![](https://img-blog.csdnimg.cn/20201014180756757.png?x-oss-process=image/resize,m_fixed,h_64,w_64)
tensorflow
zhangztSky
这个作者很懒,什么都没留下…
展开
-
Python之(axis=0 、axis=1、axis=-1)区分
https://www.cnblogs.com/rrttp/p/8028421.htmlhttps://blog.csdn.net/fangjian1204/article/details/53055219https://blog.csdn.net/weixin_42059276/article/details/106837795原创 2020-12-25 11:52:51 · 839 阅读 · 0 评论 -
imdb数据使用普通的dnn+embeding做分类
import matplotlib as mplimport matplotlib.pyplot as pltimport numpy as npimport sklearnimport pandas as pdimport osimport sysimport timeimport tensorflow as tffrom tensorflow import kerasprint(tf.__version__)print(sys.version_info)for module原创 2020-12-22 17:38:27 · 168 阅读 · 0 评论 -
python numpy reshape
https://blog.csdn.net/qq_29831163/article/details/90112000原创 2020-12-21 17:45:29 · 108 阅读 · 0 评论 -
tensorflow feature_column 讲解
官网写的很清楚我自己跑了一遍,贴个代码import numpy as npimport pandas as pdimport tensorflow as tffrom tensorflow import feature_columnfrom tensorflow.keras import layersfrom sklearn.model_selection import train_test_splitimport pathlibdataset_url = 'http://st原创 2020-12-21 14:56:44 · 330 阅读 · 0 评论 -
MaxPooling1D和GlobalMaxPooling1D的区别
import tensorflow as tffrom tensorflow import kerasinput_shape = (2, 3, 4)x = tf.random.normal(input_shape)print(x)y=keras.layers.GlobalMaxPool1D()(x)print("*"*20)print(y)''' """Global average pooling operation for temporal data. Examples:原创 2020-12-21 11:23:40 · 1727 阅读 · 0 评论 -
一维卷积二维卷积三维卷积
https://www.cnblogs.com/szxspark/p/8445406.html转载 2020-12-19 18:16:19 · 172 阅读 · 1 评论 -
tensorflow dataset API
import matplotlib as mplimport matplotlib.pyplot as pltimport numpy as npimport sklearnimport pandas as pdimport osimport sysimport timeimport tensorflow as tffrom tensorflow import keras## (x,y), (x_test, y_test) = keras.datasets.cifar100.load原创 2020-12-18 16:49:58 · 123 阅读 · 1 评论 -
tensorflow2 @tf.function
tensorflow2 @tf.function 文档如果是直接用keras 高层API的话应该用不到这玩意。原创 2020-12-17 16:30:55 · 217 阅读 · 0 评论 -
TensorFlow
s = tf.SparseTensor(indices = [[0, 1], [1, 0], [2, 3]], # 3* 4 tensor有值的地方的坐标 values = [1., 2., 3.], dense_shape = [3, 4])print(s)print(tf.sparse.to_dense(s))print("*"*20)s2 = s * 2.0print(s2)try: s3 = .原创 2020-12-17 15:33:58 · 84 阅读 · 0 评论 -
TensorFlow Ragged Tensors
https://tensorflow.google.cn/guide/ragged_tensor?hl=zh_cn原创 2020-12-17 15:21:37 · 176 阅读 · 0 评论 -
深度学习框架中的静态图vs动态图
https://www.jianshu.com/p/505e1e0142c1原创 2020-12-17 13:53:29 · 509 阅读 · 0 评论 -
计算图的理解
https://yunyaniu.blog.csdn.net/article/details/88880171转载 2020-12-17 11:13:21 · 183 阅读 · 0 评论 -
tensorflow 添加正则化项 以及callbacks
之前没找到 特此记录一下啊model=keras.models.Sequential([keras.layers.Flatten(input_shape=[28, 28]), keras.layers.Dense(300, activation='relu',kernel_regularizer=keras.regularizers.l2(1),bias_regularizer=keras.regularizers.l1(1)), .原创 2020-12-13 15:08:10 · 245 阅读 · 0 评论 -
tensorflow2 使用批归一化与dropout以及画图
import matplotlib.pyplot as pltimport numpy as npimport pandas as pdimport tensorflow as tfimport tensorflow.keras as kerasfrom sklearn.preprocessing import StandardScalerfashsion_minist = tf.keras.datasets.fashion_mnist(x_train_all, y_train_all),原创 2020-12-13 14:54:18 · 736 阅读 · 7 评论 -
tensorflow2-tensorboard使用
讲的很多,想展示什么指标都可以,只需要在metrics 处加上即可。model.compile(optimizer='adam', loss="sparse_categorical_crossentropy", metrics=["accuracy",keras.metrics.categorical_accuracy])https://www.cnblogs.com/chenhuabin/p/11736088.html...原创 2020-12-13 14:47:04 · 144 阅读 · 1 评论 -
tensorflow2 结合sklearn做网格搜索
from tensorflow import kerasimport numpy as np# load datasetfrom sklearn.datasets import fetch_california_housinghousing = fetch_california_housing()from sklearn.model_selection import train_test_splitx_train_all, x_test, y_train_all, y_test = trai原创 2020-08-26 15:32:35 · 869 阅读 · 0 评论 -
tensorflow2 函数api多输入多输出
from tensorflow import keras# load datasetfrom sklearn.datasets import fetch_california_housinghousing = fetch_california_housing()from sklearn.model_selection import train_test_splitx_train_all, x_test, y_train_all, y_test = train_test_split(housin原创 2020-08-26 15:16:39 · 589 阅读 · 0 评论 -
无监督学习之AutoEncoder
import osimport tensorflow as tfimport numpy as npfrom tensorflow import kerasfrom tensorflow.keras import Sequential, layersfrom PIL import Imagefrom matplotlib import pyplot as plttf.random.set_seed(22)np.random.seed(22)os.envi原创 2020-07-23 23:35:48 · 102 阅读 · 0 评论 -
tensorflow2.0 LSTM 情感分析
import osimport tensorflow as tfimport numpy as npfrom tensorflow import kerasfrom tensorflow.keras import layers, losses, optimizers, Sequentialtf.random.set_seed(22)np.random.seed(22)os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'assert tf.__原创 2020-07-23 15:03:01 · 370 阅读 · 2 评论 -
基于tensorflow2.0的RNN 情感分析
import osimport tensorflow as tfimport numpy as npfrom tensorflow import kerasfrom tensorflow.keras import layerstf.random.set_seed(22)np.random.seed(22)os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'assert tf.__version__.startswith('2.')batc原创 2020-07-23 09:18:45 · 409 阅读 · 0 评论 -
tensorfow2.0 实现 Resnet
import tensorflow as tffrom tensorflow import kerasfrom tensorflow.keras import layers, Sequentialclass BasicBlock(layers.Layer): def __init__(self, filter_num, stride=1): super(BasicBlock, self).__init__() self.conv1 = l原创 2020-07-22 00:05:30 · 109 阅读 · 0 评论 -
tensorflow2.0实现cnn的图像识别
import tensorflow as tffrom tensorflow.keras import layers, optimizers, datasets, Sequentialimport osos.environ['TF_CPP_MIN_LOG_LEVEL']='2'tf.random.set_seed(2345)conv_layers = [ # 5 units of conv + max pooling # unit 1 layers.Conv2D(64原创 2020-07-20 23:42:23 · 1788 阅读 · 0 评论 -
Tensorflow-keras自定义网络
import tensorflow as tffrom tensorflow.keras import datasets, layers, optimizers, Sequential, metricsfrom tensorflow import kerasimport osos.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'def preprocess(x, y): # [0~255] => [-1~1] x = 2 * tf.原创 2020-07-20 09:14:54 · 121 阅读 · 0 评论 -
tensorflow2.0之手写数字识别
不得不说,tensorflow2.0比1.x好用import tensorflow as tffrom tensorflow import kerasfrom tensorflow.keras import datasetsimport osos.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'# x: [60k, 28, 28],# y: [60k](x, y), _ = datasets.mnist.load_data()# x: [0~2原创 2020-07-18 00:38:04 · 616 阅读 · 0 评论