Code Block:

Keras "Failed to call ThenRnnForward with model config"

（重磅）Internal: Failed to call ThenRnnForward with model config 问题的解决（Keras 2.4.3 和 Tensorflow2.0 系列） https://www.cnblogs.com/xiaojieshisilang/p/13636648.html

No!!!!

Just the GPU's RAM are all occupied by another process.

nvidia-smi to check.

Still the RAM are goes to one process.

Maybe there are already some GPU load balancer could help...eg multiple user use cases

Tensorflow "Could not load dynamic library libcudart.so.10.1"

https://www.cxyzjd.com/article/qq_19707521/107384416

1. Step 1

export LD_LIBRARY_PATH=/usr/local/cuda/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}

2. Step 2

cd /usr/local/cuda/lib64

sudo cp libcublas.so.10.0 libcublas.so.10
sudo cp libcufft.so.10.0 libcufft.so.10
sudo cp libcurand.so.10.0 libcurand.so.10
sudo cp libcusolver.so.10.0 libcusolver.so.10
sudo cp libcusparse.so.10.0 libcusparse.so.10

Keras "cannot import name 'get_config'"

Instead of

import keras
import keras.xxx

use

import tensorflow.keras
import tensorflow.keras.xxx