书栈网 · BookStack 本次搜索耗时 0.041 秒,为您找到 1338 个相关结果.
  • Epilogue:A Story Between a Foodie and Cuisine

    Epilogue: A Story Between a Foodie and Cuisine Epilogue: A Story Between a Foodie and Cuisine “How to eat TensorFlow2 in 30 days?” has been finished and well concluded, and thi...
  • Compatibility with Various SQL Engines

    Compatibility with Various SQL Engines Data Operations in Go Data Retrieval Metadata Retrieval Prepare Prediction Table Translate Database Column Type to TensorFlow Feature Colu...
  • 5-3,激活函数activation

    5-3,激活函数activation 一,常用激活函数 二,在模型中使用激活函数 5-3,激活函数activation 激活函数在深度学习中扮演着非常重要的角色,它给网络赋予了非线性,从而使得神经网络能够拟合任意复杂的函数。 如果没有激活函数,无论多复杂的网络,都等价于单一的线性变换,无法对非线性函数进行拟合。 目前,深度学习中最流行的激活函...
  • TensorBoard:训练过程可视化

    TensorBoard:训练过程可视化 TensorBoard:训练过程可视化 有时,你希望查看模型训练过程中各个参数的变化情况(例如损失函数loss的值)。虽然可以通过命令行输出来查看,但有时显得不够直观。而TensorBoard就是一个能够帮助我们将训练过程可视化的工具。 首先在代码目录下建立一个文件夹(如 ./tensorboard )存放...
  • Callbacks

    Usage of callbacks Callback BaseLogger TerminateOnNaN ProgbarLogger History ModelCheckpoint EarlyStopping RemoteMonitor LearningRateScheduler ReduceLROnPlateau CSVLogger ...
  • 后端 Backend

    Keras 后端 什么是 「后端」? 从一个后端切换到另一个后端 keras.json 详细配置 使用抽象 Keras 后端编写新代码 后端函数 epsilon set_epsilon floatx set_floatx cast_to_floatx image_data_format set_image_data_format res...
  • 5-5 losses

    5-5 losses 1. Loss Function and Regularization Term 2. Pre-defined Loss Function 3. Customized Loss Function 5-5 losses In general, the target function in supervised learnin...
  • Usage

    Usage Training an Agent Single-threaded Algorithms Multi-threaded Algorithms Multi-Node Algorithms Evaluating an Agent Playing with the Environment as a Human Learning Throug...
  • FAQ

    Frequently Asked Questions (FAQ) Can I use ml5.js in the p5 web editor? Can I use ml5.js with node.js? Can I contribute? How can I contribute? Frequently Asked Questions (F...
  • 缓存冻结层

    缓存冻结层 由于冻结层不会改变,因此可以为每个训练实例缓存最上面的冻结层的输出。 由于训练贯穿整个数据集很多次,这将给你一个巨大的速度提升,因为每个训练实例只需要经过一次冻结层(而不是每个迭代一次)。 例如,你可以先运行整个训练集(假设你有足够的内存): hidden2_outputs = sess . run ( hidden2 , feed_...