技术博客tensorflow与教程covnets in.pptxVIP

下载本文档

0
0
约6.58千字
约 43页
2022-05-03 发布于北京
举报
版权申诉

技术博客tensorflow与教程covnets in.pptx

1、本文档共43页，可阅读全部内容。
2、原创力文档（book118）网站文档一经付费（服务费），不意味着购买了该文档的版权，仅供个人/单位学习、研究之用，不得用于商业用途，未经授权，严禁复制、发行、汇编、翻译或者网络传播等，侵权必究。
3、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。如您付费，意味着您自己接受本站规则且自行承担风险，本站不退款、不进行额外附加服务；查看《如何避免下载的几个坑》。如果您已付费下载过本站文档，您可以点击这里二次下载。
4、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等，请点击“版权申诉”（推荐），也可以打举报电话：400-050-0827(电话支持时间：9:00-18:30)。
5、该文档为VIP文档，如果想要下载，成为VIP会员后，下载免费。
6、成为VIP后，下载本文档将扣除1次下载权益。下载后，不支持退款、换文档。如有疑问请联系我们。
7、成为VIP后，您将拥有八大权益，权益包括：VIP文档下载权益、阅读免打扰、文档格式转换、高级专利检索、专属身份标志、高级客服、多端互通、版权登记。
8、VIP文档为合作方或网友上传，每下载1次，网站将根据用户上传文档的质量评分、类型等，对文档贡献者给予高额补贴、流量扶持。如果你也想贡献VIP文档。上传文档

Convnets in TensorFlowCS 20: TensorFlow for Deep Learning ResearchLecture 72/7/2017AgendaConvolutions without trainingConvnet with MNIST!!!tf.layersUnderstanding convolutionsConvolutions in math and physicsa function derived from two given functions by integration that expresses how the shape of one is modified by the otherConvolutions in math and physicsBrian Amberg derivative work (Wikipedia)Convolutions in math and physicsHow an input is transformed by a kernel* *also called filter/feature mapConvolutions in machine learningWe can use one single convolutional layer to modify a certain imageConvolutions in machine learning Kernel for blurring0.06250.1250.06250.1250.250.1250.06250.1250.0625Matrix multiplication of this kernel with a 3 x 3 patch of an image is a weighted sum of neighboring pixels= blurring effectConvolution without trainingKernel for blurring0.06250.1250.06250.1250.250.1250.06250.1250.0625tf.nn.conv2dinputoutputConvolutions in TensorFlowWe can use one single convolutional layer to modify a certain imagetf.nn.conv2d( input, filter, strides, padding, use_cudnn_on_gpu=True, data_format=NHWC, dilations=[1, 1, 1, 1], name=None)Convolutions in TensorFlowWe can use one single convolutional layer to modify a certain imagetf.nn.conv2d( input, Batch size (N) x Height (H) x Width (W) x Channels (C) filter, Height x Width x Input Channels x Output Channels strides, 4 element 1-D tensor, strides in each direction padding, SAME or VALID use_cudnn_on_gpu=True, data_format=NHWC, dilations=[1, 1, 1, 1], name=None)Convolutions in TensorFlowWe can use one single convolutional layer to modify a certain imagetf.nn.conv2d( image, kernel, strides=[1, 3, 3, 1], padding=SAME,)Some basic kernelsinputsharpenedgetop sobelembossSee kernels.py and 07_run_kernels.pySome basic kernelsinputsharpenedgetop sobelembossConvolutions in machine learningDon’t hard-code the values of your kernels.Learn the optimal kernels through training!ConvNet with MNISTModelFully + softmaxFully +