tensorflow的一些版本问题

tensorflow 1.5 keras 2.1.4， ,1.4和2.1.3搭配,1.2和2.11搭配

安装固定版本的tensorflow
pip install tensorflow-gpu==1.1.0

tensorflow与cudnn和cuda版本对应

Version: CPU/GPU: Python Version: Compiler: Build Tools: cuDNN: CUDA:
tensorflow-1.8.0 CPU 2.7, 3.3-3.6 GCC 4.8 Bazel 0.10.0 N/A N/A
tensorflow_gpu-1.8.0 GPU 2.7, 3.3-3.6 GCC 4.8 Bazel 0.9.0 7 9
tensorflow-1.7.0 CPU 2.7, 3.3-3.6 GCC 4.8 Bazel 0.10.0 N/A N/A
tensorflow_gpu-1.7.0 GPU 2.7, 3.3-3.6 GCC 4.8 Bazel 0.9.0 7 9
tensorflow-1.6.0 CPU 2.7, 3.3-3.6 GCC 4.8 Bazel 0.9.0 N/A N/A
tensorflow_gpu-1.6.0 GPU 2.7, 3.3-3.6 GCC 4.8 Bazel 0.9.0 7 9
tensorflow-1.5.0 CPU 2.7, 3.3-3.6 GCC 4.8 Bazel 0.8.0 N/A N/A
tensorflow_gpu-1.5.0 GPU 2.7, 3.3-3.6 GCC 4.8 Bazel 0.8.0 7 9
tensorflow-1.4.0 CPU 2.7, 3.3-3.6 GCC 4.8 Bazel 0.5.4 N/A N/A
tensorflow_gpu-1.4.0 GPU 2.7, 3.3-3.6 GCC 4.8 Bazel 0.5.4 6 8
tensorflow-1.3.0 CPU 2.7, 3.3-3.6 GCC 4.8 Bazel 0.4.5 N/A N/A
tensorflow_gpu-1.3.0 GPU 2.7, 3.3-3.6 GCC 4.8 Bazel 0.4.5 6 8
tensorflow-1.2.0 CPU 2.7, 3.3-3.6 GCC 4.8 Bazel 0.4.5 N/A N/A
tensorflow_gpu-1.2.0 GPU 2.7, 3.3-3.6 GCC 4.8 Bazel 0.4.5 5.1 8
tensorflow-1.1.0 CPU 2.7, 3.3-3.6 GCC 4.8 Bazel 0.4.2 N/A N/A
tensorflow_gpu-1.1.0 GPU 2.7, 3.3-3.6 GCC 4.8 Bazel 0.4.2 5.1 8
tensorflow-1.0.0 CPU 2.7, 3.3-3.6 GCC 4.8 Bazel 0.4.2 N/A N/A
tensorflow_gpu-1.0.0 GPU 2.7, 3.3-3.6 GCC 4.8 Bazel 0.4.2 5.1 8