光于前裕于后的博客

Keras优点多多，搭建模型快实现简单，不过建议了解神经网络基本原理后再用。 backend建议用TensorFlow，比Theano快很多。from sklearn.datasets import load_irisfrom sklearn.model_selection import train_test_splitimport kerasfrom keras.mo...

Fashion MNIST Dataset有T恤、短靴等10个类别图像，每张图片为28*28像素的灰阶图像共70000张图像，60K当作训练集，10K当作测试集# !pip install -U tensorflow_datasets# !pip install -U -i https://pypi.tuna.tsinghua.edu.cn/simple tensorflow-gpu==...

https://tensorflow.google.cn/install/source#linux

import tensorflow as tf# 设置日志输出等级tf.logging.set_verbosity(tf.logging.ERROR)import numpy as np# 输入摄氏度，输出华氏度input_ = np.array([-40, -10, 0, 8, 15, 22, 38], dtype=float)output = np.array([-40, 14...

import tensorflow as tftf.enable_eager_execution()from tensorflow import kerasimport numpy as npimport matplotlib.pyplot as pltimport osimport subprocessprint(tf.__version__)1.12.0fashio...

import tensorflow as tfF:\Anaconda3\envs\tensorflow-gpu\lib\site-packages\h5py\__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is depreca...

本文为Udacity优达学城深度学习课程笔记第三篇，使用图像增强对猫狗图像进行分类。课程地址：https://classroom.udacity.com/courses/ud187最终目的是训练CNN模型，使其能识别上面的小狗狗是小狗狗。可能遇到的问题有过拟合、输入图像的尺寸不一，不过不用担心，本文会一一解决。为了保持笔记与课程内容一致，代码没有进行修改，如需提升识别准确率可自行对模型进行修...