keras-vis：用于keras的神经网络可视化工具包
Keras可视化工具包
keras-vis是用于可视化和调试已训练的keras神经网络模型的高级工具包。 当前支持的可视化包括：
激活最大化
显着图
类激活图
默认情况下，所有可视化都支持N维图像输入。 即，它推广到模型的N维图像输入。
该工具包通过干净，易于使用和可扩展的界面将上述所有问题归纳为能量最小化问题。 与theano和tensorflow后端兼容，具有“ channels_first”，“ channels_last”数据格式。
快速链接
阅读位于的文档。 日语版为 。
加入闲暇来提问/讨论。
我们正在中跟踪新功能/任务。 如果您愿意帮助我们并提交PR，将非常乐意。
入门
在图像反向传播问题中，目标是生成使某些损失函数最小化的输入图像。 设置图像反向传播问题很容易。
定义加权损失函数
在中定义了各种有用的损失函数。 可以通过实现来定义自定义损失函数。
from vis . losses import ActivationMaximization
from vis . regularizers import TotalVariation , LPNorm
fil
文件下载
资源详情
[{"title":"（ 72 个子文件 43.42MB ） keras-vis：用于keras的神经网络可视化工具包","children":[{"title":"keras-vis-master","children":[{"title":"MANIFEST.in <span style='color:#111;'> 44B </span>","children":null,"spread":false},{"title":".gitattributes <span style='color:#111;'> 61B </span>","children":null,"spread":false},{"title":"ISSUE_TEMPLATE.md <span style='color:#111;'> 678B </span>","children":null,"spread":false},{"title":"images","children":[{"title":"opt_progress.gif <span style='color:#111;'> 13.48MB </span>","children":null,"spread":false},{"title":"attention_vis","children":[{"title":"grad-cam.png <span style='color:#111;'> 529.25KB </span>","children":null,"spread":false},{"title":"cover.png <span style='color:#111;'> 532.38KB </span>","children":null,"spread":false},{"title":"saliency_map.png <span style='color:#111;'> 118.96KB </span>","children":null,"spread":false}],"spread":true},{"title":"conv_vis","children":[{"title":"block4_conv3_filters.jpg <span style='color:#111;'> 125.42KB </span>","children":null,"spread":false},{"title":"block5_conv3_filters_no_tv.jpg <span style='color:#111;'> 183.95KB </span>","children":null,"spread":false},{"title":"block1_conv2_filters.jpg <span style='color:#111;'> 152.26KB </span>","children":null,"spread":false},{"title":"block3_conv3_filters.jpg <span style='color:#111;'> 144.40KB </span>","children":null,"spread":false},{"title":"filter_67.png <span style='color:#111;'> 73.91KB </span>","children":null,"spread":false},{"title":"cover.jpg <span style='color:#111;'> 125.42KB </span>","children":null,"spread":false},{"title":"block5_conv3_filters.jpg <span style='color:#111;'> 80.32KB </span>","children":null,"spread":false},{"title":"block2_conv2_filters.jpg <span style='color:#111;'> 109.94KB </span>","children":null,"spread":false}],"spread":true},{"title":"dense_vis","children":[{"title":"random_imagenet.png <span style='color:#111;'> 819.39KB </span>","children":null,"spread":false},{"title":"random_imagenet_no_tv.png <span style='color:#111;'> 1001.92KB </span>","children":null,"spread":false},{"title":"cover.png <span style='color:#111;'> 219.74KB </span>","children":null,"spread":false},{"title":"ouzel_vis.png <span style='color:#111;'> 219.74KB </span>","children":null,"spread":false}],"spread":true}],"spread":true},{"title":"examples","children":[{"title":"vggnet","children":[{"title":"attention.ipynb <span style='color:#111;'> 1.96MB </span>","children":null,"spread":false},{"title":"activation_maximization.ipynb <span style='color:#111;'> 23.66MB </span>","children":null,"spread":false},{"title":"images","children":[{"title":"ouzel1.jpg <span style='color:#111;'> 127.92KB </span>","children":null,"spread":false},{"title":"ouzel2.jpg <span style='color:#111;'> 143.96KB </span>","children":null,"spread":false}],"spread":true}],"spread":true},{"title":"resnet","children":[{"title":"attention.ipynb <span style='color:#111;'> 1.98MB </span>","children":null,"spread":false}],"spread":true},{"title":"mnist","children":[{"title":"attention.ipynb <span style='color:#111;'> 377.97KB </span>","children":null,"spread":false},{"title":"activation_maximization.ipynb <span style='color:#111;'> 411.46KB </span>","children":null,"spread":false}],"spread":true}],"spread":true},{"title":"CONTRIBUTING.md <span style='color:#111;'> 1.42KB </span>","children":null,"spread":false},{"title":".travis.yml <span style='color:#111;'> 1.98KB </span>","children":null,"spread":false},{"title":"pytest.ini <span style='color:#111;'> 440B </span>","children":null,"spread":false},{"title":"vis","children":[{"title":"callbacks.py <span style='color:#111;'> 2.65KB </span>","children":null,"spread":false},{"title":"utils","children":[{"title":"utils.py <span style='color:#111;'> 10.90KB </span>","children":null,"spread":false},{"title":"__init__.py <span style='color:#111;'> 0B </span>","children":null,"spread":false},{"title":"test_utils.py <span style='color:#111;'> 1.21KB </span>","children":null,"spread":false}],"spread":true},{"title":"input_modifiers.py <span style='color:#111;'> 3.54KB </span>","children":null,"spread":false},{"title":"grad_modifiers.py <span style='color:#111;'> 1.22KB </span>","children":null,"spread":false},{"title":"regularizers.py <span style='color:#111;'> 4.09KB </span>","children":null,"spread":false},{"title":"backend","children":[{"title":"theano_backend.py <span style='color:#111;'> 719B </span>","children":null,"spread":false},{"title":"__init__.py <span style='color:#111;'> 281B </span>","children":null,"spread":false},{"title":"tensorflow_backend.py <span style='color:#111;'> 4.30KB </span>","children":null,"spread":false}],"spread":false},{"title":"__init__.py <span style='color:#111;'> 0B </span>","children":null,"spread":false},{"title":"optimizer.py <span style='color:#111;'> 8.40KB </span>","children":null,"spread":false},{"title":"losses.py <span style='color:#111;'> 3.46KB </span>","children":null,"spread":false},{"title":"backprop_modifiers.py <span style='color:#111;'> 1.09KB </span>","children":null,"spread":false},{"title":"visualization","children":[{"title":"activation_maximization.py <span style='color:#111;'> 6.09KB </span>","children":null,"spread":false},{"title":"saliency.py <span style='color:#111;'> 13.31KB </span>","children":null,"spread":false},{"title":"__init__.py <span style='color:#111;'> 1.64KB </span>","children":null,"spread":false}],"spread":false}],"spread":false},{"title":"LICENSE <span style='color:#111;'> 1.06KB </span>","children":null,"spread":false},{"title":"setup.cfg <span style='color:#111;'> 67B </span>","children":null,"spread":false},{"title":"setup.py <span style='color:#111;'> 805B </span>","children":null,"spread":false},{"title":"README.md <span style='color:#111;'> 6.94KB </span>","children":null,"spread":false},{"title":"resources","children":[{"title":"imagenet_class_index.json <span style='color:#111;'> 34.53KB </span>","children":null,"spread":false}],"spread":true},{"title":"docs","children":[{"title":"update_docs.py <span style='color:#111;'> 1.11KB </span>","children":null,"spread":false},{"title":"md_autogen.py <span style='color:#111;'> 13.73KB </span>","children":null,"spread":false},{"title":"__init__.py <span style='color:#111;'> 0B </span>","children":null,"spread":false},{"title":"mkdocs.yml <span style='color:#111;'> 1.11KB </span>","children":null,"spread":false},{"title":"README.md <span style='color:#111;'> 518B </span>","children":null,"spread":false},{"title":"templates","children":[{"title":"visualizations","children":[{"title":"saliency.md <span style='color:#111;'> 5.14KB </span>","children":null,"spread":false},{"title":"activation_maximization.md <span style='color:#111;'> 4.91KB </span>","children":null,"spread":false},{"title":"class_activation_maps.md <span style='color:#111;'> 1.51KB </span>","children":null,"spread":false}],"spread":false},{"title":"css","children":[{"title":"extras.css <span style='color:#111;'> 2.42KB </span>","children":null,"spread":false}],"spread":false}],"spread":false}],"spread":false},{"title":"tests","children":[{"title":"vis","children":[{"title":"utils","children":[{"title":"test_utils.py <span style='color:#111;'> 1.05KB </span>","children":null,"spread":false}],"spread":false},{"title":"backend","children":[{"title":"test_backend.py <span style='color:#111;'> 5.98KB </span>","children":null,"spread":false}],"spread":false},{"title":"test_optimizer.py <span style='color:#111;'> 1.51KB </span>","children":null,"spread":false},{"title":"visualization","children":[{"title":"test_saliency.py <span style='color:#111;'> 1.88KB </span>","children":null,"spread":false}],"spread":false}],"spread":false}],"spread":false},{"title":"applications","children":[{"title":"self_driving","children":[{"title":"weights.hdf5 <span style='color:#111;'> 4.69MB </span>","children":null,"spread":false},{"title":"images","children":[{"title":"left.png <span style='color:#111;'> 13.55KB </span>","children":null,"spread":false},{"title":"blank.png <span style='color:#111;'> 7.20KB </span>","children":null,"spread":false},{"title":"straight.png <span style='color:#111;'> 18.53KB </span>","children":null,"spread":false}],"spread":false},{"title":"model.py <span style='color:#111;'> 845B </span>","children":null,"spread":false},{"title":"README.md <span style='color:#111;'> 271B </span>","children":null,"spread":false},{"title":"visualize_attention.ipynb <span style='color:#111;'> 276.64KB </span>","children":null,"spread":false}],"spread":false}],"spread":false},{"title":".gitignore <span style='color:#111;'> 151B </span>","children":null,"spread":false}],"spread":false}],"spread":true}]
评论信息
其他资源
- JENA教程(包括生成RDF和写本体到MySQL中)
- 基于STM32F103RCT6的无源蜂鸣器音乐播放（生日快乐歌）部分修改版
- 双目已知图像坐标求空间坐标opencv
- 竞争性自适应重加权
- stm32f2xx 库函数，及其例程，官网下载
- Using_R_for_Introductory_Econometrics_Wooldridge
- 2021年中国各省行政区高德adcode表(含有香港澳门).zip
- cad插件数字相加.lsp
- Chinese.unitypackage
- HTC shift 救砖头
- 自做项目——打地鼠C#
- unity-ui-examples：Unity Unity的UI示例集合
- 基于VMware的桌面虚拟化设计与应用
- ESQL/C 编程指南
- 数字频率计（EDA课程设计）
- Android中ListView上拉加载和下拉刷新
- butterknife7.0.1
- 多块板子利用485总线 采用MODBUS协议实现开关量采集与控制采集
- CTPTrader_穿透式Test.zip
- 用POCS方法对图像进行超分辨率重构
- UiAutomatorviewer 源码 Android 8.0
免责申明
【只为小站】的资源来自网友分享，仅供学习研究，请务必在下载后24小时内给予删除，不得用于其他任何用途，否则后果自负。基于互联网的特殊性，【只为小站】 无法对用户传输的作品、信息、内容的权属或合法性、合规性、真实性、科学性、完整权、有效性等进行实质审查；无论 【只为小站】 经营者是否已进行审查，用户均应自行承担因其传输的作品、信息、内容而可能或已经产生的侵权或权属纠纷等法律责任。
本站所有资源不代表本站的观点或立场，基于网友分享，根据中国法律《信息网络传播权保护条例》第二十二条之规定，若资源存在侵权或相关问题请联系本站客服人员，zhiweidada#qq.com，请把#换成@，本站将给予最大的支持与配合，做到及时反馈和处理。关于更多版权及免责申明参见 版权及免责申明
本站所有资源不代表本站的观点或立场，基于网友分享，根据中国法律《信息网络传播权保护条例》第二十二条之规定，若资源存在侵权或相关问题请联系本站客服人员，zhiweidada#qq.com，请把#换成@，本站将给予最大的支持与配合，做到及时反馈和处理。关于更多版权及免责申明参见 版权及免责申明