0、P-Tuning基本原理

 P-Tuning的基本思想是在prompt-tuning的基础上，对prompt部分进行进一步的编码计算，加速收敛。具体来说，PEFT支持两种编码方式，一种是LSTM，一种是MLP，与Prompt-tuning不同的是，Prompt的形式只有Soft Prompt。

1、代码实战

1.1、导包

from datasets import Dataset
from transformers import AutoTokenizer, AutoModelForCausalLM, DataCollatorForSeq2Seq, TrainingArguments, Trainer

1.2、加载数据集

ds = Dataset.load_from_disk("../Data/alpaca_data_zh/")
ds

1.3、数据集预处理

tokenizer = AutoTokenizer.from_pretrained("../Model/bloom-389m-zh")
tokenizer

def process_func(example):MAX_LENGTH = 256input_ids, attention_mask, labels = [], [], []instruction = tokenizer("\n".join(["Human: " + example["instruction"], example["input"]]).strip() + "\n\nAssistant: ")response = tokenizer(example["output"] + tokenizer.eos_token)input_ids = instruction["input_ids"] + response["input_ids"]attention_mask = instruction["attention_mask"] + response["attention_mask"]labels = [-100] * len(instruction["input_ids"]) + response["input_ids"]if len(input_ids) > MAX_LENGTH:input_ids = input_ids[:MAX_LENGTH]attention_mask = attention_mask[:MAX_LENGTH]labels = labels[:MAX_LENGTH]return {"input_ids": input_ids,"attention_mask": attention_mask,"labels": labels}

tokenized_ds = ds.map(process_func, remove_columns=ds.column_names)
tokenized_ds

1.4、创建模型

model = AutoModelForCausalLM.from_pretrained("../Model/bloom-389m-zh", low_cpu_mem_usage=True)

1.5、P-tuning*

1.5.1、配置文件

from peft import PromptEncoderConfig, TaskType, get_peft_model, PromptEncoderReparameterizationTypeconfig = PromptEncoderConfig(task_type=TaskType.CAUSAL_LM, num_virtual_tokens=10,encoder_reparameterization_type=PromptEncoderReparameterizationType.MLP,# encoder_reparameterization_type=PromptEncoderReparameterizationType.LSTM,encoder_dropout=0.1, encoder_num_layers=5, encoder_hidden_size=1024)
print(config)

1.5.2、创建模型

model = get_peft_model(model, config)

model.print_trainable_parameters()

1.6、配置训练参数

args = TrainingArguments(output_dir="./chatbot",per_device_train_batch_size=1,gradient_accumulation_steps=8,logging_steps=10,num_train_epochs=1
)

1.7、创建训练器

args = TrainingArguments(output_dir="./chatbot",per_device_train_batch_size=1,gradient_accumulation_steps=8,logging_steps=10,num_train_epochs=1
)

1.8、模型训练

args = TrainingArguments(output_dir="./chatbot",per_device_train_batch_size=1,gradient_accumulation_steps=8,logging_steps=10,num_train_epochs=1
)

1.9、模型推理

model = model.cuda()
ipt = tokenizer("Human: {}\n{}".format("数学考试有哪些技巧？", "").strip() + "\n\nAssistant: ", return_tensors="pt").to(model.device)
print(tokenizer.decode(model.generate(**ipt, max_length=256, do_sample=True)[0], skip_special_tokens=True))