没有项目描述
dffml-model-tensorflow的Python项目详细描述
tensorflow库的dffml模型
关于
由tensorflow支持的dffml模型。
演示
This demo was taken before the below steps were updated, the actual accuracy may vary as this video shows accuracy being assessed against the training data. You should try it for yourself and see!
安装
virtualenv -p python3.7 .venv. .venv/bin/activatepython3.7 -m pip install --user -U dffml[tensorflow]
用法
wget http://download.tensorflow.org/data/iris_training.csvwget http://download.tensorflow.org/data/iris_test.csvhead iris_training.csvsed -i 's/.*setosa,versicolor,virginica/SepalLength,SepalWidth,PetalLength,PetalWidth,classification/g' *.csvhead iris_training.csvdffml train \ -model dnn \ -sources csv=iris_training.csv \ -classifications 0 1 2 \ -features \ def:SepalLength:float:1 \ def:SepalWidth:float:1 \ def:PetalLength:float:1 \ def:PetalWidth:float:1 \ -num_epochs 3000 \ -steps 20000 \ -log debugdffml accuracy \ -model dnn \ -sources csv=iris_test.csv \ -classifications 0 1 2 \ -features \ def:SepalLength:float:1 \ def:SepalWidth:float:1 \ def:PetalLength:float:1 \ def:PetalWidth:float:1 \ -log criticaldffml predict all \ -model dnn \ -sources csv=iris_test.csv \ -classifications 0 1 2 \ -features \ def:SepalLength:float:1 \ def:SepalWidth:float:1 \ def:PetalLength:float:1 \ def:PetalWidth:float:1 \ -caching \ -log critical \ > results.jsonhead -n 33 results.json
许可证
dffml tensorflow模型在 MIT License。