LLaMA-Factory微调（sft）ChatGLM3-6B保姆教程
LLaMA-Factory微调（sft）ChatGLM3-6B保姆教程
准备
1、下载
- 下载LLaMA-Factory
- 下载ChatGLM3-6B
- 下载ChatGLM3
- windows下载CUDA ToolKit 12.1 （本人是在windows进行训练的，显卡GTX 1660 Ti）
CUDA安装完毕后，通过指令nvidia-smi查看
2、PyCharm打开LLaMA-Factory项目
1、选择下载目录：E:\llm-train\LLaMA-Factory，并打开
2、创建新的python环境，这里使用conda创建一个python空环境，选择python3.10
3、安装依赖
参考LLaMA-Factory的依赖安装步骤
安装LLaMA-Factory依赖
(llm) PS E:\llm-train\LLaMA-Factory> pwdPath
----
E:\llm-train\LLaMA-Factory(llm) PS E:\llm-train\LLaMA-Factory> pip install -r requirements.txt
Looking in indexes: https://pypi.tuna.tsinghua.edu.cn/simple
Collecting torch>=1.13.1 (from -r requirements.txt (line 1))Downloading https://pypi.tuna.tsinghua.edu.cn/packages/3a/81/684d99e536b20e869a7c1222cf1dd233311fb05d3628e9570992bfb65760/torch-2.2.2-cp310-cp310-win_amd64.whl (198.6 MB)━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 198.6/198.6 MB 5.0 MB/s eta 0:00:00
Collecting transformers>=4.37.2 (from -r requirements.txt (line 2))Downloading https://pypi.tuna.tsinghua.edu.cn/packages/e2/52/02271ef16713abea41bab736dfc2dbee75e5e3512cf7441e233976211ba5/transformers-4.39.2-py3-none-any.whl (8.8 MB)━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 8.8/8.8 MB 4.0 MB/s eta 0:00:00
Collecting datasets>=2.14.3 (from -r requirements.txt (line 3))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/95/fc/661a7f06e8b7d48fcbd3f55423b7ff1ac3ce59526f146fda87a1e1788ee4/datasets-2.18.0-py3-none-any.whl (510 kB)
Collecting accelerate>=0.27.2 (from -r requirements.txt (line 4))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/a0/11/9bfcf765e71a2c84bbf715719ba520aeacb2ad84113f14803ff1947ddf69/accelerate-0.28.0-py3-none-any.whl (290 kB)
Collecting peft>=0.10.0 (from -r requirements.txt (line 5))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/b4/5d/758c00ba637bc850f35fff7fad442c470ac3d606fe586d881b0bed7ef5a5/peft-0.10.0-py3-none-any.whl (199 kB)
Collecting trl>=0.8.1 (from -r requirements.txt (line 6))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/89/12/bdad48d9eb5d2f3e71fdfaa3d634737a79375be7caa927c6e95aa05f9d4a/trl-0.8.1-py3-none-any.whl (225 kB)
Collecting gradio<4.0.0,>=3.38.0 (from -r requirements.txt (line 7))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/bd/ea/ca6506e4da9b5338da3bfdd6115dc1c90ffd58c1ec50ca2792b84a7b4bdb/gradio-3.50.2-py3-none-any.whl (20.3 MB)
Collecting scipy (from -r requirements.txt (line 8))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/fd/a7/5f829b100d208c85163aecba93faf01d088d944fc91585338751d812f1e4/scipy-1.12.0-cp310-cp310-win_amd64.whl (46.2 MB)
Collecting einops (from -r requirements.txt (line 9))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/29/0b/2d1c0ebfd092e25935b86509a9a817159212d82aa43d7fb07eca4eeff2c2/einops-0.7.0-py3-none-any.whl (44 kB)
Collecting sentencepiece (from -r requirements.txt (line 10))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/85/f4/4ef1a6e0e9dbd8a60780a91df8b7452ada14cfaa0e17b3b8dfa42cecae18/sentencepiece-0.2.0-cp310-cp310-win_amd64.whl (991 kB)
Collecting protobuf (from -r requirements.txt (line 11))Downloading https://pypi.tuna.tsinghua.edu.cn/packages/8d/83/d70cb6fedb1f38318af01f0035f2103732630af0ca323c0198122b49323b/protobuf-5.26.1-cp310-abi3-win_amd64.whl (420 kB)━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 420.9/420.9 kB 3.7 MB/s eta 0:00:00
Collecting uvicorn (from -r requirements.txt (line 12))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/73/f5/cbb16fcbe277c1e0b8b3ddd188f2df0e0947f545c49119b589643632d156/uvicorn-0.29.0-py3-none-any.whl (60 kB)
Collecting pydantic (from -r requirements.txt (line 13))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/e5/f3/8296f550276194a58c5500d55b19a27ae0a5a3a51ffef66710c58544b32d/pydantic-2.6.4-py3-none-any.whl (394 kB)
Collecting fastapi (from -r requirements.txt (line 14))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/f0/f7/ea860cb8aa18e326f411e32ab537424690a53db20de6bad73d70da611fae/fastapi-0.110.0-py3-none-any.whl (92 kB)
Collecting sse-starlette (from -r requirements.txt (line 15))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/50/04/b1569f95ce91fbfe35e5e6b3b11e8373ffe4fa81e998818c8c995a873896/sse_starlette-2.0.0-py3-none-any.whl (9.0 kB)
Collecting matplotlib (from -r requirements.txt (line 16))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/e4/43/fd3cd5989d6b592af1c2e4f37bf887f74b790f10b568b2497fe874a67fc7/matplotlib-3.8.3-cp310-cp310-win_amd64.whl (7.6 MB)
Collecting fire (from -r requirements.txt (line 17))Using cached fire-0.6.0-py2.py3-none-any.whl
Collecting galore-torch (from -r requirements.txt (line 18))Downloading https://pypi.tuna.tsinghua.edu.cn/packages/2b/b9/e9e88f989c62edefaa45df07198ce280ac0d373f5e2842686c6ece2ddb1e/galore_torch-1.0-py3-none-any.whl (13 kB)
Collecting filelock (from torch>=1.13.1->-r requirements.txt (line 1))Downloading https://pypi.tuna.tsinghua.edu.cn/packages/8b/69/acdf492db27dea7be5c63053230130e0574fd8a376de3555d5f8bbc3d3ad/filelock-3.13.3-py3-none-any.whl (11 kB)
Collecting typing-extensions>=4.8.0 (from torch>=1.13.1->-r requirements.txt (line 1))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/f9/de/dc04a3ea60b22624b51c703a84bbe0184abcd1d0b9bc8074b5d6b7ab90bb/typing_extensions-4.10.0-py3-none-any.whl (33 kB)
Collecting sympy (from torch>=1.13.1->-r requirements.txt (line 1))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/d2/05/e6600db80270777c4a64238a98d442f0fd07cc8915be2a1c16da7f2b9e74/sympy-1.12-py3-none-any.whl (5.7 MB)
Collecting networkx (from torch>=1.13.1->-r requirements.txt (line 1))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/d5/f0/8fbc882ca80cf077f1b246c0e3c3465f7f415439bdea6b899f6b19f61f70/networkx-3.2.1-py3-none-any.whl (1.6 MB)
Collecting jinja2 (from torch>=1.13.1->-r requirements.txt (line 1))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/30/6d/6de6be2d02603ab56e72997708809e8a5b0fbfee080735109b40a3564843/Jinja2-3.1.3-py3-none-any.whl (133 kB)
Collecting fsspec (from torch>=1.13.1->-r requirements.txt (line 1))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/93/6d/66d48b03460768f523da62a57a7e14e5e95fdf339d79e996ce3cecda2cdb/fsspec-2024.3.1-py3-none-any.whl (171 kB)
Collecting huggingface-hub<1.0,>=0.19.3 (from transformers>=4.37.2->-r requirements.txt (line 2))Downloading https://pypi.tuna.tsinghua.edu.cn/packages/b6/51/d418bb7bb9e32845d3cb4d526012ddf1fea6bb3d55b6a1880698e4b7f19f/huggingface_hub-0.22.1-py3-none-any.whl (388 kB)━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 388.6/388.6 kB 967.7 kB/s eta 0:00:00
Collecting numpy>=1.17 (from transformers>=4.37.2->-r requirements.txt (line 2))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/19/77/538f202862b9183f54108557bfda67e17603fc560c384559e769321c9d92/numpy-1.26.4-cp310-cp310-win_amd64.whl (15.8 MB)
Collecting packaging>=20.0 (from transformers>=4.37.2->-r requirements.txt (line 2))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/49/df/1fceb2f8900f8639e278b056416d49134fb8d84c5942ffaa01ad34782422/packaging-24.0-py3-none-any.whl (53 kB)
Collecting pyyaml>=5.1 (from transformers>=4.37.2->-r requirements.txt (line 2))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/24/97/9b59b43431f98d01806b288532da38099cc6f2fea0f3d712e21e269c0279/PyYAML-6.0.1-cp310-cp310-win_amd64.whl (145 kB)
Collecting regex!=2019.12.17 (from transformers>=4.37.2->-r requirements.txt (line 2))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/83/eb/144d2db5cf2ac3989d0ea4273040218d68bd67422133548da47043423594/regex-2023.12.25-cp310-cp310-win_amd64.whl (269 kB)
Collecting requests (from transformers>=4.37.2->-r requirements.txt (line 2))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/70/8e/0e2d847013cb52cd35b38c009bb167a1a26b2ce6cd6965bf26b47bc0bf44/requests-2.31.0-py3-none-any.whl (62 kB)
Collecting tokenizers<0.19,>=0.14 (from transformers>=4.37.2->-r requirements.txt (line 2))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/c9/87/0bf37626c5f1ea2462e0398be88c287f3d40c696c255ba478bf525bdc852/tokenizers-0.15.2-cp310-none-win_amd64.whl (2.2 MB)
Collecting safetensors>=0.4.1 (from transformers>=4.37.2->-r requirements.txt (line 2))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/58/48/8d228afbcb4a48d5a9f40eca65ab250703de7fd802f5b5148e5fb1edfe6e/safetensors-0.4.2-cp310-none-win_amd64.whl (269 kB)
Collecting tqdm>=4.27 (from transformers>=4.37.2->-r requirements.txt (line 2))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/2a/14/e75e52d521442e2fcc9f1df3c5e456aead034203d4797867980de558ab34/tqdm-4.66.2-py3-none-any.whl (78 kB)
Collecting pyarrow>=12.0.0 (from datasets>=2.14.3->-r requirements.txt (line 3))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/ec/85/abca962d99950aad803bd755baf020a8183ca3be1319bb205f52bbbcce16/pyarrow-15.0.2-cp310-cp310-win_amd64.whl (24.8 MB)
Collecting pyarrow-hotfix (from datasets>=2.14.3->-r requirements.txt (line 3))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/e4/f4/9ec2222f5f5f8ea04f66f184caafd991a39c8782e31f5b0266f101cb68ca/pyarrow_hotfix-0.6-py3-none-any.whl (7.9 kB)
Collecting dill<0.3.9,>=0.3.0 (from datasets>=2.14.3->-r requirements.txt (line 3))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/c9/7a/cef76fd8438a42f96db64ddaa85280485a9c395e7df3db8158cfec1eee34/dill-0.3.8-py3-none-any.whl (116 kB)
Collecting pandas (from datasets>=2.14.3->-r requirements.txt (line 3))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/93/26/2a695303a4a3194014dca7cb5d5ce08f0d2c6baa344fb5f562c642e77b2b/pandas-2.2.1-cp310-cp310-win_amd64.whl (11.6 MB)
Collecting xxhash (from datasets>=2.14.3->-r requirements.txt (line 3))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/a5/b0/2950f76c07e467586d01d9a6cdd4ac668c13d20de5fde49af5364f513e54/xxhash-3.4.1-cp310-cp310-win_amd64.whl (29 kB)
Collecting multiprocess (from datasets>=2.14.3->-r requirements.txt (line 3))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/bc/f7/7ec7fddc92e50714ea3745631f79bd9c96424cb2702632521028e57d3a36/multiprocess-0.70.16-py310-none-any.whl (134 kB)
Collecting fsspec (from torch>=1.13.1->-r requirements.txt (line 1))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/ad/30/2281c062222dc39328843bd1ddd30ff3005ef8e30b2fd09c4d2792766061/fsspec-2024.2.0-py3-none-any.whl (170 kB)
Collecting aiohttp (from datasets>=2.14.3->-r requirements.txt (line 3))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/54/2d/746a93808c2d9d247b554155a1576bf7ed0e0029e8cfb667a9007e2d53b0/aiohttp-3.9.3-cp310-cp310-win_amd64.whl (365 kB)
Collecting psutil (from accelerate>=0.27.2->-r requirements.txt (line 4))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/93/52/3e39d26feae7df0aa0fd510b14012c3678b36ed068f7d78b8d8784d61f0e/psutil-5.9.8-cp37-abi3-win_amd64.whl (255 kB)
Collecting tyro>=0.5.11 (from trl>=0.8.1->-r requirements.txt (line 6))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/e3/0e/c421c916bc8a48a861197bda6554588f8281db3e3533083f0d4d49b80ddb/tyro-0.7.3-py3-none-any.whl (79 kB)
Collecting aiofiles<24.0,>=22.0 (from gradio<4.0.0,>=3.38.0->-r requirements.txt (line 7))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/c5/19/5af6804c4cc0fed83f47bff6e413a98a36618e7d40185cd36e69737f3b0e/aiofiles-23.2.1-py3-none-any.whl (15 kB)
Collecting altair<6.0,>=4.2.0 (from gradio<4.0.0,>=3.38.0->-r requirements.txt (line 7))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/c5/e4/7fcceef127badbb0d644d730d992410e4f3799b295c9964a172f92a469c7/altair-5.2.0-py3-none-any.whl (996 kB)
Collecting ffmpy (from gradio<4.0.0,>=3.38.0->-r requirements.txt (line 7))Using cached ffmpy-0.3.2-py3-none-any.whl
Collecting gradio-client==0.6.1 (from gradio<4.0.0,>=3.38.0->-r requirements.txt (line 7))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/7d/04/e1654ee28fb2686514ca8ae31af5e489403964d48764788f9a168e069c0f/gradio_client-0.6.1-py3-none-any.whl (299 kB)
Collecting httpx (from gradio<4.0.0,>=3.38.0->-r requirements.txt (line 7))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/41/7b/ddacf6dcebb42466abd03f368782142baa82e08fc0c1f8eaa05b4bae87d5/httpx-0.27.0-py3-none-any.whl (75 kB)
Collecting importlib-resources<7.0,>=1.3 (from gradio<4.0.0,>=3.38.0->-r requirements.txt (line 7))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/75/06/4df55e1b7b112d183f65db9503bff189e97179b256e1ea450a3c365241e0/importlib_resources-6.4.0-py3-none-any.whl (38 kB)
Collecting markupsafe~=2.0 (from gradio<4.0.0,>=3.38.0->-r requirements.txt (line 7))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/69/48/acbf292615c65f0604a0c6fc402ce6d8c991276e16c80c46a8f758fbd30c/MarkupSafe-2.1.5-cp310-cp310-win_amd64.whl (17 kB)
Collecting orjson~=3.0 (from gradio<4.0.0,>=3.38.0->-r requirements.txt (line 7))Downloading https://pypi.tuna.tsinghua.edu.cn/packages/ed/71/9b653e2ac43769385e89a30153c5374e07788c6d8fc44bb616de427a8aa2/orjson-3.10.0-cp310-none-win_amd64.whl (139 kB)━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 139.2/139.2 kB 2.1 MB/s eta 0:00:00
Collecting pillow<11.0,>=8.0 (from gradio<4.0.0,>=3.38.0->-r requirements.txt (line 7))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/ef/d8/f97270d25a003435e408e6d1e38d8eddc9b3e2c7b646719f4b3a5293685d/pillow-10.2.0-cp310-cp310-win_amd64.whl (2.6 MB)
Collecting pydub (from gradio<4.0.0,>=3.38.0->-r requirements.txt (line 7))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/a6/53/d78dc063216e62fc55f6b2eebb447f6a4b0a59f55c8406376f76bf959b08/pydub-0.25.1-py2.py3-none-any.whl (32 kB)
Collecting python-multipart (from gradio<4.0.0,>=3.38.0->-r requirements.txt (line 7))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/3d/47/444768600d9e0ebc82f8e347775d24aef8f6348cf00e9fa0e81910814e6d/python_multipart-0.0.9-py3-none-any.whl (22 kB)
Collecting semantic-version~=2.0 (from gradio<4.0.0,>=3.38.0->-r requirements.txt (line 7))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/6a/23/8146aad7d88f4fcb3a6218f41a60f6c2d4e3a72de72da1825dc7c8f7877c/semantic_version-2.10.0-py2.py3-none-any.whl (15 kB)
Collecting websockets<12.0,>=10.0 (from gradio<4.0.0,>=3.38.0->-r requirements.txt (line 7))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/98/a7/0ed69892981351e5acf88fac0ff4c801fabca2c3bdef9fca4c7d3fde8c53/websockets-11.0.3-cp310-cp310-win_amd64.whl (124 kB)
Collecting click>=7.0 (from uvicorn->-r requirements.txt (line 12))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/00/2e/d53fa4befbf2cfa713304affc7ca780ce4fc1fd8710527771b58311a3229/click-8.1.7-py3-none-any.whl (97 kB)
Collecting h11>=0.8 (from uvicorn->-r requirements.txt (line 12))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/95/04/ff642e65ad6b90db43e668d70ffb6736436c7ce41fcc549f4e9472234127/h11-0.14.0-py3-none-any.whl (58 kB)
Collecting annotated-types>=0.4.0 (from pydantic->-r requirements.txt (line 13))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/28/78/d31230046e58c207284c6b2c4e8d96e6d3cb4e52354721b944d3e1ee4aa5/annotated_types-0.6.0-py3-none-any.whl (12 kB)
Collecting pydantic-core==2.16.3 (from pydantic->-r requirements.txt (line 13))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/ec/e8/49d65816802781451af7e758bdf9ff9d976a6b3959e1aab843da9931e89f/pydantic_core-2.16.3-cp310-none-win_amd64.whl (1.9 MB)
Collecting starlette<0.37.0,>=0.36.3 (from fastapi->-r requirements.txt (line 14))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/eb/f7/372e3953b6e6fbfe0b70a1bb52612eae16e943f4288516480860fcd4ac41/starlette-0.36.3-py3-none-any.whl (71 kB)
Collecting anyio (from sse-starlette->-r requirements.txt (line 15))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/14/fd/2f20c40b45e4fb4324834aea24bd4afdf1143390242c0b33774da0e2e34f/anyio-4.3.0-py3-none-any.whl (85 kB)
Collecting contourpy>=1.0.1 (from matplotlib->-r requirements.txt (line 16))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/fd/7c/168f8343f33d861305e18c56901ef1bb675d3c7f977f435ec72751a71a54/contourpy-1.2.0-cp310-cp310-win_amd64.whl (186 kB)
Collecting cycler>=0.10 (from matplotlib->-r requirements.txt (line 16))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/e7/05/c19819d5e3d95294a6f5947fb9b9629efb316b96de511b418c53d245aae6/cycler-0.12.1-py3-none-any.whl (8.3 kB)
Collecting fonttools>=4.22.0 (from matplotlib->-r requirements.txt (line 16))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/e9/59/b95df4ad8c5657dddec2a682d105ed3d3bd8734b3b158f63340f4f76d983/fonttools-4.50.0-cp310-cp310-win_amd64.whl (2.2 MB)
Collecting kiwisolver>=1.3.1 (from matplotlib->-r requirements.txt (line 16))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/4a/a1/8a9c9be45c642fa12954855d8b3a02d9fd8551165a558835a19508fec2e6/kiwisolver-1.4.5-cp310-cp310-win_amd64.whl (56 kB)
Collecting pyparsing>=2.3.1 (from matplotlib->-r requirements.txt (line 16))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/9d/ea/6d76df31432a0e6fdf81681a895f009a4bb47b3c39036db3e1b528191d52/pyparsing-3.1.2-py3-none-any.whl (103 kB)
Collecting python-dateutil>=2.7 (from matplotlib->-r requirements.txt (line 16))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/ec/57/56b9bcc3c9c6a792fcbaf139543cee77261f3651ca9da0c93f5c1221264b/python_dateutil-2.9.0.post0-py2.py3-none-any.whl (229 kB)
Collecting six (from fire->-r requirements.txt (line 17))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/d9/5a/e7c31adbe875f2abbb91bd84cf2dc52d792b5a01506781dbcf25c91daf11/six-1.16.0-py2.py3-none-any.whl (11 kB)
Collecting termcolor (from fire->-r requirements.txt (line 17))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/d9/5f/8c716e47b3a50cbd7c146f45881e11d9414def768b7cd9c5e6650ec2a80a/termcolor-2.4.0-py3-none-any.whl (7.7 kB)
Collecting bitsandbytes (from galore-torch->-r requirements.txt (line 18))Downloading https://pypi.tuna.tsinghua.edu.cn/packages/db/5b/ce65d93d7bb60fb07e5253f8f4c2da79d0329386188374a25b0ace74aae4/bitsandbytes-0.43.0-py3-none-win_amd64.whl (101.6 MB)━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 101.6/101.6 MB 4.4 MB/s eta 0:00:00
Collecting jsonschema>=3.0 (from altair<6.0,>=4.2.0->gradio<4.0.0,>=3.38.0->-r requirements.txt (line 7))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/39/9d/b035d024c62c85f2e2d4806a59ca7b8520307f34e0932fbc8cc75fe7b2d9/jsonschema-4.21.1-py3-none-any.whl (85 kB)
Collecting toolz (from altair<6.0,>=4.2.0->gradio<4.0.0,>=3.38.0->-r requirements.txt (line 7))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/b7/8a/d82202c9f89eab30f9fc05380daae87d617e2ad11571ab23d7c13a29bb54/toolz-0.12.1-py3-none-any.whl (56 kB)
Requirement already satisfied: colorama in c:\users\administrator\appdata\roaming\python\python310\site-packages (from click>=7.0->uvicorn->-r requirements.txt (line 12)) (0.4.6)
Collecting aiosignal>=1.1.2 (from aiohttp->datasets>=2.14.3->-r requirements.txt (line 3))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/76/ac/a7305707cb852b7e16ff80eaf5692309bde30e2b1100a1fcacdc8f731d97/aiosignal-1.3.1-py3-none-any.whl (7.6 kB)
Collecting attrs>=17.3.0 (from aiohttp->datasets>=2.14.3->-r requirements.txt (line 3))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/e0/44/827b2a91a5816512fcaf3cc4ebc465ccd5d598c45cefa6703fcf4a79018f/attrs-23.2.0-py3-none-any.whl (60 kB)
Collecting frozenlist>=1.1.1 (from aiohttp->datasets>=2.14.3->-r requirements.txt (line 3))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/61/15/2b5d644d81282f00b61e54f7b00a96f9c40224107282efe4cd9d2bf1433a/frozenlist-1.4.1-cp310-cp310-win_amd64.whl (50 kB)
Collecting multidict<7.0,>=4.5 (from aiohttp->datasets>=2.14.3->-r requirements.txt (line 3))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/ef/3d/ba0dc18e96c5d83731c54129819d5892389e180f54ebb045c6124b2e8b87/multidict-6.0.5-cp310-cp310-win_amd64.whl (28 kB)
Collecting yarl<2.0,>=1.0 (from aiohttp->datasets>=2.14.3->-r requirements.txt (line 3))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/31/d4/2085272a5ccf87af74d4e02787c242c5d60367840a4637b2835565264302/yarl-1.9.4-cp310-cp310-win_amd64.whl (76 kB)
Collecting async-timeout<5.0,>=4.0 (from aiohttp->datasets>=2.14.3->-r requirements.txt (line 3))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/a7/fa/e01228c2938de91d47b307831c62ab9e4001e747789d0b05baf779a6488c/async_timeout-4.0.3-py3-none-any.whl (5.7 kB)
Collecting pytz>=2020.1 (from pandas->datasets>=2.14.3->-r requirements.txt (line 3))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/9c/3d/a121f284241f08268b21359bd425f7d4825cffc5ac5cd0e1b3d82ffd2b10/pytz-2024.1-py2.py3-none-any.whl (505 kB)
Collecting tzdata>=2022.7 (from pandas->datasets>=2.14.3->-r requirements.txt (line 3))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/65/58/f9c9e6be752e9fcb8b6a0ee9fb87e6e7a1f6bcab2cdc73f02bb7ba91ada0/tzdata-2024.1-py2.py3-none-any.whl (345 kB)
Collecting charset-normalizer<4,>=2 (from requests->transformers>=4.37.2->-r requirements.txt (line 2))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/a2/a0/4af29e22cb5942488cf45630cbdd7cefd908768e69bdd90280842e4e8529/charset_normalizer-3.3.2-cp310-cp310-win_amd64.whl (100 kB)
Collecting idna<4,>=2.5 (from requests->transformers>=4.37.2->-r requirements.txt (line 2))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/c2/e7/a82b05cf63a603df6e68d59ae6a68bf5064484a0718ea5033660af4b54a9/idna-3.6-py3-none-any.whl (61 kB)
Collecting urllib3<3,>=1.21.1 (from requests->transformers>=4.37.2->-r requirements.txt (line 2))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/a2/73/a68704750a7679d0b6d3ad7aa8d4da8e14e151ae82e6fee774e6e0d05ec8/urllib3-2.2.1-py3-none-any.whl (121 kB)
Collecting certifi>=2017.4.17 (from requests->transformers>=4.37.2->-r requirements.txt (line 2))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/ba/06/a07f096c664aeb9f01624f858c3add0a4e913d6c96257acb4fce61e7de14/certifi-2024.2.2-py3-none-any.whl (163 kB)
Collecting sniffio>=1.1 (from anyio->sse-starlette->-r requirements.txt (line 15))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/e9/44/75a9c9421471a6c4805dbf2356f7c181a29c1879239abab1ea2cc8f38b40/sniffio-1.3.1-py3-none-any.whl (10 kB)
Collecting exceptiongroup>=1.0.2 (from anyio->sse-starlette->-r requirements.txt (line 15))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/b8/9a/5028fd52db10e600f1c4674441b968cf2ea4959085bfb5b99fb1250e5f68/exceptiongroup-1.2.0-py3-none-any.whl (16 kB)
Collecting docstring-parser>=0.14.1 (from tyro>=0.5.11->trl>=0.8.1->-r requirements.txt (line 6))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/d5/7c/e9fcff7623954d86bdc17782036cbf715ecab1bec4847c008557affe1ca8/docstring_parser-0.16-py3-none-any.whl (36 kB)
Collecting rich>=11.1.0 (from tyro>=0.5.11->trl>=0.8.1->-r requirements.txt (line 6))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/87/67/a37f6214d0e9fe57f6ae54b2956d550ca8365857f42a1ce0392bb21d9410/rich-13.7.1-py3-none-any.whl (240 kB)
Collecting shtab>=1.5.6 (from tyro>=0.5.11->trl>=0.8.1->-r requirements.txt (line 6))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/e2/d1/a1d3189e7873408b9dc396aef0d7926c198b0df2aa3ddb5b539d3e89a70f/shtab-1.7.1-py3-none-any.whl (14 kB)
Collecting httpcore==1.* (from httpx->gradio<4.0.0,>=3.38.0->-r requirements.txt (line 7))Downloading https://pypi.tuna.tsinghua.edu.cn/packages/78/d4/e5d7e4f2174f8a4d63c8897d79eb8fe2503f7ecc03282fee1fa2719c2704/httpcore-1.0.5-py3-none-any.whl (77 kB)━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 77.9/77.9 kB 240.9 kB/s eta 0:00:00
Collecting mpmath>=0.19 (from sympy->torch>=1.13.1->-r requirements.txt (line 1))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/43/e3/7d92a15f894aa0c9c4b49b8ee9ac9850d6e63b03c9c32c0367a13ae62209/mpmath-1.3.0-py3-none-any.whl (536 kB)
Collecting jsonschema-specifications>=2023.03.6 (from jsonschema>=3.0->altair<6.0,>=4.2.0->gradio<4.0.0,>=3.38.0->-r requirements.txt (line 7))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/ee/07/44bd408781594c4d0a027666ef27fab1e441b109dc3b76b4f836f8fd04fe/jsonschema_specifications-2023.12.1-py3-none-any.whl (18 kB)
Collecting referencing>=0.28.4 (from jsonschema>=3.0->altair<6.0,>=4.2.0->gradio<4.0.0,>=3.38.0->-r requirements.txt (line 7))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/42/8e/ae1de7b12223986e949bdb886c004de7c304b6fa94de5b87c926c1099656/referencing-0.34.0-py3-none-any.whl (26 kB)
Collecting rpds-py>=0.7.1 (from jsonschema>=3.0->altair<6.0,>=4.2.0->gradio<4.0.0,>=3.38.0->-r requirements.txt (line 7))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/3f/17/abab0fc0ec544b89aa3de13fd9267721154a76a783279104c358e58341e2/rpds_py-0.18.0-cp310-none-win_amd64.whl (206 kB)
Collecting markdown-it-py>=2.2.0 (from rich>=11.1.0->tyro>=0.5.11->trl>=0.8.1->-r requirements.txt (line 6))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/42/d7/1ec15b46af6af88f19b8e5ffea08fa375d433c998b8a7639e76935c14f1f/markdown_it_py-3.0.0-py3-none-any.whl (87 kB)
Collecting pygments<3.0.0,>=2.13.0 (from rich>=11.1.0->tyro>=0.5.11->trl>=0.8.1->-r requirements.txt (line 6))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/97/9c/372fef8377a6e340b1704768d20daaded98bf13282b5327beb2e2fe2c7ef/pygments-2.17.2-py3-none-any.whl (1.2 MB)
Collecting mdurl~=0.1 (from markdown-it-py>=2.2.0->rich>=11.1.0->tyro>=0.5.11->trl>=0.8.1->-r requirements.txt (line 6))Using cached https://pypi.tuna.tsinghua.edu.cn/packages/b3/38/89ba8ad64ae25be8de66a6d463314cf1eb366222074cfda9ee839c56a4b4/mdurl-0.1.2-py3-none-any.whl (10.0 kB)
Installing collected packages: sentencepiece, pytz, pydub, mpmath, ffmpy, xxhash, websockets, urllib3, tzdata, typing-extensions, tqdm, toolz, termcolor, sympy, sniffio, six, shtab, semantic-version, safetensors, rpds-py, re
gex, pyyaml, python-multipart, pyparsing, pygments, pyarrow-hotfix, psutil, protobuf, pillow, packaging, orjson, numpy, networkx, multidict, mdurl, markupsafe, kiwisolver, importlib-resources, idna, h11, fsspec, frozenlist,
fonttools, filelock, exceptiongroup, einops, docstring-parser, dill, cycler, click, charset-normalizer, certifi, attrs, async-timeout, annotated-types, aiofiles, yarl, uvicorn, scipy, requests, referencing, python-dateutil,
pydantic-core, pyarrow, multiprocess, markdown-it-py, jinja2, httpcore, fire, contourpy, anyio, aiosignal, torch, starlette, rich, pydantic, pandas, matplotlib, jsonschema-specifications, huggingface-hub, httpx, aiohttp, tyr
o, tokenizers, sse-starlette, jsonschema, gradio-client, fastapi, bitsandbytes, accelerate, transformers, datasets, altair, trl, peft, gradio, galore-torch
ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
pylint 2.17.1 requires tomlkit>=0.10.1, which is not installed.
Successfully installed accelerate-0.28.0 aiofiles-23.2.1 aiohttp-3.9.3 aiosignal-1.3.1 altair-5.2.0 annotated-types-0.6.0 anyio-4.3.0 async-timeout-4.0.3 attrs-23.2.0 bitsandbytes-0.43.0 certifi-2024.2.2 charset-normalizer-3
.3.2 click-8.1.7 contourpy-1.2.0 cycler-0.12.1 datasets-2.18.0 dill-0.3.8 docstring-parser-0.16 einops-0.7.0 exceptiongroup-1.2.0 fastapi-0.110.0 ffmpy-0.3.2 filelock-3.13.3 fire-0.6.0 fonttools-4.50.0 frozenlist-1.4.1 fsspe
c-2024.2.0 galore-torch-1.0 gradio-3.50.2 gradio-client-0.6.1 h11-0.14.0 httpcore-1.0.5 httpx-0.27.0 huggingface-hub-0.22.1 idna-3.6 importlib-resources-6.4.0 jinja2-3.1.3 jsonschema-4.21.1 jsonschema-specifications-2023.12.
1 kiwisolver-1.4.5 markdown-it-py-3.0.0 markupsafe-2.1.5 matplotlib-3.8.3 mdurl-0.1.2 mpmath-1.3.0 multidict-6.0.5 multiprocess-0.70.16 networkx-3.2.1 numpy-1.26.4 orjson-3.10.0 packaging-24.0 pandas-2.2.1 peft-0.10.0 pillow
-10.2.0 protobuf-5.26.1 psutil-5.9.8 pyarrow-15.0.2 pyarrow-hotfix-0.6 pydantic-2.6.4 pydantic-core-2.16.3 pydub-0.25.1 pygments-2.17.2 pyparsing-3.1.2 python-dateutil-2.9.0.post0 python-multipart-0.0.9 pytz-2024.1 pyyaml-6.
0.1 referencing-0.34.0 regex-2023.12.25 requests-2.31.0 rich-13.7.1 rpds-py-0.18.0 safetensors-0.4.2 scipy-1.12.0 semantic-version-2.10.0 sentencepiece-0.2.0 shtab-1.7.1 six-1.16.0 sniffio-1.3.1 sse-starlette-2.0.0 starlette
-0.36.3 sympy-1.12 termcolor-2.4.0 tokenizers-0.15.2 toolz-0.12.1 torch-2.2.2 tqdm-4.66.2 transformers-4.39.2 trl-0.8.1 typing-extensions-4.10.0 tyro-0.7.3 tzdata-2024.1 urllib3-2.2.1 uvicorn-0.29.0 websockets-11.0.3 xxhash-
3.4.1 yarl-1.9.4
(llm) PS E:\llm-train\LLaMA-Factory> 为了使用cuda（GPU）加速训练和推理，不安装则使用cpu。安装torch的cuda版本：pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121
(llm) PS E:\llm-train\LLaMA-Factory> pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121
Looking in indexes: https://download.pytorch.org/whl/cu121
Requirement already satisfied: torch in c:\users\administrator\miniconda3\envs\llm\lib\site-packages (2.2.2)
Collecting torchvisionUsing cached https://download.pytorch.org/whl/cu121/torchvision-0.17.2%2Bcu121-cp310-cp310-win_amd64.whl (5.7 MB)
Collecting torchaudioUsing cached https://download.pytorch.org/whl/cu121/torchaudio-2.2.2%2Bcu121-cp310-cp310-win_amd64.whl (4.1 MB)
Requirement already satisfied: filelock in c:\users\administrator\miniconda3\envs\llm\lib\site-packages (from torch) (3.13.3)
Requirement already satisfied: typing-extensions>=4.8.0 in c:\users\administrator\miniconda3\envs\llm\lib\site-packages (from torch) (4.10.0)
Requirement already satisfied: sympy in c:\users\administrator\miniconda3\envs\llm\lib\site-packages (from torch) (1.12)
Requirement already satisfied: networkx in c:\users\administrator\miniconda3\envs\llm\lib\site-packages (from torch) (3.2.1)
Requirement already satisfied: jinja2 in c:\users\administrator\miniconda3\envs\llm\lib\site-packages (from torch) (3.1.3)
Requirement already satisfied: fsspec in c:\users\administrator\miniconda3\envs\llm\lib\site-packages (from torch) (2024.2.0)
Requirement already satisfied: numpy in c:\users\administrator\miniconda3\envs\llm\lib\site-packages (from torchvision) (1.26.4)
Collecting torchUsing cached https://download.pytorch.org/whl/cu121/torch-2.2.2%2Bcu121-cp310-cp310-win_amd64.whl (2454.8 MB)
Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in c:\users\administrator\miniconda3\envs\llm\lib\site-packages (from torchvision) (10.2.0)
Requirement already satisfied: MarkupSafe>=2.0 in c:\users\administrator\miniconda3\envs\llm\lib\site-packages (from jinja2->torch) (2.1.5)
Requirement already satisfied: mpmath>=0.19 in c:\users\administrator\miniconda3\envs\llm\lib\site-packages (from sympy->torch) (1.3.0)
Installing collected packages: torch, torchvision, torchaudioAttempting uninstall: torchFound existing installation: torch 2.2.2Uninstalling torch-2.2.2:Successfully uninstalled torch-2.2.2
Successfully installed torch-2.2.2+cu121 torchaudio-2.2.2+cu121 torchvision-0.17.2+cu121
(llm) PS E:\llm-train\LLaMA-Factory> 准备数据集
1、创建LLaMA-Factory\data\chatglm3_zh.json文件，拷贝一下内容。作为训练测试数据
[{"instruction": "","input": "安妮","output": "女仆。 精灵族\n声音温柔娇媚，嗲音。\n年龄：26岁"},{"instruction": "","input": "奥利维亚","output": "元气少女，中气十足。\n活泼可爱，心直口快。\n年龄：16岁 矮人族"},{"instruction": "","input": "维京人甲-\n维京战士","output": "野蛮无礼的维京人。\n市井气息，有些无赖。\n年龄：38岁。"},{"instruction": "","input": "维京人乙-维京射手","output": "龙套角色。\n女性\n维京人。"},{"instruction": "","input": "英灵-维京战神","output": "维京人首领。\n王者风范，霸气十足。\n声如洪钟，气势如虹。\n年龄：55岁"},{"instruction": "","input": "魔武·凛冬刃","output": "魔武。\n物品NPC"},{"instruction": "","input": "战斗教官","output": "人类。\n粗犷的女性战士，中性范。\n年龄：35岁"},{"instruction": "","input": "雅妮拉","output": "青年女性，声音甜美。\n骄横，喜欢冷嘲热讽。\n\n主角团之一。"},{"instruction": "","input": "公墓守卫队长","output": "龙套角色。\n男性，骑士团战士。"},{"instruction": "","input": "邪恶魔法师/黑巫师","output": "人类。\n丧失理智的人类法师。\n癫狂，变态。\n\n年龄：33岁"},{"instruction": "","input": "魔物-赤炎妖","output": "魔物。\n嗜血恐怖的妖魔。"},{"instruction": "","input": "英灵-裁决骑士","output": "逝去的英雄之魂。\n女性。\n言辞铿锵有力，正义之士。"},{"instruction": "","input": "布鲁克","output": "人类贵族公子，商人。\n20岁。\n天真傲慢，蛮横自大，骄傲易怒。但实际上非常弱鸡，未涉世事。\n毫无社会常识的公子哥。\n"},{"instruction": "","input": "地精仆从","output": "布鲁克的仆人。\n地精。\n胆小怕事，说话结巴。"},{"instruction": "","input": "嚣张佣兵","output": "雷根佣兵团的佣兵，狗仗人势。"},{"instruction": "","input": "神秘女神","output": "女神，端庄温柔，持稳宁静。\n不紧不慢，慢条斯理。\n年龄：20岁"},{"instruction": "","input": "托托","output": "酒馆老板，人类\n八面玲珑，诙谐幽默。\n言谈举止进退有度，侃侃而谈。\n年龄：45岁"},{"instruction": "","input": "酒馆女仆-露娜","output": "温柔可人的女仆。"},{"instruction": "","input": "狡诈魔魂","output": "人类。-灵体\n阴险狡诈的腹黑女性。\n得势时飞扬跋扈，弱势时楚楚可怜。\n年龄：29岁"},{"instruction": "","input": "火元素","output": "元素生命。\n魔物。\n龙套角色。"},{"instruction": "","input": "商会管事/商队运输理事","output": "龙套角色，均为年轻女性。"},{"instruction": "","input": "安吉尔","output": "女骑士，中性声线。\n热诚坦率，刚正不阿。"},{"instruction": "","input": "托克","output": "龙套角色。\n地精商贩 地精种族\n年龄：30岁\n青年商贩，有一点怪物腔调。"},{"instruction": "","input": "王国守卫/铁卫禁军","output": "龙套角色，普通男性战士。"},{"instruction": "","input": "爱德华","output": "人类，狮心骑士团中的精英骑士\n青壮男性，中气十足，正义之士。\n年龄：35岁"},{"instruction": "","input": "骑士团长","output": "血族\n大家风范，中性风。\n雷厉风行，刚正不阿。\n年龄：38岁"},{"instruction": "","input": "主教","output": "人类\n圣光教会主教。\n传令官，声音清晰中正，官方强调。\n年龄：30岁"},{"instruction": "","input": "女王","output": "人类。\n年轻而阴郁的贵族少女。\n郁郁寡欢。\n\n年龄：15岁"},{"instruction": "","input": "艾伦","output": "人类佣兵。——剧情主角之一。\n20岁。\n青年才俊，隐藏身份-皇室王子。\n绝顶聪明、攻于算计、略显市侩、巧于辞令。\n言语之中，充满了少年英气，但偶尔会有些轻浮。\n\n人物参考：《雪中悍刀行》徐凤年\n"},{"instruction": "","input": "雅各布","output": "半神。\n狼人始祖。（中年男子）\n傲慢无礼，嚣张跋扈，却又不失大家风范。\n\n\n人物参考：网易阴阳师里的八岐大蛇。\n"},{"instruction": "","input": "佣兵首领威尔曼","output": "人类佣兵。\n35岁。\n中年男子，沉稳持重，波澜不惊。\n"},{"instruction": "","input": "食人魔","output": "食人怪物，半兽人腔调，残暴贪婪。\n年龄：150岁"},{"instruction": "","input": "兽人首领","output": "兽人首领。\n年龄未知。（相当于成年男性）\n蛮横残暴，对人类等弱小种族表示轻蔑和无视。"},{"instruction": "","input": "霍华德","output": "半神。\n血族始祖。（神魔大陆之中非常重要的种族）\n沉默寡言，沉稳持重，有一种成熟大叔的依赖感。\n\n例子：《秦时明月》中的盖聂。"},{"instruction": "","input": "凯尼","output": "骄横佣兵，满不在乎，阴险狡诈。"},{"instruction": "","input": "西蒙斯","output": "佣兵战士，中年男性，外强中干。"},{"instruction": "","input": "迪比斯","output": "魔法师佣兵，龙套角色。"},{"instruction": "","input": "兰希亚","output": "衰弱的女神，有气无力。\n气若游丝，声音迟缓。\n年龄：45岁"},{"instruction": "","input": "树灵龙","output": "古龙级别。\n\n可类比《霍比特人》里的恶龙史矛革"},{"instruction": "","input": "世界之母","output": "世界之母，声音沉稳而空灵，声音可带回音。\n\n可参考FF14中的水晶之母，海德林。"},{"instruction": "","input": "树灵精灵","output": "较弱可爱，声线可低幼一些，软萌音。\n楚楚可怜。"},{"instruction": "","input": "狮心骑士维达","output": "外表为中老年骑士。\n老年人特有的亲和力。\n深沉慈祥，智者形象。"},{"instruction": "","input": "亚杰拉","output": "龙套角色。\n初级佣兵。\n少女。\n战斗力低下。"},{"instruction": "","input": "查克·肖特奇","output": "龙套角色。\n初级佣兵。\n青年男性。\n战斗力低下。"},{"instruction": "","input": "布里克","output": "人族，中年男性。\n狮心高阶骑士。"},{"instruction": "","input": "雷切尔","output": "人族，中年女性。\n狮心高阶骑士。"},{"instruction": "","input": "戴安娜","output": "人族，年轻女性。\n狮心高阶骑士。"},{"instruction": "","input": "巴伦德","output": "矮人，中年男性。\n狮心高阶骑士。"},{"instruction": "","input": "龙十八","output": "禅国青年才俊。\n好风雅，外在散漫放浪。\n实则极有"},{"instruction": "","input": "禅国商贩","output": "龙套角色。\n外表看上去老实巴交，实则内心满是算计的中年男子。\n"},{"instruction": "","input": "琳达·安吉","output": "龙套角色。\n禅国年轻女子。"},{"instruction": "","input": "芙娜德","output": "龙套角色。\n禅国年轻女子。"},{"instruction": "","input": "闹事者","output": "龙套角色。\n禅国年轻男子，脾气暴躁，对商会有很深的怨念。"},{"instruction": "","input": "商会守卫","output": "龙套角色。\n中年男性\n禅国商会的守卫。"},{"instruction": "","input": "穷苦孩童","output": "龙套角色。\n少年男孩。\n生活潦倒，贫穷可怜。"},{"instruction": "","input": "龙神傀儡/龙神","output": "禅国幕后的统治者。\n实际上本身是利用死者捏造的傀儡，替真正的龙神-古龙青办事。\n为人腹黑阴险，却又气度非凡。"},{"instruction": "","input": "佛陀八","output": "寿命不可知。\n龙十八以及数代会长的老师。\n暗中与龙神勾结，获得不死之身，震慑禅国的朝纲。\n常年居住于观星阁。"},{"instruction": "","input": "古龙青","output": "龙神真身，在漫长的岁月当中，生命力之间衰弱，依靠着吞噬活人续命。\n从曾经禅国的守护者，成了禅国逐渐毁灭的黑洞。"},{"instruction": "","input": "祭天坛护院/祭天坛守卫","output": "龙套角色。\n禅国祭天坛的守护者。"}
]
2、编辑LLaMA-Factory\data\dataset_info.json，添加测试数据集到配置文件
..."chatglm3_zh": {"file_name": "chatglm3_zh.json"},...
训练微调
1、启动web版本的训练
(llm) PS E:\llm-train\LLaMA-Factory> set CUDA_VISIBLE_DEVICES=0
(llm) PS E:\llm-train\LLaMA-Factory> python src/train_web.py
Running on local URL: http://0.0.0.0:7860To create a public link, set `share=True` in `launch()`.
2、调整配置，浏览器打开：http://localhost:7860/
- 语言选择中文
- 模型名称：ChatGLM3-6B-Chat
- 前面从Hugging Face下载的ChatGLM3-6B模型本地路径
- 微调方法：lora
- 训练阶段：sft
- 数据集：上面新添加的测试数据集
- 训练轮数：200，因为数据量比较小，为了能表现效果，这里使用200轮
预览命令
CUDA_VISIBLE_DEVICES=0 python src/train_bash.py \--stage sft \--do_train True \--model_name_or_path E:\llm-train\chatglm3-6b \--finetuning_type lora \--template chatglm3 \--dataset_dir data \--dataset chatglm3_zh \--cutoff_len 1024 \--learning_rate 5e-05 \--num_train_epochs 200.0 \--max_samples 100000 \--per_device_train_batch_size 2 \--gradient_accumulation_steps 8 \--lr_scheduler_type cosine \--max_grad_norm 1.0 \--logging_steps 5 \--save_steps 100 \--warmup_steps 0 \--optim adamw_torch \--output_dir saves\ChatGLM3-6B-Chat\lora\train_2024-03-29-15-09-35 \--fp16 True \--lora_rank 8 \--lora_alpha 16 \--lora_dropout 0.1 \--lora_target query_key_value \--plot_loss True
3、开始训练
[INFO|parser.py:242] 2024-03-29 15:18:18,458 >> Process rank: 0, device: cuda:0, n_gpu: 1, distributed training: False, compute dtype: torch.float16[INFO|tokenization_utils_base.py:2082] 2024-03-29 15:18:19,291 >> loading file tokenizer.model[INFO|tokenization_utils_base.py:2082] 2024-03-29 15:18:19,291 >> loading file added_tokens.json[INFO|tokenization_utils_base.py:2082] 2024-03-29 15:18:19,291 >> loading file special_tokens_map.json[INFO|tokenization_utils_base.py:2082] 2024-03-29 15:18:19,291 >> loading file tokenizer_config.json[INFO|tokenization_utils_base.py:2082] 2024-03-29 15:18:19,291 >> loading file tokenizer.json[WARNING|tokenization_chatglm.py:164] 2024-03-29 15:18:19,334 >> Setting eos_token is not supported, use the default one.[WARNING|tokenization_chatglm.py:160] 2024-03-29 15:18:19,334 >> Setting pad_token is not supported, use the default one.[WARNING|tokenization_chatglm.py:156] 2024-03-29 15:18:19,334 >> Setting unk_token is not supported, use the default one.[INFO|template.py:371] 2024-03-29 15:18:19,609 >> Add <|user|>,<|observation|> to stop words.[INFO|template.py:378] 2024-03-29 15:18:19,609 >> Cannot add this chat template to tokenizer.[INFO|loader.py:33] 2024-03-29 15:18:19,610 >> Loading dataset chatglm3_zh.json...[WARNING|utils.py:31] 2024-03-29 15:18:19,610 >> Checksum failed: missing SHA-1 hash value in dataset_info.json.[INFO|configuration_utils.py:724] 2024-03-29 15:18:21,595 >> loading configuration file E:\llm-train\chatglm3-6b\config.json[INFO|configuration_utils.py:724] 2024-03-29 15:18:21,600 >> loading configuration file E:\llm-train\chatglm3-6b\config.json[INFO|configuration_utils.py:789] 2024-03-29 15:18:21,601 >> Model config ChatGLMConfig { "_name_or_path": "E:\llm-train\chatglm3-6b", "add_bias_linear": false, "add_qkv_bias": true, "apply_query_key_layer_scaling": true, "apply_residual_connection_post_layernorm": false, "architectures": [ "ChatGLMModel" ], "attention_dropout": 0.0, "attention_softmax_in_fp32": true, "auto_map": { "AutoConfig": "configuration_chatglm.ChatGLMConfig", "AutoModel": "modeling_chatglm.ChatGLMForConditionalGeneration", "AutoModelForCausalLM": "modeling_chatglm.ChatGLMForConditionalGeneration", "AutoModelForSeq2SeqLM": "modeling_chatglm.ChatGLMForConditionalGeneration", "AutoModelForSequenceClassification": "modeling_chatglm.ChatGLMForSequenceClassification" }, "bias_dropout_fusion": true, "classifier_dropout": null, "eos_token_id": 2, "ffn_hidden_size": 13696, "fp32_residual_connection": false, "hidden_dropout": 0.0, "hidden_size": 4096, "kv_channels": 128, "layernorm_epsilon": 1e-05, "model_type": "chatglm", "multi_query_attention": true, "multi_query_group_num": 2, "num_attention_heads": 32, "num_layers": 28, "original_rope": true, "pad_token_id": 0, "padded_vocab_size": 65024, "post_layer_norm": true, "pre_seq_len": null, "prefix_projection": false, "quantization_bit": 0, "rmsnorm": true, "seq_length": 8192, "tie_word_embeddings": false, "torch_dtype": "float16", "transformers_version": "4.39.2", "use_cache": true, "vocab_size": 65024 }[INFO|modeling_utils.py:3280] 2024-03-29 15:18:21,772 >> loading weights file E:\llm-train\chatglm3-6b\model.safetensors.index.json[INFO|modeling_utils.py:1417] 2024-03-29 15:18:21,774 >> Instantiating ChatGLMForConditionalGeneration model under default dtype torch.float16.[INFO|configuration_utils.py:928] 2024-03-29 15:18:21,775 >> Generate config GenerationConfig { "eos_token_id": 2, "pad_token_id": 0 }[INFO|modeling_utils.py:4024] 2024-03-29 15:18:37,990 >> All model checkpoint weights were used when initializing ChatGLMForConditionalGeneration.[INFO|modeling_utils.py:4032] 2024-03-29 15:18:37,990 >> All the weights of ChatGLMForConditionalGeneration were initialized from the model checkpoint at E:\llm-train\chatglm3-6b. If your task is similar to the task the model of the checkpoint was trained on, you can already use ChatGLMForConditionalGeneration for predictions without further training.[INFO|modeling_utils.py:3573] 2024-03-29 15:18:37,999 >> Generation config file not found, using a generation config created from the model config.[INFO|patcher.py:259] 2024-03-29 15:18:38,004 >> Gradient checkpointing enabled.[INFO|adapter.py:90] 2024-03-29 15:18:38,004 >> Fine-tuning method: LoRA[INFO|loader.py:126] 2024-03-29 15:18:38,190 >> trainable params: 1949696 || all params: 6245533696 || trainable%: 0.0312[INFO|trainer.py:607] 2024-03-29 15:18:38,204 >> Using auto half precision backend[INFO|trainer.py:1969] 2024-03-29 15:18:38,350 >> ***** Running training *****[INFO|trainer.py:1970] 2024-03-29 15:18:38,351 >> Num examples = 59[INFO|trainer.py:1971] 2024-03-29 15:18:38,351 >> Num Epochs = 200[INFO|trainer.py:1972] 2024-03-29 15:18:38,351 >> Instantaneous batch size per device = 2[INFO|trainer.py:1975] 2024-03-29 15:18:38,351 >> Total train batch size (w. parallel, distributed & accumulation) = 16[INFO|trainer.py:1976] 2024-03-29 15:18:38,351 >> Gradient Accumulation steps = 8[INFO|trainer.py:1977] 2024-03-29 15:18:38,352 >> Total optimization steps = 600[INFO|trainer.py:1978] 2024-03-29 15:18:38,354 >> Number of trainable parameters = 1,949,696
4、训练完毕
导出模型：需要设置训练的最后一个检查点。
(llm) PS E:\llm-train\LLaMA-Factory> python src/export_model.py --model_name_or_path E:\\llm-train\\chatglm3-6b --adapter_name_or_path E:\\llm-train\\LLaMA-Factory\\saves\\ChatGLM3-6B-Chat\\lora\\train_glm3\\checkpoint-200 -
-template default --finetuning_type lora --export_dir E:\\llm-train\\chatglm3-trained --export_size 2 --export_legacy_format False
[INFO|tokenization_utils_base.py:2082] 2024-03-30 17:18:44,968 >> loading file tokenizer.model
[INFO|tokenization_utils_base.py:2082] 2024-03-30 17:18:44,968 >> loading file added_tokens.json
[INFO|tokenization_utils_base.py:2082] 2024-03-30 17:18:44,968 >> loading file special_tokens_map.json
[INFO|tokenization_utils_base.py:2082] 2024-03-30 17:18:44,969 >> loading file tokenizer_config.json
[INFO|tokenization_utils_base.py:2082] 2024-03-30 17:18:44,969 >> loading file tokenizer.json
Setting eos_token is not supported, use the default one.
Setting pad_token is not supported, use the default one.
Setting unk_token is not supported, use the default one.
[INFO|configuration_utils.py:724] 2024-03-30 17:18:45,127 >> loading configuration file E:\\llm-train\\chatglm3-6b\config.json
[INFO|configuration_utils.py:724] 2024-03-30 17:18:45,132 >> loading configuration file E:\\llm-train\\chatglm3-6b\config.json
[INFO|configuration_utils.py:789] 2024-03-30 17:18:45,133 >> Model config ChatGLMConfig {"_name_or_path": "E:\\\\llm-train\\\\chatglm3-6b","add_bias_linear": false,"add_qkv_bias": true,"apply_query_key_layer_scaling": true,"apply_residual_connection_post_layernorm": false,"architectures": ["ChatGLMModel"],"attention_dropout": 0.0,"attention_softmax_in_fp32": true,"auto_map": {"AutoConfig": "configuration_chatglm.ChatGLMConfig","AutoModel": "modeling_chatglm.ChatGLMForConditionalGeneration","AutoModelForCausalLM": "modeling_chatglm.ChatGLMForConditionalGeneration","AutoModelForSeq2SeqLM": "modeling_chatglm.ChatGLMForConditionalGeneration","AutoModelForSequenceClassification": "modeling_chatglm.ChatGLMForSequenceClassification"},"bias_dropout_fusion": true,"classifier_dropout": null,"eos_token_id": 2,"ffn_hidden_size": 13696,"fp32_residual_connection": false,"hidden_dropout": 0.0,"hidden_size": 4096,"kv_channels": 128,"layernorm_epsilon": 1e-05,"model_type": "chatglm","multi_query_attention": true,"multi_query_group_num": 2,"num_attention_heads": 32,"num_layers": 28,"original_rope": true,"pad_token_id": 0,"padded_vocab_size": 65024,"post_layer_norm": true,"pre_seq_len": null,"prefix_projection": false,"quantization_bit": 0,"rmsnorm": true,"seq_length": 8192,"tie_word_embeddings": false,"torch_dtype": "float16","transformers_version": "4.39.2","use_cache": true,"vocab_size": 65024
}03/30/2024 17:18:45 - INFO - llmtuner.model.patcher - Using KV cache for faster generation.
[INFO|modeling_utils.py:3280] 2024-03-30 17:18:45,287 >> loading weights file E:\\llm-train\\chatglm3-6b\model.safetensors.index.json
[INFO|modeling_utils.py:1417] 2024-03-30 17:18:45,288 >> Instantiating ChatGLMForConditionalGeneration model under default dtype torch.float16.
[INFO|configuration_utils.py:928] 2024-03-30 17:18:45,289 >> Generate config GenerationConfig {"eos_token_id": 2,"pad_token_id": 0
}Loading checkpoint shards: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 7/7 [00:04<00:00, 1.60it/s]
[INFO|modeling_utils.py:4024] 2024-03-30 17:18:49,890 >> All model checkpoint weights were used when initializing ChatGLMForConditionalGeneration.[INFO|modeling_utils.py:4032] 2024-03-30 17:18:49,890 >> All the weights of ChatGLMForConditionalGeneration were initialized from the model checkpoint at E:\\llm-train\\chatglm3-6b.
If your task is similar to the task the model of the checkpoint was trained on, you can already use ChatGLMForConditionalGeneration for predictions without further training.
[INFO|modeling_utils.py:3573] 2024-03-30 17:18:49,895 >> Generation config file not found, using a generation config created from the model config.
WARNING:root:Some parameters are on the meta device device because they were offloaded to the cpu.
03/30/2024 17:18:49 - INFO - llmtuner.model.adapter - Fine-tuning method: LoRA
INFO:llmtuner.model.adapter:Fine-tuning method: LoRA
03/30/2024 17:18:53 - INFO - llmtuner.model.adapter - Merged 1 adapter(s).
INFO:llmtuner.model.adapter:Merged 1 adapter(s).
03/30/2024 17:18:53 - INFO - llmtuner.model.adapter - Loaded adapter(s): E:\\llm-train\\LLaMA-Factory\\saves\\ChatGLM3-6B-Chat\\lora\\train_glm3\\checkpoint-200
INFO:llmtuner.model.adapter:Loaded adapter(s): E:\\llm-train\\LLaMA-Factory\\saves\\ChatGLM3-6B-Chat\\lora\\train_glm3\\checkpoint-200
03/30/2024 17:18:53 - INFO - llmtuner.model.loader - all params: 6243584000
INFO:llmtuner.model.loader:all params: 6243584000
[INFO|configuration_utils.py:471] 2024-03-30 17:18:53,066 >> Configuration saved in E:\\llm-train\\chatglm3-trained\config.json
[INFO|configuration_utils.py:697] 2024-03-30 17:18:53,066 >> Configuration saved in E:\\llm-train\\chatglm3-trained\generation_config.json
[INFO|modeling_utils.py:2482] 2024-03-30 17:19:08,637 >> The model is bigger than the maximum size per checkpoint (2GB) and is going to be split in 7 checkpoint shards. You can find where each parameters has been saved in th
e index located at E:\\llm-train\\chatglm3-trained\model.safetensors.index.json.
[INFO|tokenization_utils_base.py:2502] 2024-03-30 17:19:08,642 >> tokenizer config file saved in E:\\llm-train\\chatglm3-trained\tokenizer_config.json
[INFO|tokenization_utils_base.py:2511] 2024-03-30 17:19:08,643 >> Special tokens file saved in E:\\llm-train\\chatglm3-trained\special_tokens_map.json
(llm) PS E:\llm-train\LLaMA-Factory>
测试模型
1、修改训练后的模型路径：MODEL_PATH = os.environ.get(‘MODEL_PATH’, ‘E:\llm-train\chatglm3-trained’)
2、运行：ChatGLM3\basic_demo\cli_demo.py
用户：>? 安妮
人类。
年龄大约二十岁。
性格开朗。
青年女子。
青年女优。
音色甜美的声音。
用户：>? 魔魂
ChatGLM：亡灵。
年龄大约三十岁。
性格狡猾。
亡灵法师。
膂力过人。
声音低沉。
用户：
相关文章：
LLaMA-Factory微调（sft）ChatGLM3-6B保姆教程
LLaMA-Factory微调（sft）ChatGLM3-6B保姆教程 准备 1、下载 下载LLaMA-Factory下载ChatGLM3-6B下载ChatGLM3windows下载CUDA ToolKit 12.1 （本人是在windows进行训练的，显卡GTX 1660 Ti） CUDA安装完毕后，…...
Web安全-浏览器安全策略及跨站脚本攻击与请求伪造漏洞原理
Web安全-浏览器安全策略及跨站脚本攻击与请求伪造漏洞原理 Web服务组件分层概念 静态层 ：web前端框架：Bootstrap，jQuery,HTML5框架等，主要存在跨站脚本攻击脚本层：web应用，web开发框架，web服务…...
蓝桥杯B组C++省赛——飞机降落（DFS）
题目连接：https://www.lanqiao.cn/problems/3511/learning/ 思路：由于数据范围很小，所有选择用DFS枚举所有飞机的所有的降落顺序，看哪个顺序可以让所有飞机顺利降落，有的话就算成功方案，输出了“YES”。 …...
Java 中的 Map集合
文章目录 添加和修改元素获取元素检查元素删除元素获取所有键 / 值 / 键值对大小 在 Java 中，Map 接口是 Java 集合框架的一部分，它存储键值对（key-value pairs）。Map 接口有许多常用的方法，用于添加、删除、获取元素&…...
基于springboot大学生兼职平台管理系统（完整源码+数据库）
一、项目简介 本项目是一套基于springboot大学生兼职平台管理系统 包含：项目源码、数据库脚本等，该项目附带全部源码可作为毕设使用。 项目都经过严格调试，eclipse或者idea 确保可以运行！ 该系统功能完善、界面美观、操作简单、功…...
C#学生信息管理系统
一、引言 学生信息管理系统是现代学校管理的重要组成部分，它能够有效地管理学生的基本信息、课程信息、成绩信息等，提高学校管理的效率和质量。本文将介绍如何使用SQL Server数据库和C#语言在.NET平台上开发一个学生信息管理系统的课程设计项目。 二、项…...
双机 Cartogtapher 建图文件配置
双机cartogtapher建图 最近在做硕士毕设的最后一个实验，其中涉及到多机建图，经过调研最终采用cartographer建图算法，其中配置多机建图的文件有些麻烦，特此博客以记录 非常感谢我的同门 ”叶少“ 山上的稻草人-CSDN博客的帮助&am…...
VMware提示 该虚拟机似乎正在使用中,如何解决？
VMware提示 该虚拟机似乎正在使用中,如何解决？ 问题描述解决方法1.找到安装VMware的文件目录2.在VMware目录下.lck后缀的文件夹删除或重命名3.运行VMware 问题描述 该虚拟机似乎正在使用中。 如果该虚拟机未在使用，请按“获取所有权(T)”按钮获取它的所…...
阿里云短信服务业务
一、了解阿里云用户权限操作 1.注册账号、实名认证； 2.使用AccessKey 步骤一 点击头像，权限安全的AccessKey 步骤二 设置子用户AccessKey 步骤三 添加用户组和用户 步骤四 添加用户组记得绑定短信服务权限 步骤五 添加用户记得勾选openApi访问 添加…...
ElasticSearch的DSL查询
ElasticSearch的DSL查询 准备工作 创建测试方法，初始化测试结构。 import org.apache.http.HttpHost; import org.apache.lucene.search.TotalHits; import org.elasticsearch.action.search.SearchRequest; import org.elasticsearch.action.search.SearchRespo…...
每天定时杀spark进程
##编写shell脚本 #!/bin/bash arr(“zhangsan” “lisi” “wangwu”) for i in “${arr[]}” do processps -ef|grep ${i}| grep -v "grep"| awk {print $2} kill -9 ${process} done ##每日定时杀手动启动的进程 0 19 * * * cd /kill_process && sh kil…...
win10 安装kubectl，配置config连接k8s集群
安装kubectl 按照官方文档安装：https://kubernetes.io/docs/tasks/tools/install-kubectl-windows/ curl安装 （1）下载curl安装压缩包: curl for Windows （2）配置环境变量： 用户变量： Path变…...
Calico IPIP和BGP TOR的数据包走向
IPIP Mesh全网互联 文字描述 APOD eth0 10.7.75.132 -----> APOD 网关 -----> A宿主机 cali76174826315网卡 -----> Atunl0 10.7.75.128 封装 ----> Aeth0 10.120.181.20 -----> 通过网关 10.120.181.254 -----> 下一跳 BNODE eth0 10.120.179.8 解封装 --…...
静态成员主要用于提供与类本身相关的功能或数据，有什么应用场景
静态成员（包括静态方法和静态属性）在JavaScript中常用于多种应用场景，它们为类提供了与类本身直接相关而不是与实例相关的功能或数据。以下是一些常见的应用场景： 工厂方法 静态方法可以作为工厂方法，用于创建类的实…...
在线考试|基于Springboot的在线考试管理系统设计与实现(源码+数据库+文档)
在线考试管理系统目录 目录 基于Springboot的在线考试管理系统设计与实现 一、前言 二、系统设计 三、系统功能设计 1、前台： 2、后台 管理员功能 四、数据库设计 五、核心代码 六、论文参考 七、最新计算机毕设选题推荐 八、源码获取： 博主…...
C语言-----数据在内存中的存储(1)
1.整数在内存中的存储 我们之前就了解过整数的二进制写法分别有3种，分别为原码，反码，补码。整型在内存中存储的是补码。 原码，反码，补码都有自己的符号位和数值位，符号位为1时，则表示负数&…...
Ribbon有哪些负载均衡策略
负载均衡类都实现了IRule接口。 RandomRule：随机的选用一个实例 RoundRobinRule：轮询的使用实例 RetryRule：在轮询的基础上加了一个错误重试机制，在deadline时间内会不断的重试 WeightResponeTimeRule：根据权重去做…...
websocket多级nginx代理
在使用多层Nginx代理时，WebSocket的连接可能会遇到一些问题，因为WebSocket连接是持久化的，它需要Upgrade头部来确认升级到WebSocket协议。在多层代理的情况下，每层代理可能会修改或丢失这个Upgrade头部信息。 为了确保WebSocket能…...
【python从入门到精通】-- 第四战：语句汇总
🌈 个人主页：白子寰 🔥 分类专栏：python从入门到精通，魔法指针，进阶C，C语言，C语言题集，C语言实现游戏👈 希望得到您的订阅和支持~ 💡 坚持创作博文…...
【NC50937】货仓选址
题目 货仓选址 二分，前缀和，数学推导 思路 由题意可知货仓的位置是可以和商店的位置重合的。首先应该将商店的坐标从小到大排序，然后假设商店的坐标为 a i a_i ai，货仓的坐标为 x x x，货仓左侧第一家商店&#x…...
大数据学习栈记——Neo4j的安装与使用
本文介绍图数据库Neofj的安装与使用，操作系统：Ubuntu24.04，Neofj版本：2025.04.0。 Apt安装 Neofj可以进行官网安装：Neo4j Deployment Center - Graph Database & Analytics 我这里安装是添加软件源的方法 最新版…...
Lombok 的 @Data 注解失效，未生成 getter/setter 方法引发的HTTP 406 错误
HTTP 状态码 406 (Not Acceptable) 和 500 (Internal Server Error) 是两类完全不同的错误，它们的含义、原因和解决方法都有显著区别。以下是详细对比： 1. HTTP 406 (Not Acceptable) 含义： 客户端请求的内容类型与服务器支持的内容类型不匹…...
day52 ResNet18 CBAM
在深度学习的旅程中，我们不断探索如何提升模型的性能。今天，我将分享我在 ResNet18 模型中插入 CBAM（Convolutional Block Attention Module）模块，并采用分阶段微调策略的实践过程。通过这个过程，我不仅提升…...
解决Ubuntu22.04 VMware失败的问题 ubuntu入门之二十八
现象1 打开VMware失败 Ubuntu升级之后打开VMware上报需要安装vmmon和vmnet，点击确认后如下提示 最终上报fail 解决方法 内核升级导致，需要在新内核下重新下载编译安装 查看版本 $ vmware -v VMware Workstation 17.5.1 build-23298084$ lsb_release…...
系统设计 --- MongoDB亿级数据查询优化策略
系统设计 --- MongoDB亿级数据查询分表策略 背景Solution --- 分表 背景 使用audit log实现Audi Trail功能 Audit Trail范围: 六个月数据量: 每秒5-7条audi log，共计7千万 – 1亿条数据需要实现全文检索按照时间倒序因为license问题，不能使用ELK只能使用…...
04-初识css
一、css样式引入 1.1.内部样式 <div style"width: 100px;"></div>1.2.外部样式 1.2.1.外部样式1 <style>.aa {width: 100px;} </style> <div class"aa"></div>1.2.2.外部样式2 <!-- rel内表面引入的是style样…...
【Java_EE】Spring MVC
目录 Spring Web MVC 编辑注解 RestController RequestMapping RequestParam RequestParam RequestBody PathVariable RequestPart 参数传递 注意事项 编辑参数重命名 RequestParam 编辑编辑传递集合 RequestParam 传递JSON数据 编辑RequestBody …...
汇编常见指令
汇编常见指令 一、数据传送指令 指令功能示例说明MOV数据传送MOV EAX, 10将立即数 10 送入 EAXMOV [EBX], EAX将 EAX 值存入 EBX 指向的内存LEA加载有效地址LEA EAX, [EBX4]将 EBX4 的地址存入 EAX（不访问内存）XCHG交换数据XCHG EAX, EBX交换 EAX 和 EB…...
Caliper 配置文件解析：config.yaml
Caliper 是一个区块链性能基准测试工具，用于评估不同区块链平台的性能。下面我将详细解释你提供的 fisco-bcos.json 文件结构，并说明它与 config.yaml 文件的关系。 fisco-bcos.json 文件解析 这个文件是针对 FISCO-BCOS 区块链网络的 Caliper 配置文件，主要包含以下几个部…...
【论文阅读28】-CNN-BiLSTM-Attention-（2024）
本文把滑坡位移序列拆开、筛优质因子，再用 CNN-BiLSTM-Attention 来动态预测每个子序列，最后重构出总位移，预测效果超越传统模型。 文章目录 1 引言2 方法2.1 位移时间序列加性模型2.2 变分模态分解 (VMD) 具体步骤2.3.1 样本熵（S…...