Merge 59658fb60c into cdf0a93e02

[fix] 用 AST 求值器替换工具里的 eval()，并修复 Windows 上工具调用全部静默失败

scripts/eval_toolcall.py

@@ -12,7 +12,7 @@ from datetime import datetime
 from transformers import AutoTokenizer, AutoModelForCausalLM, TextStreamer
 from openai import OpenAI
 from model.model_minimind import MiniMindConfig, MiniMindForCausalLM
-from trainer.trainer_utils import setup_seed, get_model_params
+from trainer.trainer_utils import setup_seed, get_model_params, safe_math_eval
 warnings.filterwarnings('ignore')
 
 TOOLS = [
@@ -27,7 +27,7 @@ TOOLS = [
 ]
 
 MOCK_RESULTS = {
-    "calculate_math": lambda args: {"result": str(eval(str(args.get("expression", "0")).replace("^", "**").replace("×", "*").replace("÷", "/").replace("−", "-").replace("²", "**2").replace("³", "**3").replace("（", "(").replace("）", ")")))},
+    "calculate_math": lambda args: {"result": str(safe_math_eval(args.get("expression", "0")))},
     "get_current_time": lambda args: {"datetime": datetime.now().strftime("%Y-%m-%d %H:%M:%S"), "timezone": args.get("timezone", "Asia/Shanghai")},
     "random_number": lambda args: {"result": random.randint(int(args.get("min", 0)), int(args.get("max", 100)))},
     "text_length": lambda args: {"characters": len(args.get("text", "")), "words": len(args.get("text", "").split())},

scripts/serve_openai_api.py

@@ -53,7 +53,7 @@ class ChatRequest(BaseModel):
     temperature: float = 0.7
     top_p: float = 0.92
     max_tokens: int = 8192
-    stream: bool = True
+    stream: bool = False  # 对齐 OpenAI 规范：默认非流式，避免标准客户端不传 stream 时误收 SSE
     tools: list = Field(default_factory=list)
     open_thinking: bool = False
     chat_template_kwargs: dict = None

scripts/web_demo.py

@@ -2,6 +2,7 @@ import random
 import re
 import json
 import os
+import sys
 from threading import Thread
 
 import torch
@@ -9,6 +10,9 @@ import numpy as np
 import streamlit as st
 from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
 
+sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__), '..')))
+from trainer.trainer_utils import safe_math_eval
+
 st.set_page_config(page_title="MiniMind", initial_sidebar_state="collapsed")
 
 st.markdown("""
@@ -125,7 +129,7 @@ def execute_tool(tool_name, args):
     import datetime
     try:
         if tool_name == 'calculate_math':
-            return {"result": eval(args.get('expression', '0'))}
+            return {"result": safe_math_eval(args.get('expression', '0'))}
         elif tool_name == 'get_current_time':
             tz = args.get('timezone', 'Asia/Shanghai')
             return {"result": datetime.datetime.now().strftime('%Y-%m-%d %H:%M:%S')}

trainer/train_agent.py

@@ -10,7 +10,6 @@ import gc
 import json
 import math
 import random
-import signal
 import argparse
 import warnings
 import torch
@@ -24,7 +23,7 @@ from torch.optim.lr_scheduler import CosineAnnealingLR
 from transformers import AutoTokenizer
 from model.model_minimind import MiniMindConfig, MiniMindForCausalLM
 from dataset.lm_dataset import AgentRLDataset
-from trainer.trainer_utils import Logger, is_main_process, lm_checkpoint, init_distributed_mode, setup_seed, SkipBatchSampler, init_model, LMForRewardModel
+from trainer.trainer_utils import Logger, is_main_process, lm_checkpoint, init_distributed_mode, setup_seed, SkipBatchSampler, init_model, LMForRewardModel, safe_math_eval
 from trainer.rollout_engine import create_rollout_engine, compute_per_token_logps
 
 warnings.filterwarnings('ignore')
@@ -55,7 +54,7 @@ UNIT_DATA = {"km_miles": 0.621371, "miles_km": 1.60934, "kg_pounds": 2.20462, "p
 
 # ======== 模拟执行 ========
 MOCK_RESULTS = {
-    "calculate_math": lambda args: {"result": str(eval(str(args.get("expression", "0")).replace("^", "**").replace("×", "*").replace("÷", "/").replace("−", "-").replace("（", "(").replace("）", ")"), {"__builtins__": {}, "math": math}))},
+    "calculate_math": lambda args: {"result": str(safe_math_eval(args.get("expression", "0")))},
     "unit_converter": lambda args: {"result": round(float(args.get("value", 0)) * UNIT_DATA.get(f"{args.get('from_unit', '').lower()}_{args.get('to_unit', '').lower()}", 1), 4)},
     "get_current_weather": lambda args: (lambda w: {"city": args.get("location"), "temperature": w[0], "humidity": "65%", "condition": w[1]})(WEATHER_DATA.get(args.get("location"), ("22°C", "晴"))),
     "get_current_time": lambda args: {"datetime": TIME_DATA.get(args.get("timezone", "Asia/Shanghai"), "2025-03-07 14:30:00"), "timezone": args.get("timezone", "Asia/Shanghai")},
@@ -85,14 +84,9 @@ def execute_tool(name, args):
     fn = MOCK_RESULTS.get(name)
     if not fn: return None
     try:
-        signal.signal(signal.SIGALRM, lambda *_: (_ for _ in ()).throw(TimeoutError()))
-        signal.alarm(1)
         return fn(args)
-    except:
+    except Exception:
         return None
-    finally:
-        try: signal.alarm(0)
-        except: pass
 
 # ======== 多轮 Rollout ========
 def rollout_single(rollout_engine, tokenizer, messages, tools, max_turns=3, max_new_tokens=256, thinking_ratio=0.5, device="cuda"):
@@ -278,9 +272,10 @@ def rl_train_epoch(epoch, loader, iters, rollout_engine, ref_model, reward_model
 
         rewards = calculate_rewards(prompts, completions, gt_batch, tools_batch, args.num_generations, reward_model, device=args.device, turn_outputs_batch=turn_outputs_batch, unfinished_batch=unfinished_batch)
 
-        model_unwrapped = model.module if isinstance(model, DistributedDataParallel) else model
         with autocast_ctx:
-            res = model_unwrapped(input_ids, attention_mask=full_mask)
+            # 反向传播必须经过 DDP 包装后的模块：直接调用 .module 会跳过
+            # DDP 的 prepare_for_backward，梯度不会 all-reduce，各卡静默发散。
+            res = model(input_ids, attention_mask=full_mask)
             aux_loss = res.aux_loss if lm_config.use_moe else torch.tensor(0.0, device=args.device)
             logits = res.logits[:, :-1, :]
             per_token_logps = F.log_softmax(logits, dim=-1).gather(2, input_ids[:, 1:].unsqueeze(-1)).squeeze(-1)

trainer/train_grpo.py

@@ -95,9 +95,10 @@ def grpo_train_epoch(epoch, loader, iters, rollout_engine, ref_model, reward_mod
 
         rewards = calculate_rewards(prompts, completions, reward_model).to(args.device)  # [B*num_gen]
 
-        model_unwrapped = model.module if isinstance(model, DistributedDataParallel) else model
         with autocast_ctx:
-            res = model_unwrapped(outputs, attention_mask=full_mask)
+            # 反向传播必须经过 DDP 包装后的模块：直接调用 .module 会跳过
+            # DDP 的 prepare_for_backward，梯度不会 all-reduce，各卡静默发散。
+            res = model(outputs, attention_mask=full_mask)
             aux_loss = res.aux_loss if lm_config.use_moe else torch.tensor(0.0, device=args.device)
             per_token_logps = F.log_softmax(res.logits[:, :-1, :], dim=-1).gather(2, outputs[:, 1:].unsqueeze(-1)).squeeze(-1).gather(1, logp_pos)
 

trainer/train_ppo.py

@@ -159,8 +159,6 @@ def ppo_train_epoch(epoch, loader, iters, rollout_engine, ref_model, actor_sched
         clipfrac_sum = 0.0
         aux_loss_sum = 0.0
         log_count = 0
-        actor_unwrapped = actor_model.module if isinstance(actor_model, DistributedDataParallel) else actor_model
-        critic_unwrapped = critic_model.module if isinstance(critic_model, DistributedDataParallel) else critic_model
         for ppo_epoch in range(args.ppo_update_iters):
             if stop_ppo:
                 break
@@ -168,11 +166,13 @@ def ppo_train_epoch(epoch, loader, iters, rollout_engine, ref_model, actor_sched
             for i in range(0, B, mb_size):
                 inds = b_inds[i:i + mb_size]
                 
-                mb_values_seq = critic_unwrapped(input_ids=gen_out[inds], attention_mask=full_mask[inds])
+                # 反向传播必须经过 DDP 包装后的模块：直接调用 .module 会跳过
+                # DDP 的 prepare_for_backward，梯度不会 all-reduce，各卡静默发散。
+                mb_values_seq = critic_model(input_ids=gen_out[inds], attention_mask=full_mask[inds])
                 mb_resp_values = mb_values_seq.gather(1, logp_pos[inds])
 
                 with autocast_ctx:
-                    res = actor_unwrapped(input_ids=gen_out[inds], attention_mask=full_mask[inds])
+                    res = actor_model(input_ids=gen_out[inds], attention_mask=full_mask[inds])
                     aux_loss = res.aux_loss if lm_config.use_moe else torch.tensor(0.0, device=args.device)
                     # 在 autocast 内计算 log_softmax，避免直接对 fp16/bf16 logits
                     # 计算造成额外数值偏差。
@@ -191,7 +191,7 @@ def ppo_train_epoch(epoch, loader, iters, rollout_engine, ref_model, actor_sched
                                f"ratio_max={torch.exp(_lrv).max().item():.6f} "
                                f"ratio_min={torch.exp(_lrv).min().item():.6f} "
                                f"dropout={getattr(lm_config, 'dropout', None)} "
-                               f"training={actor_unwrapped.training}")
+                               f"training={actor_model.training}")
                 approx_kl = (0.5 * (log_ratio ** 2) * resp_policy_mask[inds]).sum() / resp_policy_mask[inds].sum().clamp(min=1)
                 
                 # 同步各卡的 approx_kl，防止某卡 break 而其它卡继续导致 DDP 死锁

trainer/trainer_utils.py

@@ -5,6 +5,8 @@ import os
 import sys
 __package__ = "trainer"
 sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__), '..')))
+import ast
+import operator
 import random
 import math
 import numpy as np
@@ -175,3 +177,52 @@ class LMForRewardModel:
         ]
         score = self.model.get_score(self.tokenizer, eval_messages)
         return max(min(score, 3.0), -3.0)
+
+
+# ===== 数学表达式安全求值（替代 eval，避免执行模型生成的任意代码） =====
+_MATH_FUNCS = {name: getattr(math, name) for name in (
+    'sqrt', 'pow', 'exp', 'log', 'log2', 'log10', 'sin', 'cos', 'tan', 'asin', 'acos', 'atan', 'atan2',
+    'sinh', 'cosh', 'tanh', 'floor', 'ceil', 'trunc', 'fabs', 'fmod', 'hypot', 'gcd', 'degrees', 'radians'
+)}
+_MATH_CONSTS = {'pi': math.pi, 'e': math.e, 'tau': math.tau}
+_BIN_OPS = {ast.Add: operator.add, ast.Sub: operator.sub, ast.Mult: operator.mul, ast.Div: operator.truediv,
+            ast.FloorDiv: operator.floordiv, ast.Mod: operator.mod, ast.Pow: operator.pow}
+_UNARY_OPS = {ast.UAdd: operator.pos, ast.USub: operator.neg}
+_MAX_EXPR_LEN, _MAX_EXP, _MAX_DIGITS = 512, 1e4, 100  # 表达式长度 / 指数 / 结果位数上限
+
+
+def _check_pow(base, exp):  # 防止 9**9**9 这类表达式把进程算死
+    if abs(exp) > _MAX_EXP or (abs(base) > 1 and abs(exp) * math.log10(abs(base)) > _MAX_DIGITS):
+        raise ValueError('幂运算结果过大')
+
+
+def _resolve_name(node):  # 同时兼容 sqrt(4) 与 math.sqrt(4) 两种写法
+    if isinstance(node, ast.Name): return node.id
+    if isinstance(node, ast.Attribute) and isinstance(node.value, ast.Name) and node.value.id == 'math': return node.attr
+    return None
+
+
+def _eval_ast(node):
+    if isinstance(node, ast.Constant) and isinstance(node.value, (int, float)) and not isinstance(node.value, bool):
+        return node.value
+    if isinstance(node, (ast.Name, ast.Attribute)) and _resolve_name(node) in _MATH_CONSTS:
+        return _MATH_CONSTS[_resolve_name(node)]
+    if isinstance(node, ast.UnaryOp) and type(node.op) in _UNARY_OPS:
+        return _UNARY_OPS[type(node.op)](_eval_ast(node.operand))
+    if isinstance(node, ast.BinOp) and type(node.op) in _BIN_OPS:
+        left, right = _eval_ast(node.left), _eval_ast(node.right)
+        if isinstance(node.op, ast.Pow): _check_pow(left, right)
+        return _BIN_OPS[type(node.op)](left, right)
+    if isinstance(node, ast.Call) and _resolve_name(node.func) in _MATH_FUNCS and not node.keywords:
+        name, args = _resolve_name(node.func), [_eval_ast(arg) for arg in node.args]
+        if name == 'pow' and len(args) == 2: _check_pow(*args)
+        return _MATH_FUNCS[name](*args)
+    raise ValueError(f'不支持的表达式语法: {type(node).__name__}')
+
+
+def safe_math_eval(expression):
+    """对模型生成的数学表达式求值：只放行算术运算与 math 白名单函数，不执行任意代码。"""
+    expr = str(expression).replace('^', '**').replace('×', '*').replace('÷', '/').replace('−', '-')
+    expr = expr.replace('²', '**2').replace('³', '**3').replace('（', '(').replace('）', ')').strip()
+    if not expr or len(expr) > _MAX_EXPR_LEN: raise ValueError('表达式为空或过长')
+    return _eval_ast(ast.parse(expr, mode='eval').body)