Win10操作系统，conda 4.7.12，python3.7.5

1.使用管理员权限，打开cmd，进入Anaconda路径

D:\Anaconda3\

2. Create a virtual environment

$ conda create -n venv pip python=3.7 #select your python wersion

$ conda activate venv #activate the virtual enveronment

(venv)$ pip install --ignore-installed --upgrade packageURL #install the TensorFlow pip package using its complete URL

(venv)$ conda deactivate #exit virtualenv

completeURL: https://www.tensorflow.org/install/pip?lang=python3#package-location

记住选择‘CPU-only’

如果是python版本3.7，可以使用packageURL：https://storage.googleapis.com/tensorflow/windows/cpu/tensorflow-2.0.0-cp37-cp37m-win_amd64.whl

3.Install the TensorFlow pip package

(venv)$ pip install --upgrade tensorflow

(venv)$ python -c "import tensorflow as tf;print(tf.reduce_sum(tf.random.normal([1000, 1000])))" #verity the install

安装的是CPU版本，在验证代码中加入以下代码，忽略警告：

import os

os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'

4. 配置Anaconda，使用Jupyter

打开Anaconda，修改environment至venv：

重新install Jupyter并打开就可以使用了：

5.测试

TensorFlow官方代码，可以作为测试：

import tensorflow as tf

mnist = tf.keras.datasets.mnist

(x_train, y_train),(x_test, y_test) = mnist.load_data()

x_train, x_test = x_train / 255.0, x_test / 255.0

model = tf.keras.models.Sequential([

tf.keras.layers.Flatten(input_shape=(28, 28)),

tf.keras.layers.Dense(128, activation='relu'),

tf.keras.layers.Dropout(0.2),

tf.keras.layers.Dense(10, activation='softmax')

])

model.compile(optimizer='adam',

loss='sparse_categorical_crossentropy',

metrics=['accuracy'])

model.fit(x_train, y_train, epochs=5)

model.evaluate(x_test, y_test)