我刚刚安装了启用Tensorflow-GPU的软件包(tensorflow-gpu
),因为我具有CUDA支持的GPU和Keras包装器(软件包Keras
),并试图测试一切是否正常。当我尝试运行示例脚本时收到错误Failed to load the native TensorFlow runtime
,如下所示:
from keras.models import Sequential
from keras.layers import Dense
import tensorflow as tf
def main():
print(tf.VERSION)
print(tf.keras.__version__)
if __name__ == '__main__':
main()
Keras后端配置已设置为Tensorflow (因为它是默认值):
keras.json
{
"floatx": "float32",
"epsilon": 1e-07,
"backend": "tensorflow",
"image_data_format": "channels_last"
}
软件堆栈:
操作系统:Windows 10 (64 bit)
tensorflow-gpu
:1.13.1
Keras
:2.2.4