{"id":543,"date":"2025-11-23T10:46:38","date_gmt":"2025-11-23T02:46:38","guid":{"rendered":"https:\/\/www.idcyz.com\/blog\/?p=543"},"modified":"2025-11-23T10:46:38","modified_gmt":"2025-11-23T02:46:38","slug":"%e5%85%b3%e4%ba%8efrom-tensorboardx-import-summarywriter%e7%9a%84%e6%8a%a5%e9%94%99%e6%8c%87%e5%8d%97%ef%bc%8c2025%e5%b9%b4%e6%8e%a8%e8%8d%90%e7%9a%84%e8%a7%a3%e5%86%b3%e6%96%b9%e6%a1%88","status":"publish","type":"post","link":"https:\/\/www.idcyz.com\/blog\/543.html","title":{"rendered":"\u5173\u4e8efrom tensorboardx import summarywriter\u7684\u62a5\u9519\u6307\u5357\uff0c2025\u5e74\u63a8\u8350\u7684\u89e3\u51b3\u65b9\u6848"},"content":{"rendered":"

\"\u5173\u4e8efrom<\/p>\n

1. from tensorboardx import summarywriter \u62a5\u9519\u6982\u8ff0<\/h3>\n<\/p>\n

\u9047\u5230 “from tensorboardx import summarywriter” \u62a5\u9519\u65f6\uff0c\u901a\u5e38\u662f\u56e0\u4e3a TensorBoardX \u5e93\u6ca1\u6709\u6b63\u786e\u5b89\u88c5\u6216\u4e0e\u5f53\u524d Python \u73af\u5883\u4e0d\u517c\u5bb9\u3002\u8981\u89e3\u51b3\u6b64\u95ee\u9898\uff0c\u9996\u5148\u8981\u786e\u4fdd\u5b89\u88c5\u4e86\u6b63\u786e\u7684\u7248\u672c\u3002\u5efa\u8bae\u4f7f\u7528\u4ee5\u4e0b\u547d\u4ee4\u6765\u5b89\u88c5 TensorBoardX\uff0c\u786e\u4fdd\u60a8\u5df2\u7ecf\u5728 Python \u7684\u865a\u62df\u73af\u5883\u4e2d\uff1a <\/p>\n

pip install tensorboardX<\/code><\/pre>\n<\/p>\n
\u540c\u6837\uff0c\u68c0\u67e5 TensorFlow \u6216 PyTorch \u7684\u4f7f\u7528\u7248\u672c\u662f\u5426\u4e0e TensorBoardX \u517c\u5bb9\u3002\u4f8b\u5982\uff0cTensorBoardX \u7684\u67d0\u4e9b\u7248\u672c\u53ef\u80fd\u53ea\u652f\u6301\u5bf9\u5e94\u7248\u672c\u7684 TensorFlow \u6216 PyTorch\u3002<\/p>\n<\/p>\n
\n
2. \u4f9d\u8d56\u5173\u7cfb\u68c0\u67e5<\/h3>\n<\/p>\n
\u5f53\u9047\u5230\u5bfc\u5165\u9519\u8bef\u65f6\uff0c\u4e5f\u53ef\u80fd\u662f\u56e0\u4e3a\u7f3a\u5c11 TensorBoardX \u7684\u4f9d\u8d56\u9879\u3002\u5efa\u8bae\u4f7f\u7528\u4ee5\u4e0b\u547d\u4ee4\u5b89\u88c5\u76f8\u5173\u4f9d\u8d56\u5e93\uff1a  <\/p>\n
pip install numpy protobuf<\/code><\/pre>\n<\/p>\n
\u786e\u4fdd\u8fd9\u4e9b\u57fa\u7840\u5e93\u7684\u7248\u672c\u4e0e TensorBoardX \u517c\u5bb9\u3002\u53ef\u4ee5\u4f7f\u7528\u4ee5\u4e0b\u547d\u4ee4\u67e5\u770b\u5df2\u5b89\u88c5\u5e93\u7684\u7248\u672c\uff1a  <\/p>\n
pip show numpy protobuf tensorboardX<\/code><\/pre>\n<\/p>\n
\u5982\u679c\u8fd9\u4e9b\u5e93\u7684\u7248\u672c\u4e0d\u5339\u914d\uff0c\u53ef\u80fd\u4f1a\u5bfc\u81f4\u529f\u80fd\u5f02\u5e38\u6216\u5bfc\u5165\u5931\u8d25\u3002<\/p>\n<\/p>\n
\n
3. \u5b89\u88c5\u5728\u4e0d\u540c\u73af\u5883\u4e2d\u7684\u95ee\u9898<\/h3>\n<\/p>\n
\u5728\u591a\u4e2a Python \u7248\u672c\u6216\u865a\u62df\u73af\u5883\u4e2d\u5de5\u4f5c\u65f6\uff0c\u5bb9\u6613\u51fa\u73b0\u6b64\u7c7b\u5bfc\u5165\u9519\u8bef\u3002\u8981\u68c0\u67e5\u5f53\u524d Python \u73af\u5883\u548c\u5305\uff0c\u53ef\u4ee5\u5148\u67e5\u770b Python \u7684\u5f53\u524d\u7248\u672c\uff1a  <\/p>\n
python --version<\/code><\/pre>\n<\/p>\n
\u786e\u4fdd\u60a8\u5728\u6b63\u786e\u7684\u73af\u5883\u4e2d\u5b89\u88c5\u4e86 TensorBoardX\u3002\u6b64\u5916\uff0c\u53ef\u4ee5\u4f7f\u7528\u4ee5\u4e0b\u547d\u4ee4\u6765\u5217\u51fa\u5f53\u524d\u73af\u5883\u4e0b\u7684\u6240\u6709\u5305\uff1a  <\/p>\n
pip list<\/code><\/pre>\n<\/p>\n
\u786e\u4fdd\u5728\u5217\u8868\u4e2d\u80fd\u591f\u627e\u5230 tensorboardX\u3002<\/p>\n<\/p>\n
\n
4. \u4f7f\u7528\u7684\u793a\u4f8b\u53ca\u4ee3\u7801\u7247\u6bb5<\/h3>\n<\/p>\n
\u5728\u89e3\u51b3\u95ee\u9898\u65f6\uff0c\u53ef\u4ee5\u5c1d\u8bd5\u8fd0\u884c\u4ee5\u4e0b\u793a\u4f8b\u4ee3\u7801\u6bb5\u6765\u68c0\u67e5 TensorBoardX \u662f\u5426\u6b63\u5e38\u5de5\u4f5c\uff1a  <\/p>\n
import torch<\/p>\r\n
from tensorboardX import SummaryWriter<\/p>\r\n\r\n
writer = SummaryWriter('logs')<\/p>\r\n
x = torch.randn(100, 100)<\/p>\r\n
writer.add_graph(model, (x,))<\/p>\r\n
writer.close()<\/code><\/pre>\n<\/p>\n
\u5982\u679c\u8fd0\u884c\u8be5\u793a\u4f8b\u65f6\u6ca1\u6709\u62a5\u9519\uff0c\u8bf4\u660e\u5b89\u88c5\u548c\u4f7f\u7528\u90fd\u6b63\u5e38\u3002\u5982\u679c\u51fa\u73b0\u9519\u8bef\u4fe1\u606f\uff0c\u9700\u8981\u6839\u636e\u9519\u8bef\u4fe1\u606f\u8fdb\u4e00\u6b65\u5206\u6790\u539f\u56e0\u3002<\/p>\n<\/p>\n
\n
5. \u5982\u4f55\u9a8c\u8bc1 TensorBoardX \u5b89\u88c5\u662f\u5426\u6210\u529f\uff1f<\/h3>\n<\/p>\n
\u6211\u5e94\u8be5\u5982\u4f55\u786e\u8ba4 TensorBoardX \u5df2\u88ab\u6210\u529f\u5b89\u88c5\uff1f<\/b>  <\/p>\n
\u53ef\u4ee5\u901a\u8fc7\u5c1d\u8bd5\u8fd0\u884c\u4e00\u4e9b\u57fa\u7840\u7684\u4ee3\u7801\uff0c\u6bd4\u5982\u521b\u5efa\u4e00\u4e2a SummaryWriter \u5b9e\u4f8b\u5e76\u8bb0\u5f55\u4e00\u4e2a\u7b80\u5355\u7684\u6807\u91cf\u3002\u8fd0\u884c\u4ee5\u4e0b\u4ee3\u7801\uff1a  <\/p>\n
from tensorboardX import SummaryWriter<\/p>\r\n\r\n
writer = SummaryWriter('example_logs')<\/p>\r\n
writer.add_scalar('example_scalar', 1, 0)<\/p>\r\n
writer.close()<\/code><\/pre>\n<\/p>\n
\u5982\u679c\u6ca1\u6709\u62a5\u9519\uff0c\u5e76\u4e14\u5728\u6307\u5b9a\u8def\u5f84\u4e0b\u751f\u6210\u4e86\u65e5\u5fd7\u6587\u4ef6\uff0c\u5373\u53ef\u786e\u8ba4\u5df2\u6210\u529f\u5b89\u88c5\u3002<\/p>\n<\/p>\n
\n
6. \u662f\u5426\u6709\u5176\u4ed6\u53ef\u80fd\u5f15\u8d77\u6b64\u9519\u8bef\u7684\u539f\u56e0\uff1f<\/h3>\n<\/p>\n
\u9664\u4e86\u4f9d\u8d56\u9879\u5916\uff0c\u8fd8\u6709\u54ea\u4e9b\u56e0\u7d20\u53ef\u80fd\u5bfc\u81f4\u6b64\u9519\u8bef\u51fa\u73b0\uff1f<\/b>  <\/p>\n
\u6b64\u5916\uff0c\u53ef\u80fd\u7684\u95ee\u9898\u8fd8\u5305\u62ec Python \u7248\u672c\u662f\u5426\u517c\u5bb9\u3002TensorBoardX \u5bf9 Python 3.6 \u4ee5\u4e0a\u7684\u7248\u672c\u6709\u4e00\u5b9a\u8981\u6c42\uff0c\u56e0\u6b64\u5982\u679c\u4f7f\u7528\u7684\u662f\u8f83\u65e7\u7684 Python \u7248\u672c\uff0c\u9700\u8981\u8003\u8651\u5347\u7ea7\u3002\u6b64\u5916\uff0c\u53ef\u80fd\u662f\u4e0d\u540c\u7684\u73af\u5883\u914d\u7f6e\u5f15\u8d77\u7684\uff0c\u6bd4\u5982\u4f7f\u7528 Anaconda \u65f6\u9700\u8981\u786e\u4fdd\u5728\u6fc0\u6d3b\u7684\u73af\u5883\u4e2d\u6b63\u786e\u5b89\u88c5\u5305\u3002<\/p>\n<\/p>\n
\n
7. \u5982\u679c\u7ee7\u7eed\u51fa\u73b0\u95ee\u9898\uff0c\u6211\u8be5\u5411\u54ea\u91cc\u5bfb\u6c42\u5e2e\u52a9\uff1f<\/h3>\n<\/p>\n
\u5f53\u9047\u5230\u4ece tensorboardx \u5bfc\u5165\u9519\u8bef\u4e14\u627e\u4e0d\u5230\u89e3\u51b3\u65b9\u6848\u65f6\uff0c\u6211\u5e94\u8be5\u600e\u4e48\u529e\uff1f<\/b>  <\/p>\n
\u53ef\u4ee5\u8bbf\u95ee TensorBoardX \u7684 GitHub \u9875\u9762\uff0c\u67e5\u770b\u662f\u5426\u6709\u4eba\u9047\u5230\u8fc7\u7c7b\u4f3c\u7684\u95ee\u9898\u5e76\u627e\u5230\u4e86\u89e3\u51b3\u65b9\u6848\u3002\u793e\u533a\u53cd\u9988\u4f1a\u5f88\u6709\u5e2e\u52a9\uff0c\u4e5f\u53ef\u4ee5\u76f4\u63a5\u5728 issue \u533a\u57df\u63d0\u95ee\u3002\u5728\u8be2\u95ee\u65f6\uff0c\u8bb0\u5f97\u63d0\u4f9b\u8be6\u7ec6\u4fe1\u606f\uff0c\u5982\u9519\u8bef\u4fe1\u606f\u3001\u60a8\u4f7f\u7528\u7684\u73af\u5883\uff0c\u4ee5\u53ca\u5c1d\u8bd5\u8fc7\u7684\u89e3\u51b3\u65b9\u6848\uff0c\u4ee5\u4fbf\u5176\u4ed6\u5f00\u53d1\u8005\u53ef\u4ee5\u66f4\u6709\u6548\u5730\u5e2e\u52a9\u60a8\u89e3\u51b3\u95ee\u9898\u3002<\/p><\/p>\n","protected":false},"excerpt":{"rendered":"
1. from tensorboardx import summarywriter \u62a5\u9519\u6982\u8ff0 \u9047\u5230 “from tensorboardx import summarywriter” \u62a5\u9519\u65f6\uff0c\u901a\u5e38\u662f\u56e0\u4e3a TensorB…<\/p>\n","protected":false},"author":1,"featured_media":544,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[288,286,287],"topic":[],"class_list":["post-543","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-jiushu","tag-import","tag-tensorboardx","tag-287"],"_links":{"self":[{"href":"https:\/\/www.idcyz.com\/blog\/wp-json\/wp\/v2\/posts\/543","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.idcyz.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.idcyz.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.idcyz.com\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.idcyz.com\/blog\/wp-json\/wp\/v2\/comments?post=543"}],"version-history":[{"count":1,"href":"https:\/\/www.idcyz.com\/blog\/wp-json\/wp\/v2\/posts\/543\/revisions"}],"predecessor-version":[{"id":546,"href":"https:\/\/www.idcyz.com\/blog\/wp-json\/wp\/v2\/posts\/543\/revisions\/546"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.idcyz.com\/blog\/wp-json\/wp\/v2\/media\/544"}],"wp:attachment":[{"href":"https:\/\/www.idcyz.com\/blog\/wp-json\/wp\/v2\/media?parent=543"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.idcyz.com\/blog\/wp-json\/wp\/v2\/categories?post=543"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.idcyz.com\/blog\/wp-json\/wp\/v2\/tags?post=543"},{"taxonomy":"topic","embeddable":true,"href":"https:\/\/www.idcyz.com\/blog\/wp-json\/wp\/v2\/topic?post=543"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}