Merge ccb352f901 into 949488fdb4

[fix] keep DPO preference-pair think formatting aligned

README.md

@@ -230,6 +230,10 @@ minimind2系列旧模型均经过权重映射+（微调训练）QKVO线性层校
 # 克隆仓库、安装依赖
 git clone --depth 1 https://github.com/jingyaogong/minimind
 cd minimind && pip install -r requirements.txt -i https://mirrors.aliyun.com/pypi/simple
+#部分老版本Windows powershell不能识别&&，Windows用户应将&&链式命令改为分行写法
+git clone --depth 1 https://github.com/jingyaogong/minimind
+cd minimind 
+pip install -r requirements.txt -i https://mirrors.aliyun.com/pypi/simple
 ```
 
 ## Ⅰ 🚀 模型推理
@@ -242,6 +246,13 @@ cd minimind && pip install -r requirements.txt -i https://mirrors.aliyun.com/pyp
 modelscope download --model gongjy/minimind-3 --local_dir ./minimind-3
 # 方式2
 git clone https://huggingface.co/jingyaogong/minimind-3
+# 国内网络进入huggingface网站存在困难，可使用镜像站
+# Windows
+$env:HF_ENDPOINT = "https://hf-mirror.com" 
+git clone https://hf-mirror.com/jingyaogong/minimind-3
+#Linux
+export HF_ENDPOINT="https://hf-mirror.com"
+git clone https://hf-mirror.com/jingyaogong/minimind-3
 ```
 
 ### 2' CLI 推理

README_en.md

@@ -229,6 +229,10 @@ After this update, maintenance for the entire `minimind-v1` series will be disco
 # Clone repository and install dependencies
 git clone --depth 1 https://github.com/jingyaogong/minimind
 cd minimind && pip install -r requirements.txt -i https://mirrors.aliyun.com/pypi/simple
+#Tip:Some older versions of Windows PowerShell (v5.1 and below) do not recognize the && operator. 
+git clone --depth 1 https://github.com/jingyaogong/minimind
+cd minimind 
+pip install -r requirements.txt -i https://mirrors.aliyun.com/pypi/simple
 ```
 
 ## Ⅰ 🚀 Model Inference

dataset/lm_dataset.py

@@ -28,10 +28,13 @@ def pre_processing_chat(conversations, add_system_ratio=0.2):
             return [{'role': 'system', 'content': random.choice(SYSTEM_PROMPTS)}] + conversations
     return conversations
 
-def post_processing_chat(prompt_content, empty_think_ratio=0.2):
+def post_processing_chat(prompt_content, empty_think_ratio=0.2, remove_empty_think=None):
     # 以80%概率移除空思考标签
-    if '<think>\n\n</think>\n\n' in prompt_content and random.random() > empty_think_ratio:
+    if '<think>\n\n</think>\n\n' in prompt_content:
-        prompt_content = prompt_content.replace('<think>\n\n</think>\n\n', '')
+        if remove_empty_think is None:
+            remove_empty_think = random.random() > empty_think_ratio
+        if remove_empty_think:
+            prompt_content = prompt_content.replace('<think>\n\n</think>\n\n', '')
     return prompt_content
 
 class PretrainDataset(Dataset):
@@ -139,12 +142,13 @@ class DPODataset(Dataset):
         chosen_prompt = self.tokenizer.apply_chat_template(
             chosen, tokenize=False, add_generation_prompt=False
         )
-        chosen_prompt = post_processing_chat(chosen_prompt)
 
         rejected_prompt = self.tokenizer.apply_chat_template(
             rejected, tokenize=False, add_generation_prompt=False
         )
-        rejected_prompt = post_processing_chat(rejected_prompt)
+        remove_empty_think = random.random() > 0.2
+        chosen_prompt = post_processing_chat(chosen_prompt, remove_empty_think=remove_empty_think)
+        rejected_prompt = post_processing_chat(rejected_prompt, remove_empty_think=remove_empty_think)
         chosen_encoding = self.tokenizer(
             chosen_prompt, truncation=True, max_length=self.max_length, padding='max_length'
         )

model/model_minimind.py

@@ -125,10 +125,14 @@ class Attention(nn.Module):
         if self.flash and (seq_len > 1) and (not self.is_causal or past_key_value is None) and (attention_mask is None or torch.all(attention_mask == 1)):
             output = F.scaled_dot_product_attention(xq, xk, xv, dropout_p=self.dropout if self.training else 0.0, is_causal=self.is_causal)
         else:
-            scores = (xq @ xk.transpose(-2, -1)) / math.sqrt(self.head_dim)
+            # 先转 float32 再加 mask：-1e9 超出 float16 的表示范围（min≈-65504），
+            # 在 float16 张量上累加会饱和成 -inf。若某个 query 在因果范围内可见的
+            # key 全被 mask 掉（左 padding 时必然出现），softmax 整行就是 NaN，
+            # 并在下一层通过 K/V 投影扩散到同序列里有效的 token 上。
+            scores = (xq @ xk.transpose(-2, -1)).float() / math.sqrt(self.head_dim)
             if self.is_causal: scores[:, :, :, -seq_len:] += torch.full((seq_len, seq_len), float("-inf"), device=scores.device).triu(1)
             if attention_mask is not None: scores += (1.0 - attention_mask.unsqueeze(1).unsqueeze(2)) * -1e9
-            output = self.attn_dropout(F.softmax(scores.float(), dim=-1).type_as(xq)) @ xv
+            output = self.attn_dropout(F.softmax(scores, dim=-1).type_as(xq)) @ xv
         output = output.transpose(1, 2).reshape(bsz, seq_len, -1)
         output = self.resid_dropout(self.o_proj(output))
         return output, past_kv

requirements.txt

@@ -28,6 +28,8 @@ streamlit==1.50.0
 einops==0.8.1
 swanlab==0.9.8
 modelscope==1.37.0
+fastapi==0.120.1
+uvicorn==0.38.0
 # torch==2.6.0
 # torchvision==0.21.0
 # peft==0.7.1

scripts/serve_openai_api.py

@@ -202,7 +202,7 @@ async def chat_completions(request: ChatRequest):
             with torch.no_grad():
                 generated_ids = model.generate(
                     inputs["input_ids"],
-                    max_length=inputs["input_ids"].shape[1] + request.max_tokens,
+                    max_new_tokens=request.max_tokens,
                     do_sample=True,
                     attention_mask=inputs["attention_mask"],
                     pad_token_id=tokenizer.pad_token_id,

trainer/train_tokenizer.py

@@ -8,18 +8,39 @@ DATA_PATH = '../dataset/sft_t2t_mini.jsonl'
 TOKENIZER_DIR = '../model_learn_tokenizer/'
 VOCAB_SIZE = 6400
 SPECIAL_TOKENS_NUM = 36
+MAX_LINES = 0  # 0 表示读取全部；设为正数则只取前 N 条（快速试跑）
 
-def get_texts(data_path):
+def get_texts(data_path, max_lines=MAX_LINES):
+    """逐行产出用于训练 BPE 的文本，兼容两种主线数据格式。
+
+    pretrain_t2t(_mini).jsonl 是 {"text": ...}，
+    sft_t2t(_mini).jsonl 是 {"conversations": [{"role":..., "content":...}, ...]}。
+    原先只解析 conversations，所以把 DATA_PATH 指向预训练语料时一条文本都取不到，
+    BPE 仅基于 initial_alphabet 训练，产出一个没有任何 merge 的词表，且全程不报错。
+    """
+    used = 0
     with open(data_path, 'r', encoding='utf-8', errors='ignore') as f:
-        for i, line in enumerate(f):
+        for line in f:
-            if i >= 10000: break # 选10000行测试
+            if max_lines and used >= max_lines:
+                break
             try:
                 data = json.loads(line)
-                contents = [item.get('content') for item in data.get('conversations', []) if item.get('content')]
-                if contents:
-                    yield "\n".join(contents)
             except json.JSONDecodeError:
                 continue
+            if 'text' in data:
+                text = str(data['text'])
+            else:
+                contents = [item.get('content') for item in data.get('conversations', []) if item.get('content')]
+                text = "\n".join(contents) if contents else ''
+            if text.strip():
+                used += 1
+                yield text
+    if used == 0:
+        raise ValueError(
+            f'{data_path} 中没有可用文本：每行应为 {{"text": ...}} 或 '
+            f'{{"conversations": [{{"role":..., "content":...}}, ...]}}。'
+            f'继续训练只会得到一个空词表。'
+        )
 
 def train_tokenizer(data_path, tokenizer_dir, vocab_size, special_tokens_num=SPECIAL_TOKENS_NUM):
     tokenizer = Tokenizer(models.BPE())