生成式机器人

1、导入相关包

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

2、加载、划分、处理数据集

# 加载
ds = Dataset.load_from_disk("./alpaca_data_zh/")
# 划分

# 数据处理
tokenizer = AutoTokenizer.from_pretrained("../bloom-389m-zh")
def process_func(example):
    MAX_LENGTH = 256
    input_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)    

3、创建模型

model = AutoModelForCausalLM.from_pretrained("../bloom-389m-zh")

4、创建评估函数

5、创建TrainingArguments、Trainer

args = TrainingArguments(
    output_dir="./chatbot",
    per_device_train_batch_size=4,
    gradient_accumulation_steps=8,
    logging_steps=10,
    num_train_epochs=2
)

trainer = Trainer(
    model=model,
    args=args,
    tokenizer=tokenizer,
    train_dataset=tokenized_ds,
    data_collator=DataCollatorForSeq2Seq(tokenizer=tokenizer, padding=True)
)

6、模型训练、评估、预测

# 模型训练
trainer.train()
# 模型评估
trainer.evaluate()

7、模型预测

# 模型预测

# 测试
from transformers import pipeline

pipe = pipeline("text-generation", model=model, tokenizer=tokenizer, device=0)

ipt = "Human: {}\n{}".format("考试有哪些技巧？", "").strip() + "\n\nAssistant: "
pipe(ipt, max_length=256, do_sample=True, )