP_400_BAD_REQUEST, detail="Only supports u/a/u/a/u...")
for i in range(len(messages)):
if i % 2 == 0 and messages[i]["role"] not in [Role.USER, Role.TOOL]:
input_messages = []
for i, message in enumerate(request.messages):
input_messages.append({"role": role_mapping[message.role], "content": message.content})
if i % 2 == 0 and input_messages[i]["role"] not in [DataRole.USER, DataRole.OBSERVATION]:
raise HTTPException(status_code=status.HTTP_400_BAD_REQUEST, detail="Invalid role")
elif i % 2 == 1 and messages[i]["role"] not in [Role.ASSISTANT, Role.FUNCTION]:
elif i % 2 == 1 and input_messages[i]["role"] not in [DataRole.ASSISTANT, DataRole.FUNCTION]:
raise HTTPException(status_code=status.HTTP_400_BAD_REQUEST, detail="Invalid role")
elif messages[i]["role"] == Role.TOOL:
messages[i]["role"] = DataRole.OBSERVATION
tool_list = request.tools
if len(tool_list):
if isinstance(tool_list, list) and len(tool_list):
try:
tools = json.dumps([tool["function"] for tool in tool_list], ensure_ascii=False)
except Exception:
@@ -116,7 +122,7 @@ def create_app(chat_model: "ChatModel") -> "FastAPI":
async with semaphore:
loop = asyncio.get_running_loop()
return await loop.run_in_executor(None, chat_completion, messages, system, tools, request)
return await loop.run_in_executor(None, chat_completion, input_messages, system, tools, request)
def chat_completion(messages: Sequence[Dict[str, str]], system: str, tools: str, request: ChatCompletionRequest):
if request.stream:

2
src/llmtuner/data/utils.py
View File

@@ -20,8 +20,8 @@ class Role(str, Enum):
USER = "user"
ASSISTANT = "assistant"
SYSTEM = "system"
OBSERVATION = "observation"
FUNCTION = "function"
OBSERVATION = "observation"
def checksum(data_files: List[str], file_sha1: Optional[str] = None) -> None:

3
src/llmtuner/extras/misc.py
View File

@@ -10,6 +10,7 @@ from transformers.utils import (
WEIGHTS_NAME,
is_torch_bf16_gpu_available,
is_torch_cuda_available,
is_torch_mps_available,
is_torch_npu_available,
is_torch_xpu_available,
)
@@ -133,6 +134,8 @@ def get_current_device() -> torch.device:
device = "xpu:{}".format(os.environ.get("LOCAL_RANK", "0"))
elif is_torch_npu_available():
device = "npu:{}".format(os.environ.get("LOCAL_RANK", "0"))
elif is_torch_mps_available():
device = "mps:{}".format(os.environ.get("LOCAL_RANK", "0"))
elif is_torch_cuda_available():
device = "cuda:{}".format(os.environ.get("LOCAL_RANK", "0"))
else:

44
src/llmtuner/hparams/data_args.py
View File

@@ -9,30 +9,40 @@ class DataArguments:
"""
template: Optional[str] = field(
default=None, metadata={"help": "Which template to use for constructing prompts in training and inference."}
default=None,
metadata={"help": "Which template to use for constructing prompts in training and inference."},
)
dataset: Optional[str] = field(
default=None,
metadata={"help": "The name of provided dataset(s) to use. Use commas to separate multiple datasets."},
)
dataset_dir: Optional[str] = field(
default="data", metadata={"help": "Path to the folder containing the datasets."}
default="data",
metadata={"help": "Path to the folder containing the datasets."},
)
split: Optional[str] = field(
default="train", metadata={"help": "Which dataset split to use for training and evaluation."}
default="train",
metadata={"help": "Which dataset split to use for training and evaluation."},
)
cutoff_len: Optional[int] = field(
default=1024, metadata={"help": "The cutoff length of the model inputs after tokenization."}
default=1024,
metadata={"help": "The cutoff length of the model inputs after tokenization."},
)
reserved_label_len: Optional[int] = field(
default=1, metadata={"help": "The minimum cutoff length reserved for label after tokenization."}
default=1,
metadata={"help": "The minimum cutoff length reserved for label after tokenization."},
)
train_on_prompt: Optional[bool] = field(
default=False, metadata={"help": "Whether to disable the mask on the prompt or not."}
default=False,
metadata={"help": "Whether to disable the mask on the prompt or not."},
)
streaming: Optional[bool] = field(
default=False,
metadata={"help": "Enable dataset streaming."},
)
streaming: Optional[bool] = field(default=False, metadata={"help": "Enable dataset streaming."})
buffer_size: Optional[int] = field(
default=16384, metadata={"help": "Size of the buffer to randomly sample examples from in dataset streaming."}
default=16384,
metadata={"help": "Size of the buffer to randomly sample examples from in dataset streaming."},
)
mix_strategy: Optional[Literal["concat", "interleave_under", "interleave_over"]] = field(
default="concat",
@@ -43,13 +53,16 @@ class DataArguments:
metadata={"help": "Probabilities to sample data from datasets. Use commas to separate multiple datasets."},
)
overwrite_cache: Optional[bool] = field(
default=False, metadata={"help": "Overwrite the cached training and evaluation sets."}
default=False,
metadata={"help": "Overwrite the cached training and evaluation sets."},
)
preprocessing_num_workers: Optional[int] = field(
default=None, metadata={"help": "The number of processes to use for the preprocessing."}
default=None,
metadata={"help": "The number of processes to use for the preprocessing."},
)
max_samples: Optional[int] = field(
default=None, metadata={"help": "For debugging purposes, truncate the number of examples for each dataset."}
default=None,
metadata={"help": "For debugging purposes, truncate the number of examples for each dataset."},
)
eval_num_beams: Optional[int] = field(
default=None,
@@ -62,13 +75,16 @@ class DataArguments:
},
)
val_size: Optional[float] = field(
default=0, metadata={"help": "Size of the development set, should be an integer or a float in range `[0,1)`."}
default=0,
metadata={"help": "Size of the development set, should be an integer or a float in range `[0,1)`."},
)
sft_packing: Optional[bool] = field(
default=False, metadata={"help": "Packing the questions and answers in the supervised fine-tuning stage."}
default=False,
metadata={"help": "Packing the questions and answers in the supervised fine-tuning stage."},
)
cache_path: Optional[str] = field(
default=None, metadata={"help": "Path to save or load the preprocessed datasets."}
default=None,
metadata={"help": "Path to save or load the preprocessed datasets."},
)
def __post_init__(self):

34
src/llmtuner/hparams/evaluation_args.py
View File

@@ -11,15 +11,33 @@ class EvaluationArguments:
Arguments pertaining to specify the evaluation parameters.
"""
task: str = field(metadata={"help": "Name of the evaluation task."})
task_dir: Optional[str] = field(
default="evaluation", metadata={"help": "Path to the folder containing the evaluation datasets."}
task: str = field(
metadata={"help": "Name of the evaluation task."},
)
task_dir: Optional[str] = field(
default="evaluation",
metadata={"help": "Path to the folder containing the evaluation datasets."},
)
batch_size: Optional[int] = field(
default=4,
metadata={"help": "The batch size per GPU for evaluation."},
)
seed: Optional[int] = field(
default=42,
metadata={"help": "Random seed to be used with data loaders."},
)
lang: Optional[Literal["en", "zh"]] = field(
default="en",
metadata={"help": "Language used at evaluation."},
)
n_shot: Optional[int] = field(
default=5,
metadata={"help": "Number of examplars for few-shot learning."},
)
save_dir: Optional[str] = field(
default=None,
metadata={"help": "Path to save the evaluation results."},
)
batch_size: Optional[int] = field(default=4, metadata={"help": "The batch size per GPU for evaluation."})
seed: Optional[int] = field(default=42, metadata={"help": "Random seed to be used with data loaders."})
lang: Optional[Literal["en", "zh"]] = field(default="en", metadata={"help": "Language used at evaluation."})
n_shot: Optional[int] = field(default=5, metadata={"help": "Number of examplars for few-shot learning."})
save_dir: Optional[str] = field(default=None, metadata={"help": "Path to save the evaluation results."})
download_mode: Optional[DownloadMode] = field(
default=DownloadMode.REUSE_DATASET_IF_EXISTS,
metadata={"help": "Download mode used for the evaluation datasets."},

120
src/llmtuner/hparams/finetuning_args.py
View File

@@ -10,20 +10,25 @@ class FreezeArguments:
"""
name_module_trainable: Optional[str] = field(
default="mlp",
default=None,
metadata={
"help": 'Name of trainable modules for partial-parameter (freeze) fine-tuning. \
Use commas to separate multiple modules. \
LLaMA choices: ["mlp", "self_attn"], \
BLOOM & Falcon & ChatGLM choices: ["mlp", "self_attention"], \
Qwen choices: ["mlp", "attn"], \
Phi choices: ["mlp", "mixer"], \
InternLM2 choices: ["feed_forward", "attention"], \
Others choices: the same as LLaMA.'
"help": """Name of trainable modules for partial-parameter (freeze) fine-tuning. \
Use commas to separate multiple modules. \
Use "all" to specify all the available modules. \
LLaMA choices: ["mlp", "self_attn"], \
BLOOM & Falcon & ChatGLM choices: ["mlp", "self_attention"], \
Qwen choices: ["mlp", "attn"], \
InternLM2 choices: ["feed_forward", "attention"], \
Others choices: the same as LLaMA."""
},
)
num_layer_trainable: Optional[int] = field(
default=3, metadata={"help": "The number of trainable layers for partial-parameter (freeze) fine-tuning."}
default=3,
metadata={"help": "The number of trainable layers for partial-parameter (freeze) fine-tuning."},
)
use_llama_pro: Optional[bool] = field(
default=False,
metadata={"help": "Whether or not to use llama pro for partial-parameter (freeze) fine-tuning."},
)
@@ -40,27 +45,42 @@ class LoraArguments:
},
)
lora_alpha: Optional[int] = field(
default=None, metadata={"help": "The scale factor for LoRA fine-tuning (default: lora_rank * 2)."}
default=None,
metadata={"help": "The scale factor for LoRA fine-tuning (default: lora_rank * 2)."},
)
lora_dropout: Optional[float] = field(
default=0.0,
metadata={"help": "Dropout rate for the LoRA fine-tuning."},
)
lora_rank: Optional[int] = field(
default=8,
metadata={"help": "The intrinsic dimension for LoRA fine-tuning."},
)
lora_dropout: Optional[float] = field(default=0.0, metadata={"help": "Dropout rate for the LoRA fine-tuning."})
lora_rank: Optional[int] = field(default=8, metadata={"help": "The intrinsic dimension for LoRA fine-tuning."})
lora_target: Optional[str] = field(
default=None,
metadata={
"help": 'Name(s) of target modules to apply LoRA. Use commas to separate multiple modules. \
LLaMA choices: ["q_proj", "k_proj", "v_proj", "o_proj", "gate_proj", "up_proj", "down_proj"], \
BLOOM & Falcon & ChatGLM choices: ["query_key_value", "dense", "dense_h_to_4h", "dense_4h_to_h"], \
Baichuan choices: ["W_pack", "o_proj", "gate_proj", "up_proj", "down_proj"], \
Qwen choices: ["c_attn", "attn.c_proj", "w1", "w2", "mlp.c_proj"], \
Phi choices: ["Wqkv", "out_proj", "fc1", "fc2"], \
Others choices: the same as LLaMA.'
"help": """Name(s) of target modules to apply LoRA. \
Use commas to separate multiple modules. \
Use "all" to specify all the available modules. \
LLaMA choices: ["q_proj", "k_proj", "v_proj", "o_proj", "gate_proj", "up_proj", "down_proj"], \
BLOOM & Falcon & ChatGLM choices: ["query_key_value", "dense", "dense_h_to_4h", "dense_4h_to_h"], \
Baichuan choices: ["W_pack", "o_proj", "gate_proj", "up_proj", "down_proj"], \
Qwen choices: ["c_attn", "attn.c_proj", "w1", "w2", "mlp.c_proj"], \
InternLM2 choices: ["wqkv", "wo", "w1", "w2", "w3"], \
Others choices: the same as LLaMA."""
},
)
lora_bf16_mode: Optional[bool] = field(
default=False, metadata={"help": "Whether or not to train lora adapters in bf16 precision."}
default=False,
metadata={"help": "Whether or not to train lora adapters in bf16 precision."},
)
use_rslora: Optional[bool] = field(
default=False,
metadata={"help": "Whether or not to use the rank stabilization scaling factor for LoRA layer."},
)
create_new_adapter: Optional[bool] = field(
default=False, metadata={"help": "Whether or not to create a new adapter with randomly initialized weight."}
default=False,
metadata={"help": "Whether or not to create a new adapter with randomly initialized weight."},
)
@@ -70,49 +90,65 @@ class RLHFArguments:
Arguments pertaining to the PPO and DPO training.
"""
dpo_beta: Optional[float] = field(default=0.1, metadata={"help": "The beta parameter for the DPO loss."})
dpo_beta: Optional[float] = field(
default=0.1,
metadata={"help": "The beta parameter for the DPO loss."},
)
dpo_loss: Optional[Literal["sigmoid", "hinge", "ipo", "kto"]] = field(
default="sigmoid", metadata={"help": "The type of DPO loss to use."}
default="sigmoid",
metadata={"help": "The type of DPO loss to use."},
)
dpo_ftx: Optional[float] = field(
default=0, metadata={"help": "The supervised fine-tuning loss coefficient in DPO training."}
default=0,
metadata={"help": "The supervised fine-tuning loss coefficient in DPO training."},
)
ppo_buffer_size: Optional[int] = field(
default=1,
metadata={"help": "The number of mini-batches to make experience buffer in a PPO optimization step."},
)
ppo_epochs: Optional[int] = field(
default=4, metadata={"help": "The number of epochs to perform in a PPO optimization step."}
default=4,
metadata={"help": "The number of epochs to perform in a PPO optimization step."},
)
ppo_logger: Optional[str] = field(
default=None, metadata={"help": 'Log with either "wandb" or "tensorboard" in PPO training.'}
default=None,
metadata={"help": 'Log with either "wandb" or "tensorboard" in PPO training.'},
)
ppo_score_norm: Optional[bool] = field(
default=False, metadata={"help": "Use score normalization in PPO training."}
default=False,
metadata={"help": "Use score normalization in PPO training."},
)
ppo_target: Optional[float] = field(
default=6.0, metadata={"help": "Target KL value for adaptive KL control in PPO training."}
default=6.0,
metadata={"help": "Target KL value for adaptive KL control in PPO training."},
)
ppo_whiten_rewards: Optional[bool] = field(
default=False, metadata={"help": "Whiten the rewards before compute advantages in PPO training."}
default=False,
metadata={"help": "Whiten the rewards before compute advantages in PPO training."},
)
ref_model: Optional[str] = field(
default=None, metadata={"help": "Path to the reference model used for the PPO or DPO training."}
default=None,
metadata={"help": "Path to the reference model used for the PPO or DPO training."},
)
ref_model_adapters: Optional[str] = field(
default=None, metadata={"help": "Path to the adapters of the reference model."}
default=None,
metadata={"help": "Path to the adapters of the reference model."},
)
ref_model_quantization_bit: Optional[int] = field(
default=None, metadata={"help": "The number of bits to quantize the reference model."}
default=None,
metadata={"help": "The number of bits to quantize the reference model."},
)
reward_model: Optional[str] = field(
default=None, metadata={"help": "Path to the reward model used for the PPO training."}
default=None,
metadata={"help": "Path to the reward model used for the PPO training."},
)
reward_model_adapters: Optional[str] = field(
default=None, metadata={"help": "Path to the adapters of the reward model."}
default=None,
metadata={"help": "Path to the adapters of the reward model."},
)
reward_model_quantization_bit: Optional[int] = field(
default=None, metadata={"help": "The number of bits to quantize the reward model."}
default=None,
metadata={"help": "The number of bits to quantize the reward model."},
)
reward_model_type: Optional[Literal["lora", "full", "api"]] = field(
default="lora",
@@ -127,16 +163,20 @@ class FinetuningArguments(FreezeArguments, LoraArguments, RLHFArguments):
"""
stage: Optional[Literal["pt", "sft", "rm", "ppo", "dpo"]] = field(
default="sft", metadata={"help": "Which stage will be performed in training."}
default="sft",
metadata={"help": "Which stage will be performed in training."},
)
finetuning_type: Optional[Literal["lora", "freeze", "full"]] = field(
default="lora", metadata={"help": "Which fine-tuning method to use."}
default="lora",
metadata={"help": "Which fine-tuning method to use."},
)
disable_version_checking: Optional[bool] = field(
default=False, metadata={"help": "Whether or not to disable version checking."}
default=False,
metadata={"help": "Whether or not to disable version checking."},
)
plot_loss: Optional[bool] = field(
default=False, metadata={"help": "Whether or not to save the training loss curves."}
default=False,
metadata={"help": "Whether or not to save the training loss curves."},
)
def __post_init__(self):

15
src/llmtuner/hparams/generating_args.py
View File

@@ -9,10 +9,12 @@ class GeneratingArguments:
"""
do_sample: Optional[bool] = field(
default=True, metadata={"help": "Whether or not to use sampling, use greedy decoding otherwise."}
default=True,
metadata={"help": "Whether or not to use sampling, use greedy decoding otherwise."},
)
temperature: Optional[float] = field(
default=0.95, metadata={"help": "The value used to modulate the next token probabilities."}
default=0.95,
metadata={"help": "The value used to modulate the next token probabilities."},
)
top_p: Optional[float] = field(
default=0.7,
@@ -25,7 +27,8 @@ class GeneratingArguments:
metadata={"help": "The number of highest probability vocabulary tokens to keep for top-k filtering."},
)
num_beams: Optional[int] = field(
default=1, metadata={"help": "Number of beams for beam search. 1 means no beam search."}
default=1,
metadata={"help": "Number of beams for beam search. 1 means no beam search."},
)
max_length: Optional[int] = field(
default=512,
@@ -36,10 +39,12 @@ class GeneratingArguments:
metadata={"help": "The maximum numbers of tokens to generate, ignoring the number of tokens in the prompt."},
)
repetition_penalty: Optional[float] = field(
default=1.0, metadata={"help": "The parameter for repetition penalty. 1.0 means no penalty."}
default=1.0,
metadata={"help": "The parameter for repetition penalty. 1.0 means no penalty."},
)
length_penalty: Optional[float] = field(
default=1.0, metadata={"help": "Exponential penalty to the length that is used with beam-based generation."}
default=1.0,
metadata={"help": "Exponential penalty to the length that is used with beam-based generation."},
)
def to_dict(self) -> Dict[str, Any]:

78
src/llmtuner/hparams/model_args.py
View File

@@ -9,10 +9,13 @@ class ModelArguments:
"""
model_name_or_path: str = field(
metadata={"help": "Path to the model weight or identifier from huggingface.co/models or modelscope.cn/models."}
metadata={
"help": "Path to the model weight or identifier from huggingface.co/models or modelscope.cn/models."
},
)
adapter_name_or_path: Optional[str] = field(
default=None, metadata={"help": "Path to the adapter weight or identifier from huggingface.co/models."}
default=None,
metadata={"help": "Path to the adapter weight or identifier from huggingface.co/models."},
)
cache_dir: Optional[str] = field(
default=None,
@@ -23,7 +26,8 @@ class ModelArguments:
metadata={"help": "Whether or not to use one of the fast tokenizer (backed by the tokenizers library)."},
)
resize_vocab: Optional[bool] = field(
default=False, metadata={"help": "Whether or not to resize the tokenizer vocab and the embedding layers."}
default=False,
metadata={"help": "Whether or not to resize the tokenizer vocab and the embedding layers."},
)
split_special_tokens: Optional[bool] = field(
default=False,
@@ -34,60 +38,88 @@ class ModelArguments:
metadata={"help": "The specific model version to use (can be a branch name, tag name or commit id)."},
)
quantization_bit: Optional[int] = field(
default=None, metadata={"help": "The number of bits to quantize the model."}
default=None,
metadata={"help": "The number of bits to quantize the model."},
)
quantization_type: Optional[Literal["fp4", "nf4"]] = field(
default="nf4", metadata={"help": "Quantization data type to use in int4 training."}
default="nf4",
metadata={"help": "Quantization data type to use in int4 training."},
)
double_quantization: Optional[bool] = field(
default=True, metadata={"help": "Whether or not to use double quantization in int4 training."}
default=True,
metadata={"help": "Whether or not to use double quantization in int4 training."},
)
rope_scaling: Optional[Literal["linear", "dynamic"]] = field(
default=None, metadata={"help": "Which scaling strategy should be adopted for the RoPE embeddings."}
default=None,
metadata={"help": "Which scaling strategy should be adopted for the RoPE embeddings."},
)
flash_attn: Optional[bool] = field(
default=False, metadata={"help": "Enable FlashAttention-2 for faster training."}
default=False,
metadata={"help": "Enable FlashAttention-2 for faster training."},
)
shift_attn: Optional[bool] = field(
default=False, metadata={"help": "Enable shift short attention (S^2-Attn) proposed by LongLoRA."}
default=False,
metadata={"help": "Enable shift short attention (S^2-Attn) proposed by LongLoRA."},
)
use_unsloth: Optional[bool] = field(
default=False, metadata={"help": "Whether or not to use unsloth's optimization for the LoRA training."}
default=False,
metadata={"help": "Whether or not to use unsloth's optimization for the LoRA training."},
)
disable_gradient_checkpointing: Optional[bool] = field(
default=False, metadata={"help": "Whether or not to disable gradient checkpointing."}
default=False,
metadata={"help": "Whether or not to disable gradient checkpointing."},
)
upcast_layernorm: Optional[bool] = field(
default=False, metadata={"help": "Whether or not to upcast the layernorm weights in fp32."}
default=False,
metadata={"help": "Whether or not to upcast the layernorm weights in fp32."},
)
upcast_lmhead_output: Optional[bool] = field(
default=False, metadata={"help": "Whether or not to upcast the output of lm_head in fp32."}
default=False,
metadata={"help": "Whether or not to upcast the output of lm_head in fp32."},
)
hf_hub_token: Optional[str] = field(
default=None,
metadata={"help": "Auth token to log in with Hugging Face Hub."},
)
ms_hub_token: Optional[str] = field(
default=None,
metadata={"help": "Auth token to log in with ModelScope Hub."},
)
hf_hub_token: Optional[str] = field(default=None, metadata={"help": "Auth token to log in with Hugging Face Hub."})
ms_hub_token: Optional[str] = field(default=None, metadata={"help": "Auth token to log in with ModelScope Hub."})
export_dir: Optional[str] = field(
default=None, metadata={"help": "Path to the directory to save the exported model."}
default=None,
metadata={"help": "Path to the directory to save the exported model."},
)
export_size: Optional[int] = field(
default=1, metadata={"help": "The file shard size (in GB) of the exported model."}
default=1,
metadata={"help": "The file shard size (in GB) of the exported model."},
)
export_quantization_bit: Optional[int] = field(
default=None, metadata={"help": "The number of bits to quantize the exported model."}
default=None,
metadata={"help": "The number of bits to quantize the exported model."},
)
export_quantization_dataset: Optional[str] = field(
default=None, metadata={"help": "Path to the dataset or dataset name to use in quantizing the exported model."}
default=None,
metadata={"help": "Path to the dataset or dataset name to use in quantizing the exported model."},
)
export_quantization_nsamples: Optional[int] = field(
default=128, metadata={"help": "The number of samples used for quantization."}
default=128,
metadata={"help": "The number of samples used for quantization."},
)
export_quantization_maxlen: Optional[int] = field(
default=1024, metadata={"help": "The maximum length of the model inputs used for quantization."}
default=1024,
metadata={"help": "The maximum length of the model inputs used for quantization."},
)
export_legacy_format: Optional[bool] = field(
default=False, metadata={"help": "Whether or not to save the `.bin` files instead of `.safetensors`."}
default=False,
metadata={"help": "Whether or not to save the `.bin` files instead of `.safetensors`."},
)
export_hub_model_id: Optional[str] = field(
default=None, metadata={"help": "The name of the repository if push the model to the Hugging Face hub."}
default=None,
metadata={"help": "The name of the repository if push the model to the Hugging Face hub."},
)
print_param_status: Optional[bool] = field(
default=False,
metadata={"help": "For debugging purposes, print the status of the parameters in the model."},
)
def __post_init__(self):

34
src/llmtuner/hparams/parser.py
View File

@@ -30,12 +30,15 @@ _EVAL_ARGS = [ModelArguments, DataArguments, EvaluationArguments, FinetuningArgu
_EVAL_CLS = Tuple[ModelArguments, DataArguments, EvaluationArguments, FinetuningArguments]
def _check_dependencies():
require_version("transformers>=4.37.2", "To fix: pip install transformers>=4.37.2")
require_version("datasets>=2.14.3", "To fix: pip install datasets>=2.14.3")
require_version("accelerate>=0.21.0", "To fix: pip install accelerate>=0.21.0")
require_version("peft>=0.7.0", "To fix: pip install peft>=0.7.0")
require_version("trl>=0.7.6", "To fix: pip install trl>=0.7.6")
def _check_dependencies(disabled: bool) -> None:
if disabled:
logger.warning("Version checking has been disabled, may lead to unexpected behaviors.")
else:
require_version("transformers>=4.37.2", "To fix: pip install transformers>=4.37.2")
require_version("datasets>=2.14.3", "To fix: pip install datasets>=2.14.3")
require_version("accelerate>=0.21.0", "To fix: pip install accelerate>=0.21.0")
require_version("peft>=0.8.2", "To fix: pip install peft>=0.8.2")
require_version("trl>=0.7.6", "To fix: pip install trl>=0.7.6")
def _parse_args(parser: "HfArgumentParser", args: Optional[Dict[str, Any]] = None) -> Tuple[Any]:
@@ -130,6 +133,13 @@ def get_train_args(args: Optional[Dict[str, Any]] = None) -> _TRAIN_CLS:
if training_args.do_train and training_args.predict_with_generate:
raise ValueError("`predict_with_generate` cannot be set as True while training.")
if (
training_args.do_train
and finetuning_args.finetuning_type == "freeze"
and finetuning_args.name_module_trainable is None
):
raise ValueError("Please specify `name_module_trainable` in Freeze training.")
if training_args.do_train and finetuning_args.finetuning_type == "lora" and finetuning_args.lora_target is None:
raise ValueError("Please specify `lora_target` in LoRA training.")
@@ -137,9 +147,7 @@ def get_train_args(args: Optional[Dict[str, Any]] = None) -> _TRAIN_CLS:
raise ValueError("Install Unsloth: https://github.com/unslothai/unsloth")
_verify_model_args(model_args, finetuning_args)
if not finetuning_args.disable_version_checking:
_check_dependencies()
_check_dependencies(disabled=finetuning_args.disable_version_checking)
if (
training_args.do_train
@@ -240,13 +248,11 @@ def get_infer_args(args: Optional[Dict[str, Any]] = None) -> _INFER_CLS:
_set_transformers_logging()
_verify_model_args(model_args, finetuning_args)
_check_dependencies(disabled=finetuning_args.disable_version_checking)
if data_args.template is None:
raise ValueError("Please specify which `template` to use.")
if not finetuning_args.disable_version_checking:
_check_dependencies()
return model_args, data_args, finetuning_args, generating_args
@@ -255,13 +261,11 @@ def get_eval_args(args: Optional[Dict[str, Any]] = None) -> _EVAL_CLS:
_set_transformers_logging()
_verify_model_args(model_args, finetuning_args)
_check_dependencies(disabled=finetuning_args.disable_version_checking)
if data_args.template is None:
raise ValueError("Please specify which `template` to use.")
if not finetuning_args.disable_version_checking:
_check_dependencies()
transformers.set_seed(eval_args.seed)
return model_args, data_args, eval_args, finetuning_args

4
src/llmtuner/model/__init__.py
View File

@@ -1,5 +1,5 @@
from .loader import load_model_and_tokenizer
from .utils import dispatch_model, get_modelcard_args, load_valuehead_params
from .utils import dispatch_model, load_valuehead_params
__all__ = ["load_model_and_tokenizer", "dispatch_model", "get_modelcard_args", "load_valuehead_params"]
__all__ = ["load_model_and_tokenizer", "dispatch_model", "load_valuehead_params"]

51
src/llmtuner/model/adapter.py
View File

@@ -1,8 +1,7 @@
import inspect
from typing import TYPE_CHECKING
import torch
from peft import LoraConfig, PeftModel, TaskType, get_peft_model
from peft import LoraConfig, LoraModel, PeftModel, TaskType, get_peft_model
from transformers.integrations import is_deepspeed_zero3_enabled
from ..extras.logging import get_logger
@@ -47,12 +46,22 @@ def init_adapter(
if not num_layers:
raise ValueError("Current model does not support freeze tuning.")
if finetuning_args.num_layer_trainable > 0: # fine-tuning the last n layers if num_layer_trainable > 0
trainable_layer_ids = [num_layers - k - 1 for k in range(finetuning_args.num_layer_trainable)]
else: # fine-tuning the first n layers if num_layer_trainable < 0
trainable_layer_ids = [k for k in range(-finetuning_args.num_layer_trainable)] # noqa: C416
if finetuning_args.use_llama_pro:
if num_layers % finetuning_args.num_layer_trainable != 0:
raise ValueError(
"`num_layers` {} should be divisible by `num_layer_trainable` {}.".format(
num_layers, finetuning_args.num_layer_trainable
)
)
freeze_modules = set()
stride = num_layers // finetuning_args.num_layer_trainable
trainable_layer_ids = range(stride - 1, num_layers + stride - 1, stride)
elif finetuning_args.num_layer_trainable > 0: # fine-tuning the last n layers if num_layer_trainable > 0
trainable_layer_ids = range(num_layers - finetuning_args.num_layer_trainable, num_layers)
else: # fine-tuning the first n layers if num_layer_trainable < 0
trainable_layer_ids = range(-finetuning_args.num_layer_trainable)
freeze_modules = {"all"}
for name, _ in model.named_modules():
if "0." in name:
freeze_modules.add(name.split("0.")[-1].split(".")[0])
@@ -65,13 +74,13 @@ def init_adapter(
)
for idx in trainable_layer_ids:
trainable_layers.append("{:d}.{}".format(idx, module_name))
trainable_layers.append(".{:d}.{}".format(idx, module_name if module_name != "all" else ""))
for name, param in model.named_parameters():
if not any(trainable_layer in name for trainable_layer in trainable_layers):
param.requires_grad_(False)
else:
if any(trainable_layer in name for trainable_layer in trainable_layers):
param.data = param.data.to(torch.float32)
else:
param.requires_grad_(False)
if finetuning_args.finetuning_type == "lora":
logger.info("Fine-tuning method: LoRA")
@@ -94,7 +103,7 @@ def init_adapter(
adapter_to_merge = model_args.adapter_name_or_path
for adapter in adapter_to_merge:
model = PeftModel.from_pretrained(model, adapter)
model: "LoraModel" = PeftModel.from_pretrained(model, adapter)
model = model.merge_and_unload()
if len(adapter_to_merge) > 0:
@@ -114,22 +123,14 @@ def init_adapter(
"target_modules": target_modules,
"lora_alpha": finetuning_args.lora_alpha,
"lora_dropout": finetuning_args.lora_dropout,
"use_rslora": finetuning_args.use_rslora,
}
if model_args.use_unsloth:
from unsloth import FastLlamaModel, FastMistralModel # type: ignore
from unsloth import FastLanguageModel # type: ignore
unsloth_peft_kwargs = {"model": model, "max_seq_length": model_args.model_max_length}
if "loftq_config" in inspect.signature(FastLlamaModel.get_peft_model).parameters:
unsloth_peft_kwargs["loftq_config"] = {}
if getattr(model.config, "model_type", None) == "llama":
model = FastLlamaModel.get_peft_model(**peft_kwargs, **unsloth_peft_kwargs)
elif getattr(model.config, "model_type", None) == "mistral":
model = FastMistralModel.get_peft_model(**peft_kwargs, **unsloth_peft_kwargs)
else:
raise NotImplementedError
model = FastLanguageModel.get_peft_model(**peft_kwargs, **unsloth_peft_kwargs)
else:
lora_config = LoraConfig(
task_type=TaskType.CAUSAL_LM,
@@ -142,7 +143,7 @@ def init_adapter(
for param in filter(lambda p: p.requires_grad, model.parameters()):
param.data = param.data.to(torch.bfloat16 if finetuning_args.lora_bf16_mode else torch.float32)
if model_args.adapter_name_or_path is not None:
logger.info("Loaded adapter(s): {}".format(",".join(model_args.adapter_name_or_path)))
if model_args.adapter_name_or_path is not None:
logger.info("Loaded adapter(s): {}".format(",".join(model_args.adapter_name_or_path)))
return model

31
src/llmtuner/model/loader.py
View File

@@ -55,7 +55,7 @@ def load_model_and_tokenizer(
model = None
if is_trainable and model_args.use_unsloth:
from unsloth import FastLlamaModel, FastMistralModel # type: ignore
from unsloth import FastLanguageModel # type: ignore
unsloth_kwargs = {
"model_name": model_args.model_name_or_path,
@@ -63,14 +63,12 @@ def load_model_and_tokenizer(
"dtype": model_args.compute_dtype,
"load_in_4bit": model_args.quantization_bit == 4,
"token": model_args.hf_hub_token,
"device_map": get_current_device(),
"device_map": {"": get_current_device()},
"rope_scaling": getattr(config, "rope_scaling", None),
}
if getattr(config, "model_type", None) == "llama":
model, _ = FastLlamaModel.from_pretrained(**unsloth_kwargs)
elif getattr(config, "model_type", None) == "mistral":
model, _ = FastMistralModel.from_pretrained(**unsloth_kwargs)
else:
try:
model, _ = FastLanguageModel.from_pretrained(**unsloth_kwargs)
except NotImplementedError:
logger.warning("Unsloth does not support model type {}.".format(getattr(config, "model_type", None)))
model_args.use_unsloth = False
@@ -87,17 +85,6 @@ def load_model_and_tokenizer(
**config_kwargs,
)
# Add llama-factory tag to push these tags on the Hub.
# the feature is available since 4.37.0 but adding the check
# just in case
if hasattr(model, "add_model_tags"):
model.add_model_tags(["llama-factory"])
else:
logger.warning_once(
"Was not able to properly tag the model, if you want to use the model tagging feature, make sure to "
"have transformers>=4.37.0 installed on your environment."
)
patch_model(model, tokenizer, model_args, is_trainable)
register_autoclass(config, model, tokenizer)
@@ -134,4 +121,12 @@ def load_model_and_tokenizer(
if not is_trainable:
logger.info("This IS expected that the trainable params is 0 if you are using model for inference only.")
if model_args.print_param_status:
for name, param in model.named_parameters():
print(
"name: {}, dtype: {}, device: {}, trainable: {}".format(
name, param.dtype, param.device, param.requires_grad
)
)
return model, tokenizer

5
src/llmtuner/model/patcher.py
View File

@@ -300,6 +300,11 @@ def patch_model(
if is_trainable:
patch_mixtral_replace_moe_impl()
try:
model.add_model_tags(["llama-factory"])
except Exception:
logger.warning("Cannot properly tag the model.")
def patch_valuehead_model(model: "AutoModelForCausalLMWithValueHead") -> None:
def tie_weights(self: "AutoModelForCausalLMWithValueHead") -> None:

16
src/llmtuner/model/utils.py
View File

@@ -1,5 +1,5 @@
import inspect
from typing import TYPE_CHECKING, Any, Dict, List
from typing import TYPE_CHECKING, Dict, List
import torch
from transformers import PreTrainedModel
@@ -13,7 +13,7 @@ from ..extras.misc import get_current_device
if TYPE_CHECKING:
from transformers import PretrainedConfig, PreTrainedTokenizer
from ..hparams import DataArguments, FinetuningArguments, ModelArguments
from ..hparams import ModelArguments
logger = get_logger(__name__)
@@ -76,18 +76,6 @@ def find_all_linear_modules(model: "PreTrainedModel") -> List[str]:
return list(module_names)
def get_modelcard_args(
model_args: "ModelArguments", data_args: "DataArguments", finetuning_args: "FinetuningArguments"
) -> Dict[str, Any]:
return {
"tasks": "text-generation",
"license": "other",
"finetuned_from": model_args.model_name_or_path,
"dataset": [dataset.strip() for dataset in data_args.dataset.split(",")],
"tags": ["llama-factory"] + (["lora"] if finetuning_args.finetuning_type == "lora" else []),
}
def load_valuehead_params(path_or_repo_id: str, model_args: "ModelArguments") -> Dict[str, torch.Tensor]:
r"""
Loads value head parameters from Hugging Face Hub or local disk.

22
src/llmtuner/train/utils.py
View File

@@ -4,7 +4,7 @@ import torch
from ..extras.logging import get_logger
from ..hparams import FinetuningArguments, ModelArguments
from ..model import get_modelcard_args, load_model_and_tokenizer, load_valuehead_params
from ..model import load_model_and_tokenizer, load_valuehead_params
if TYPE_CHECKING:
@@ -25,14 +25,18 @@ def create_modelcard_and_push(
training_args: "Seq2SeqTrainingArguments",
finetuning_args: "FinetuningArguments",
) -> None:
if training_args.do_train:
if training_args.push_to_hub:
trainer.push_to_hub(**get_modelcard_args(model_args, data_args, finetuning_args))
return
try:
trainer.create_model_card(**get_modelcard_args(model_args, data_args, finetuning_args))
except Exception as err:
logger.warning("Failed to create model card: {}".format(str(err)))
kwargs = {
"tasks": "text-generation",
"finetuned_from": model_args.model_name_or_path,
"dataset": [dataset.strip() for dataset in data_args.dataset.split(",")],
"tags": ["llama-factory", finetuning_args.finetuning_type],
}
if not training_args.do_train:
pass
elif training_args.push_to_hub:
trainer.push_to_hub(**kwargs)
else:
trainer.create_model_card(license="other", **kwargs) # prevent from connecting to hub
def create_ref_model(