ChatGLM2-6B微调记录【2】

  • 模型推理测试
    • 微调前的chatglm2-6b模型
    • 运行python predict.py --mode glm2 --model_path chatglm2-6b/
    • 运行结果记录
bash 复制代码
/data/user23262833/.conda/envs/chatglm/lib/python3.8/site-packages/transformers/utils/generic.py:311: FutureWarning: `torch.utils._pytree._register_pytree_node` is deprecated. Please use `torch.utils._pytree.register_pytree_node` instead.
  torch.utils._pytree._register_pytree_node(
2024-11-08 17:21:16.440515: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 AVX512F AVX512_VNNI FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2024-11-08 17:21:16.562434: I tensorflow/core/util/port.cc:104] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2024-11-08 17:21:17.121689: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda-11.8/lib64:/usr/local/cuda-11.8/lib64:
2024-11-08 17:21:17.121775: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda-11.8/lib64:/usr/local/cuda-11.8/lib64:
2024-11-08 17:21:17.121784: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.
/data/user23262833/.conda/envs/chatglm/lib/python3.8/site-packages/transformers/utils/generic.py:311: FutureWarning: `torch.utils._pytree._register_pytree_node` is deprecated. Please use `torch.utils._pytree.register_pytree_node` instead.
  torch.utils._pytree._register_pytree_node(
Loading checkpoint shards:   0%|                                                                             | 0/7 [00:00<?, ?it/s]/data/user23262833/.conda/envs/chatglm/lib/python3.8/site-packages/transformers/modeling_utils.py:488: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  return torch.load(checkpoint_file, map_location=map_location)
Loading checkpoint shards: 100%|█████████████████████████████████████████████████████████████████████| 7/7 [00:10<00:00,  1.52s/it]
根据您提供的文本,我可以抽取出以下三个相关三元组:

1. 性能故障 -> 发动机水温高
2. 部件故障 -> 风扇始终是低速转动
3. 组成 -> 高速档不工作
4. 检测工具 -> 开空调

其中,第一个三元组表示发动机出现了性能故障,导致发动机水温高。第二个三元组表示风扇出现了部件故障,始终是低速转动。第三个三元组表示高速档不工作,可能是由于某个部件出现了问题。最后一个三元组表示开空调时出现了性能故障。
  • 微调后的chatglm2-6b模型
    • 运行python predict.py --mode glm2 --model_path output-glm2/epoch-1-step-90/
    • 运行结果记录
bash 复制代码
/data/user23262833/.conda/envs/chatglm/lib/python3.8/site-packages/transformers/utils/generic.py:311: FutureWarning: `torch.utils._pytree._register_pytree_node` is deprecated. Please use `torch.utils._pytree.register_pytree_node` instead.
  torch.utils._pytree._register_pytree_node(
2024-11-08 17:30:12.764824: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 AVX512F AVX512_VNNI FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2024-11-08 17:30:12.887176: I tensorflow/core/util/port.cc:104] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2024-11-08 17:30:13.455307: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda-11.8/lib64:/usr/local/cuda-11.8/lib64:
2024-11-08 17:30:13.455391: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda-11.8/lib64:/usr/local/cuda-11.8/lib64:
2024-11-08 17:30:13.455400: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.
/data/user23262833/.conda/envs/chatglm/lib/python3.8/site-packages/transformers/utils/generic.py:311: FutureWarning: `torch.utils._pytree._register_pytree_node` is deprecated. Please use `torch.utils._pytree.register_pytree_node` instead.
  torch.utils._pytree._register_pytree_node(
Loading checkpoint shards:   0%|                                                                             | 0/7 [00:00<?, ?it/s]/data/user23262833/.conda/envs/chatglm/lib/python3.8/site-packages/transformers/modeling_utils.py:488: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  return torch.load(checkpoint_file, map_location=map_location)
Loading checkpoint shards: 100%|█████████████████████████████████████████████████████████████████████| 7/7 [00:10<00:00,  1.55s/it]
/data/user23262833/.conda/envs/chatglm/lib/python3.8/site-packages/peft/utils/save_and_load.py:198: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  adapters_weights = torch.load(filename, map_location=torch.device(device))
发动机_部件故障_水温高
风扇_部件故障_始终是低速转动
高速档_部件故障_不工作
  • 运行python predict.py --mode glm2 --model_path output-glm2/epoch-2-step-180/
  • 运行结果记录
bash 复制代码
/data/user23262833/.conda/envs/chatglm/lib/python3.8/site-packages/transformers/utils/generic.py:311: FutureWarning: `torch.utils._pytree._register_pytree_node` is deprecated. Please use `torch.utils._pytree.register_pytree_node` instead.
  torch.utils._pytree._register_pytree_node(
2024-11-08 17:34:23.878094: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 AVX512F AVX512_VNNI FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2024-11-08 17:34:24.001155: I tensorflow/core/util/port.cc:104] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.
2024-11-08 17:34:24.560807: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda-11.8/lib64:/usr/local/cuda-11.8/lib64:
2024-11-08 17:34:24.560894: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda-11.8/lib64:/usr/local/cuda-11.8/lib64:
2024-11-08 17:34:24.560903: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.
/data/user23262833/.conda/envs/chatglm/lib/python3.8/site-packages/transformers/utils/generic.py:311: FutureWarning: `torch.utils._pytree._register_pytree_node` is deprecated. Please use `torch.utils._pytree.register_pytree_node` instead.
  torch.utils._pytree._register_pytree_node(
Loading checkpoint shards:   0%|                                                                             | 0/7 [00:00<?, ?it/s]/data/user23262833/.conda/envs/chatglm/lib/python3.8/site-packages/transformers/modeling_utils.py:488: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  return torch.load(checkpoint_file, map_location=map_location)
Loading checkpoint shards: 100%|█████████████████████████████████████████████████████████████████████| 7/7 [00:10<00:00,  1.55s/it]
/data/user23262833/.conda/envs/chatglm/lib/python3.8/site-packages/peft/utils/save_and_load.py:198: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  adapters_weights = torch.load(filename, map_location=torch.device(device))
发动机_部件故障_水温高
风扇_部件故障_低速转动
高速档_部件故障_不工作
相关推荐
zhuangzhao1191 小时前
第一性原理大模型课程 · 第 21 课:从 Prompt 到 Agent:四件套、错误累积与 ReAct 手算
大模型·react·ai agent·错误累积·大模型课程
智联视频超融合平台14 小时前
视频 AI 的误报之战,正在从调参数卷到改架构
人工智能·大模型·边缘计算·视频监控·安防
谙弆悕博士1 天前
多智能体架构详解:层级树状架构(分级主从型)
运维·架构·大模型·agent·多智能体·langgraph·层级树状
Chuck.li1 天前
AI 长篇创作工具选型:工作流对照与自测清单
人工智能·大模型·ai写作·网文作者·ai网文·豆包写作
谙弆悕博士1 天前
多智能体架构详解:黑板共享架构(共享存储交互型)
人工智能·架构·大模型·agent·多智能体·黑板架构·anggraph
学无止境小袁同学1 天前
OpenAI Agents SDK 模板操作手册(可直接操作)
大模型·openai·agent·sdk·大模型应用
bryant_meng2 天前
ChatGPT Desktop Model Evolution (2025): From GPT-4.1 to GPT-5.1
gpt·chatgpt·gpt4·gpt4o·gpt5
九卷2 天前
本体(Ontology)是什么?在 AI 中,本体的作用到底是什么?.md
大模型·agent
佛系1892 天前
从服务端日志精确分离搜索抓取、监控探测与真实流量的实操清单
ai·大模型·seo·geo·结构化数据·indexnow
MSTcheng.2 天前
【AI应用实战】用 FastAPI + SSE 做一个自己的 AI 聊天网页:流式输出、多轮对话与完整源码
人工智能·python·大模型·fastapi·sse·ai聊天助手