Skip to content
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.

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

多轮对话数据处理 #349

Open
447428054 opened this issue Jul 6, 2023 · 1 comment
Open

多轮对话数据处理 #349

447428054 opened this issue Jul 6, 2023 · 1 comment

Comments

@447428054
Copy link

447428054 commented Jul 6, 2023

@linonetwo @jsl9208 @xpqiu @meta-tabchen 您好

      with open(os.path.join(self.data_dir, f'{self.data_type}.jsonl'), 'r') as f:
                for line in f:
                    sample = json.loads(line)

                    chat = sample['chat']
                    num_turns = int(sample['num_turns'])

                    meta_instruction = sample['meta_instruction']
                    instruction_ids = self.tokenizer.encode(meta_instruction)
                    assert isinstance(instruction_ids, list) and len(instruction_ids) > 0
                    
                    input_ids = copy.deepcopy(instruction_ids)
                    no_loss_spans = [(0, len(instruction_ids))]

                    for i in range(num_turns):
                        cur_turn_ids = []
                        cur_no_loss_spans = []
                        cur_turn = chat[f'turn_{i+1}']
                        for key, value in cur_turn.items():

                            cur_ids = self.tokenizer.encode(value)

                            if key == 'Tool Responses':
                                # The format tokens (<|Results|>:...<eor>\n) should have losses. 
                                cur_no_loss_spans.append((len(input_ids + cur_turn_ids) + 5, len(input_ids + cur_turn_ids + cur_ids) - 2))    

                            assert isinstance(cur_ids, list) and len(cur_ids) > 0

                            cur_turn_ids.extend(cur_ids)

                        if len(input_ids + cur_turn_ids) > 2048:
                            break

                        input_ids.extend(cur_turn_ids)
                        no_loss_spans.extend(cur_no_loss_spans)

                    if len(input_ids) == len(instruction_ids):
                        continue

                    assert len(input_ids) > 0 and len(input_ids) <= 2048

                    self.data.append(input_ids)
                    self.no_loss_spans.append(no_loss_spans)

请问这种多轮对话只放到一条样本能过加速训练过程吗?
我理解这种方式,减少了前向传播的次数,但是增加了反向传播的长度,而反向传播计算梯度要更耗时,这种方式会不会比每一轮单独抽出来训练要慢呢?

@YanZiBuGuiCHunShiWan
Copy link

我也好奇,还有请问为什么causal lm 不把多轮对话的human提问部分的损失忽略呢?这样不会有影响么?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants