{"id":20316,"date":"2025-08-06T12:17:42","date_gmt":"2025-08-06T04:17:42","guid":{"rendered":"https:\/\/aicats.wiki\/sites\/20316.html"},"modified":"2025-08-06T12:17:56","modified_gmt":"2025-08-06T04:17:56","slug":"huggingface","status":"publish","type":"sites","link":"https:\/\/aicats.wiki\/tw\/sites\/20316.html","title":{"rendered":"HuggingFace"},"content":{"rendered":"

\u8fd1\u5e74\u6765\uff0c\u4eba\u5de5\u667a\u80fd\u4e0e\u81ea\u7136\u8bed\u8a00\u5904\u7406\u6280\u672f\u98de\u901f\u53d1\u5c55\uff0c\u800c\u805a\u5408\u521b\u65b0\uff0c\u9a71\u52a8\u884c\u4e1a\u751f\u6001\u7684HuggingFace\u6beb\u65e0\u7591\u95ee\u6210\u4e86\u4e1a\u754c\u5173\u6ce8\u7684\u7126\u70b9\u3002\u4f5c\u4e3a\u5168\u7403\u6700\u91cd\u8981\u7684\u5f00\u6e90AI\u5e73\u53f0<\/a>\u4e4b\u4e00\uff0cHuggingFace\u4e0d\u4ec5\u88ab\u8a89\u4e3a\u201cAI\u754c\u7684GitHub\u201d\uff0c\u66f4\u5df2\u6210\u4e3aAI\u8bad\u7ec3\u6a21\u578b\u4e0e\u4ea7\u4e1a\u5408\u4f5c\u7684\u67a2\u7ebd\u3002<\/strong>\u672c\u62a5\u5bfc\u5c06\u5168\u9762\u8be0\u91caHuggingFace\u7684\u6838\u5fc3\u529f\u80fd\u3001\u4ef7\u683c\u4f53\u7cfb\u3001\u5b9e\u9645\u5e94\u7528\u65b9\u6cd5\u3001\u9002\u7528\u5bf9\u8c61\u53ca\u5176\u5bf9\u884c\u4e1a\u53d1\u5c55\u7684\u91cd\u5927\u5f71\u54cd\u3002<\/p>\n\n\n\n

HuggingFace\u5e73\u53f0\u7b80\u4ecb<\/p>\n\n\n\n

HuggingFace\u6210\u7acb\u4e8e2016\u5e74\uff0c\u603b\u90e8\u4f4d\u4e8e\u7f8e\u56fd\u7ebd\u7ea6\u3002\u6700\u521d\u662f\u4e00\u4e2a\u9762\u5411\u9752\u5c11\u5e74\u7684\u804a\u5929\u673a\u5668\u4eba\u521b\u4e1a\u516c\u53f8\uff0c\u540e\u56e0\u5f00\u6e90\u7b56\u7565\u8fc5\u901f\u8f6c\u578b\u4e3aAI\u4ea7\u4e1a\u5f00\u6e90\u793e\u533a\u5e73\u53f0\u3002<\/strong>\u4eca\u65e5\uff0cHuggingFace\u805a\u5408\u4e86\u6765\u81ea\u5168\u7403\u7684\u5f00\u53d1\u8005\u3001\u7814\u7a76\u8005\u4e0e\u4f01\u4e1a\uff0c\u4e3aAI\u79d1\u7814\u3001\u4ea7\u4e1a\u8f6c\u578b\u4e0e\u521b\u65b0\u521b\u4e1a\u63d0\u4f9b\u5168\u65b9\u4f4d\u6280\u672f\u548c\u793e\u533a\u652f\u6491\u3002<\/p>\n\n\n\n

\u8bbf\u95eeHuggingFace\u5b98\u65b9\u5e73\u53f0\uff1ahttps:\/\/huggingface.co\/<\/a><\/p>\n\n\n\n

\u5176\u6700\u5927\u7279\u8272\u5728\u4e8e\u201c\u5f00\u653e\u5171\u4eab\u201d\uff0c\u4e0d\u4ec5\u5f00\u653eNLP\u3001CV\u3001\u97f3\u9891\u7b49\u5404\u9886\u57dfAI\u8bad\u7ec3\u6a21\u578b\u3001\u6570\u636e\u96c6\u3001\u5de5\u5177\u53caApp\u5b9e\u4f8b\uff0c\u8fd8\u652f\u6301\u5728\u7ebf\u8fd0\u884c\u3001\u534f\u4f5c\u5f00\u53d1\u4e0e\u5b89\u5168\u6258\u7ba1\u3002<\/strong><\/p>\n\n\n\n

HuggingFace\u7684\u4e3b\u8981\u529f\u80fd<\/h2>\n\n\n\n

HuggingFace\u7684\u529f\u80fd\u4f53\u7cfb\u6781\u4e3a\u4e30\u5bcc\uff0c\u6db5\u76d6\u4e86\u4ece\u6570\u636e\u5230\u6a21\u578b\u518d\u5230\u5e94\u7528\u843d\u5730\u7684\u5168\u94fe\u8def\u3002\u5982\u4e0b\u8868\u6c47\u603b\u4e86\u5176\u6838\u5fc3\u4ea7\u54c1\u4e0e\u6a21\u5757\u4ee5\u53ca\u4ee3\u8868\u6027\u63cf\u8ff0\uff1a<\/p>\n\n\n\n

\"HuggingFace\u5b98\u65b9\u5e73\u53f0\"
\u5716\uff0fHuggingFace\u5b98\u65b9\u5e73\u53f0<\/a><\/figcaption><\/figure>\n\n\n\n
\u7d44\u4ef6<\/td>\u529f\u80fd\u6982\u8ff0<\/td>\u5feb\u901f\u9023\u7d50<\/td><\/tr>
Hugging Face Hub<\/td>\u6a21\u578b\u3001\u6570\u636e\u96c6\u3001Spaces\u6258\u7ba1\uff0c\u793e\u533a\u534f\u4f5c<\/td>Hub\u6587\u6863\u5165\u53e3<\/a><\/td><\/tr>
Transformers\u5e93<\/td>\u652f\u6301\u6587\u672c\u3001\u56fe\u50cf\u3001\u97f3\u9891\u7684\u9884\u8bad\u7ec3\u548c\u63a8\u7406<\/td>Transformers\u6307\u5357<\/a><\/td><\/tr>
Datasets\u5e93<\/td>\u4f18\u8d28\u6807\u51c6\u5316\u6570\u636e\u96c6\u6d4f\u89c8\u3001\u4e0b\u8f7d\u4e0e\u7ba1\u7406<\/td>Datasets\u6307\u5357<\/a><\/td><\/tr>
Tokenizers<\/td>\u9ad8\u6548\u5206\u8bcd\u3001\u5b50\u8bcd\u5207\u5206\uff0c\u9002\u914d\u591a\u8bed\u8a00<\/td>Tokenizers\u6307\u5357<\/a><\/td><\/tr>
Diffusers<\/td>\u56fe\u50cf\/\u89c6\u9891\u6269\u6563\u6a21\u578b\uff0c\u751f\u6210\u5f0fAI\u9886\u57df\u6838\u5fc3<\/td>Diffusers\u4ecb\u7ecd<\/a><\/td><\/tr>
Evaluate<\/td>\u63d0\u4f9b\u6a21\u578b\u8bc4\u4f30\u5de5\u5177<\/td>Evaluate\u6587\u6863<\/a><\/td><\/tr>
Gradio<\/td>\u79d2\u7ea7\u642d\u5efaAI Web UI Demo<\/td>Gradio\u6587\u6863<\/a><\/td><\/tr>
Safetensors<\/td>\u5b89\u5168\u9ad8\u6548\u7684\u6743\u91cd\u6587\u4ef6\u683c\u5f0f<\/td>Safetensors\u8bf4\u660e<\/a><\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n

\u6838\u5fc3\u4eae\u70b9\u529f\u80fd\u68b3\u7406<\/h3>\n\n\n\n

1. HuggingFace Hub\uff1a<\/strong>
\u4ee5Git\u4e3a\u6838\u5fc3\uff0cHuggingFace Hub\u652f\u6491\u6a21\u578b\u3001\u6570\u636e\u96c6\u3001Space App\u7684\u6258\u7ba1\u3001\u68c0\u7d22\u548c\u7248\u672c\u7ba1\u7406\u3002\u7528\u6237\u4e0d\u4ec5\u53ef\u514d\u8d39\u4e0a\u4f20\u3001\u7ba1\u7406\u6a21\u578b\u4e0e\u8d44\u6e90\uff0c\u8fd8\u80fd\u4e0e\u5168\u7403\u5f00\u53d1\u8005\u534f\u540c\u8fed\u4ee3\u3002<\/p>\n\n\n\n

2. Transformers\u5e93\uff1a<\/strong>
\u8fd9\u662fNLP\u7b49\u9886\u57df\u6570\u4e07\u5f00\u53d1\u56e2\u961f\u7684\u9996\u9009\u3002\u6db5\u76d6BERT\u3001GPT\u3001T5\u3001Llama\u3001StableLM\u7b49\u9884\u8bad\u7ec3\u5927\u6a21\u578b<\/strong>\uff0c\u652f\u6301\u6587\u672c\u5206\u7c7b\u3001\u6458\u8981\u3001\u5b9e\u4f53\u8bc6\u522b\u3001\u6587\u672c\u751f\u6210\u3001\u673a\u5668\u7ffb\u8bd1\u3001\u56fe\u50cf\u5206\u7c7b\u3001\u8bed\u97f3\u8bc6\u522b\u7b49\u9ad8\u6548\u529f\u80fd\u3002\u652f\u6301Python\/PyTorch\u3001Tensorflow\u548cJAX\u7b49\u4e3b\u6d41\u6846\u67b6\u3002
\u76f8\u95dc\u9023\u7d50\uff1aTransformers\u5e93\u8be6\u60c5<\/a><\/p>\n\n\n\n

3. Datasets\u5e93\uff1a<\/strong>
\u652f\u6301\u6570\u4e07\u6761NLP\u3001CV\u3001\u97f3\u9891\u7b49\u573a\u666f\u7684\u6570\u636e\u96c6\uff0c\u63d0\u4f9b\u667a\u80fd\u8fc7\u6ee4\u4e0b\u8f7d\u3001\u4e00\u952e\u52a0\u8f7d\uff08\u6d41\u5f0f\u8bbf\u95ee\uff09\uff0c\u8fd8\u5141\u8bb8\u81ea\u5b9a\u4e49\u4e0a\u4f20\u6570\u636e\u96c6\uff0c\u6784\u5efa\u4e2a\u6027\u5316AI\u8bad\u7ec3pipeline\u3002
\u67e5\u770b\u793a\u4f8b\uff1aDatasets\u6d4f\u89c8<\/a><\/p>\n\n\n\n

\"HuggingFace
\u5716\uff0fHuggingFace Transformers\u5e93 \u9875\u9762<\/a><\/figcaption><\/figure>\n\n\n\n

4. Tokenizers\uff1a<\/strong>
\u81ea\u7814\u9ad8\u6027\u80fd\u5206\u8bcd\u5f15\u64ce\uff0c\u52a0\u6301\u6df1\u5ea6\u5b66\u4e60\u6587\u672c\u5206\u6790\uff0c\u6781\u5927\u63d0\u5347\u8bad\u7ec3\u6548\u7387\u4e0e\u56fd\u9645\u5316\u9002\u914d\u529b\uff08\u5982BPE\u3001WordPiece\u3001SentencePiece\u7b49\u5206\u8bcd\u6cd5\uff09\u3002<\/p>\n\n\n\n

5. Diffusers\u4e0e\u751f\u6210\u5f0fAI\u9886\u57df\uff1a<\/strong>
\u9488\u5bf9\u751f\u6210\u5f0fAI\uff0cDiffusers\u6a21\u5757\u652f\u6301Stable Diffusion\u3001ControlNet\u3001DALL-E\u7b49\uff0c\u9002\u7528\u4e8e\u6587\u751f\u56fe\u3001\u89c6\u9891\u4e0e\u97f3\u9891\u751f\u6210\u521b\u65b0\u5e94\u7528\u3002
\u5165\u53e3\uff1aDiffusers\u5e93<\/a><\/p>\n\n\n\n

6. Spaces & Online Demo\u5ba4<\/strong>
\u5e73\u53f0\u5185\u7f6e\u6570\u4e07\u4e2aSpaces\u5e94\u7528(\u4f4e\u4ee3\u7801\/\u5728\u7ebfdemo\/\u5c0f\u578b\u7a0b\u5e8f)\uff0c\u53ef\u5728\u7ebf\u8bd5\u73a9\u6587\u672c\u751f\u6210\u3001\u56fe\u50cf\u751f\u6210\u3001\u95ee\u7b54\u3001\u97f3\u9891\u8bc6\u522b\u7b49\u529f\u80fd\u3002
\u8a2a\u554f\uff1aSpaces\u8bd5\u73a9<\/a><\/p>\n\n\n\n

HuggingFace\u7684\u4ef7\u683c & \u65b9\u6848<\/h2>\n\n\n\n

HuggingFace\u5927\u90e8\u5206\u529f\u80fd\u5bf9\u4e2a\u4eba\u4e0e\u7814\u7a76\u8005\u514d\u8d39\u5f00\u653e\uff0c\u4f46\u9488\u5bf9\u66f4\u9ad8\u9636\u7684\u9700\u6c42\u53ca\u4f01\u4e1a\u573a\u666f\uff0c\u63d0\u4f9b\u7075\u6d3b\u7684\u4ed8\u8d39\u9009\u9879\u3002<\/strong>
\u8be6\u7ec6\u5b9a\u4ef7\u6587\u6863\uff1ahttps:\/\/huggingface.co\/pricing<\/a><\/p>\n\n\n\n

\"HuggingFace\u5b9a\u4ef7\u9875\u9762\"
\u5716\uff0fHuggingFace\u5b9a\u4ef7\u9875\u9762<\/a><\/figcaption><\/figure>\n\n\n\n

\u4e3b\u8981\u65b9\u6848\u6bd4\u8f03\u5982\u4e0b\uff1a<\/strong><\/p>\n\n\n\n

\u8ba1\u5212\u7c7b\u578b<\/td>\u8d39\u7528\uff08\u7f8e\u5143\/\u6708\/\u7528\u6237\uff09<\/td>\u4e3b\u8981\u7279\u8272<\/td><\/tr>
Free \u514d\u8d39\u65b9\u6848<\/td>$0<\/td>\u57fa\u7840Hub\u8bbf\u95ee\u3001\u4e0b\u8f7d\u6a21\u578b\/\u6570\u636e\u96c6\u3001\u5f00\u6e90\u793e\u533a\u53c2\u4e0e<\/td><\/tr>
Pro \u5c08\u696d\u7248<\/td>$9<\/td>API\u8c03\u7528\u989d\u5ea6\u63d0\u5347\u3001ZeroGPU\u7b49\u9ad8\u9636\u63a8\u7406\u8bd5\u7528\u3001\u65b0\u529f\u80fd\u62a2\u5148\u4f53\u9a8c<\/td><\/tr>
Enterprise Hub<\/td>$20<\/td>\u66f4\u5f3a\u9690\u79c1\u3001\u6743\u9650\u914d\u7f6e\u3001\u4f01\u4e1a\u5b89\u5168\u3001\u6570\u636e\u9694\u79bb\u652f\u6301<\/td><\/tr>
Spaces Hardware<\/td>$0\u8d77\/\u5c0f\u65f6<\/td>CPU\/GPU\/\u52a0\u901f\u5361\u8d44\u6e90\u52a8\u6001\u6269\u5bb9\u3001\u652f\u6301\u5927\u89c4\u6a21\u8fd0\u7b97\u5b9e\u9a8c<\/td><\/tr>
Inference Endpoints<\/td>\u6309\u7528\u91cf\u8a08\u8cbb<\/td>\u4e91\u7aef\u6a21\u578b\u5b9e\u65f6\u63a8\u7406\u670d\u52a1\u3001\u81ea\u52a8\u5f39\u6027\u6269\u5bb9<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n

\u4f01\u4e1a\u53ef\u6309\u9700\u63a5\u5165\u79c1\u6709\u5316\u90e8\u7f72\u3001\u5b9a\u5236API\u3001\u4e13\u9879\u5b89\u5168\u52a0\u5bc6\u4e0e\u5408\u89c4\u4fdd\u969c\u3002<\/strong>
\u8be6\u7ec6\u4ef7\u683c\u53c2\u8003\uff1a\u5b9a\u50f9\u8aaa\u660e<\/a><\/p>\n\n\n\n

\u5982\u4f55\u4f7f\u7528HuggingFace<\/h2>\n\n\n\n

1. \u5e33\u865f\u8a3b\u518a\u8207\u74b0\u5883\u6e96\u5099<\/strong>
– \u6ce8\u518c\u8d26\u53f7\uff1a\u8a3b\u518a\u9801\u9762<\/a>
– \u83b7\u53d6API Token\uff1aToken\u8bbe\u7f6e\u9875<\/a><\/p>\n\n\n\n

2. \u67e5\u627e\u4e0e\u52a0\u8f7d\u6a21\u578b\/\u6570\u636e\u96c6<\/strong>
– \u901a\u8fc7Hub\u4e3b\u9875<\/a>\u8f93\u5165\u5173\u952e\u8bcd\u667a\u80fd\u68c0\u7d22\u6216\u4f7f\u7528\u8fc7\u6ee4\u5668\u7b5b\u9009\u6a21\u578b\u7c7b\u522b\u548c\u4f5c\u8005\u3002
– \u5bf9\u4e8ePython\u7528\u6237\uff0c\u4e00\u884c\u4ee3\u7801\u5373\u53ef\u81ea\u52a8\u52a0\u8f7d\u4e0e\u5fae\u8c03\u6a21\u578b\uff0c\u4f8b\u5982\u6587\u672c\u751f\u6210\uff08\u4ee5`transformers`\u5e93\u4e3a\u4f8b\uff09\uff1a<\/p>\n\n\n\n\n

from transformers import pipeline\n\nclassifier = pipeline('sentiment-analysis', model='distilbert-base-uncased-finetuned-sst-2-english')\nresult = classifier(\"HuggingFace\u771f\u7684\u5f88\u5f3a\u5927\uff01\")\nprint(result)\n<\/code><\/pre>\n\n\n\n\n
\"\u8a3b\u518a\u9801\u9762\"
\u5716\uff0f\u8a3b\u518a\u9801\u9762<\/a><\/figcaption><\/figure>\n\n\n\n
– \u52a0\u8f7d\u6570\u636e\u96c6\u53c2\u8003\uff1a<\/p>\n\n\n\n\n
from datasets import load_dataset\n\ndataset = load_dataset(\"imdb\")\nprint(dataset[\"trai<\/a>n\"][:2])\n<\/code><\/pre>\n\n\n\n\n
3. \u4e0b\u8f7d\u4e0e\u672c\u5730\u90e8\u7f72<\/strong>
HuggingFace\u652f\u6301\u76f4\u63a5\u7f51\u9875\u4e0b\u8f7d\u3001\u547d\u4ee4\u884c(huggingface-cli)\u4e0b\u8f7d\u3001git clone\uff0c\u6216\u4f7f\u7528API\/SDK\u6279\u91cf\u540c\u6b65\u6a21\u578b\u4e0e\u6570\u636e\u96c6\u3002
\u8be6\u7ec6\u64cd\u4f5c\u65b9\u6cd5\u53ef\u67e5\u9605\u5b98\u65b9\u6307\u5357<\/a>\u53ca HF-Mirror\u955c\u50cf\u52a0\u901f\u8be6\u89e3<\/a>\u3002<\/p>\n\n\n\n
4. \u4f53\u9a8cSpaces\u548cDemo<\/strong>
\u65e0\u9700\u4ee3\u7801\u57fa\u7840\uff0c\u76f4\u63a5\u4e0a\u7ebfSpaces\u5e94\u7528\uff0c\u5728\u7ebf\u5c1d\u8bd5\u5404\u79cdAI\u8bad\u7ec3\u6a21\u578b\u6548\u679c\u3002
\u5165\u53e3\uff1ahttps:\/\/huggingface.co\/spaces<\/a><\/p>\n\n\n\n
5. \u79c1\u6709\u6a21\u578b\/\u4f01\u4e1a\u534f\u4f5c<\/strong>
\u4f01\u4e1a\u53ef\u81ea\u5efaPrivate Hub\uff0c\u5728\u7ebf\u914d\u7f6e\u591a\u7ea7\u6743\u9650\u4e0e\u6570\u636e\u9694\u79bb\uff0c\u9ad8\u6548\u534f\u4f5c\u3002<\/p>\n\n\n\n
HuggingFace\u7684\u9002\u7528\u4eba\u7fa4<\/h2>\n\n\n\n
\"huggingface
\u5716\uff0fhuggingface space<\/a><\/figcaption><\/figure>\n\n\n\n
HuggingFace\u5e73\u53f0\u51e0\u4e4e\u9002\u7528\u4e8e\u6240\u6709\u5173\u6ce8AI\u521b\u65b0\u7684\u7528\u6237\u7fa4\u4f53\uff1a<\/strong><\/p>\n\n\n\n