tensorflow

Tensorflow-IO Dataset input pipeline with very large HDF5 files

由 佐手、 提交于 2021-02-10 12:18:10
问题 I have very big training (30Gb) files. Since all the data does not fit in my available RAM, I want to read the data by batch. I saw that there is Tensorflow-io package which implemented a way to read HDF5 into Tensorflow this way thanks to the function tfio.IODataset.from_hdf5() Then, since tf.keras.model.fit() takes a tf.data.Dataset as input containing both samples and targets, I need to zip my X and Y together and then use .batch and .prefetch to load in memory just the necessary data. For

TensorFlow function to check whether a value is in given range?

由 ε祈祈猫儿з 提交于 2021-02-10 12:14:41
问题 I know there is tf.greater(x,y) which will return true if x > y (element-wise). Is there a function that returns true if lower_bound < x < upper_bound (element-wise) for a tensor x? 回答1: There's not a specific function for that, but you can use a combination of tf.greater , tf.less , and tf.logical_and to get the same result. lower_tensor = tf.greater(x, lower) upper_tensor = tf.less(x, upper) in_range = tf.logical_and(lower_tensor, upper_tensor) 来源： https://stackoverflow.com/questions

TensorFlow function to check whether a value is in given range?

由 丶灬走出姿态 提交于 2021-02-10 12:14:30
问题 I know there is tf.greater(x,y) which will return true if x > y (element-wise). Is there a function that returns true if lower_bound < x < upper_bound (element-wise) for a tensor x? 回答1: There's not a specific function for that, but you can use a combination of tf.greater , tf.less , and tf.logical_and to get the same result. lower_tensor = tf.greater(x, lower) upper_tensor = tf.less(x, upper) in_range = tf.logical_and(lower_tensor, upper_tensor) 来源： https://stackoverflow.com/questions

How can we use lbfgs_minimize in TensorFlow 2.0

由 馋奶兔 提交于 2021-02-10 12:00:33
问题 I have been unable to reproduce this example from tensorflow having tensorflow 2.0 installed. This is the original snippet: # A high-dimensional quadratic bowl. ndims = 60 minimum = np.ones([ndims], dtype='float64') scales = np.arange(ndims, dtype='float64') + 1.0 # The objective function and the gradient. def quadratic(x): value = tf.reduce_sum(scales * (x - minimum) ** 2) return value, tf.gradients(value, x)[0] start = np.arange(ndims, 0, -1, dtype='float64') optim_results = tfp.optimizer

How can we use lbfgs_minimize in TensorFlow 2.0

由 余生长醉 提交于 2021-02-10 11:58:16
问题 I have been unable to reproduce this example from tensorflow having tensorflow 2.0 installed. This is the original snippet: # A high-dimensional quadratic bowl. ndims = 60 minimum = np.ones([ndims], dtype='float64') scales = np.arange(ndims, dtype='float64') + 1.0 # The objective function and the gradient. def quadratic(x): value = tf.reduce_sum(scales * (x - minimum) ** 2) return value, tf.gradients(value, x)[0] start = np.arange(ndims, 0, -1, dtype='float64') optim_results = tfp.optimizer

用 tensorflow实现DeepFM

由 删除回忆录丶 提交于 2021-02-10 11:27:21
http://www.fabwrite.com/deepfm 文章 DeepFM: A Factorization-Machine based Neural Network for CTR Prediction 介绍了一种深度学习模型，以实现点击率预估。用 tensorflow 试着写了 DeepFM,见 https://github.com/zgw21cn/DeepFM 。 1. FNN、PNN、wide&deep等此前几种深度模型 见下图。 (1) FNN，见图左边。用 FM 预训练embedding layer，然后DNN训练。作者认为有两点局限：embedding layer 的参数会受到 FM 的影响；预训练计算量大，效率问题。同时 FNN仅能捕捉高阶特征，相比之下，DeepFM 无需预训练，且能捕捉高阶和低阶特征。 (2) PNN，见图中间。为了捕捉高阶特征间的组合作用，PNN在 embedding layer 和首层 hidden layer 间引入了 product layer。根据 product 类型的不同，有多种类型的 product layer。product 类型有内积、外积、内积和外积的混合。 PNN的局限在于，外积由于是近似计算会丢失信息而比内积不可靠；内积更可靠，但是仍会计算量大。类似于 FNN，PNN忽略了低阶特征。 (3) Wide&Deep

毕业设计之「神经网络与深度学习概述」 （一）

由 ε祈祈猫儿з 提交于 2021-02-10 09:31:53
阅读本文大概需要 18 分钟。 前言 本章节属于我 毕业设计系列推文 中的基础篇内容，是对 神经网络与深度学习基础 的一个概述，共包括两节内容（分别是毕业设计之「神经网络与深度学习概述」 （一）和 毕业设计之「神经网络与深度学习概述」（二） ），我会分成两次推文进行推送，本文是「神经网络与深度学习概述」的第一篇，主要内容将围绕 神经网络的基础 进行展开。 需要注意的一点，由于本小节推文中涉及较多的公式、特殊符号、表格等，从而造成在排版上有较大难度，因此我以截图的形式进行本次推文的推送，各位读者请见谅。 神经网络基础 - End - 往期推荐 🔗 近期推文 TensorFlow 2.0 概述 毕业设计之「神经网络与深度学习概述」（二） 毕业设计之「测试实验及结果分析」（二） 微信拍一拍效果的原理分析及代码实现 Python 开发植物大战僵尸游戏 数据结构与算法系列 数据结构与算法-1 ：旋转图像 数据结构与算法 -2 ：罗马数字与整数的相互转换 数据结构与算法 -3 ：复原IP地址 数据结构与算法 -4、5 ：两数相加&&两数之和 毕业设计系列 TensorFlow环境搭建 毕业设计（基于TensorFlow的深度学习与研究）之文献翻译篇 TensorFlow 2.0 概述 毕业设计（基于TensorFlow的深度学习与研究）之核心篇CNN-AlexNet详解 毕业设计之

How to get/set weights for a supervised model in tensorflow.js?

由 混江龙づ霸主 提交于 2021-02-10 07:53:15
问题 I'd like to change the weights of a supervised model but I get the same exact result after changing the weights. What am I doing wrong? const model = tf.sequential(); model.add(tf.layers.dense({...})); model.add(tf.layers.dense({...})); model.add(tf.layers.dense({...})); model.compile({...}); model.fit({}); const result1 = model.predict(tf.tensor2d(...)).dataSync(); const newWeights = []; model.layers.map((layer, i) => { newWeights[i] = [] const weights = layer.getWeights(); newWeights[i][0]

How can I use GPU for running a tflite model (*.tflite) using tf.lite.Interpreter (in python)?

由 大城市里の小女人 提交于 2021-02-10 07:25:36
问题 I have converted a tensorflow inference graph to tflite model file (*.tflite), according to instructions from https://www.tensorflow.org/lite/convert. I tested the tflite model on my GPU server, which has 4 Nvidia TITAN GPUs. I used the tf.lite.Interpreter to load and run tflite model file. It works as the former tensorflow graph, however, the problem is that the inference became too slow. When I checked out the reason, I found that the GPU utilization is simply 0% when tf.lite.Interpreter is

hickle NameError: name 'file' is not defined

由 柔情痞子 提交于 2021-02-10 07:25:29
问题 please explain to me why this error getting for me. I have install python 3.5 in ubuntu import os import hickle as hkl import numpy as np array_obj = np.ones(32768, dtype='float32') hkl.dump(array_obj, 'test.hkl', mode='w') the result is Traceback (most recent call last): File "<stdin>", line 1, in <module> File "C:\Users\Admin\AppData\Local\Programs\Python\Python35\lib\site-packages\hickle.py", line 308, in dump h5f = file_opener(file_obj, mode, track_times) File "C:\Users\Admin\AppData
订阅 tensorflow