level 1
sctracy03
楼主
服务器环境：tensorflow1.2/cuda8/cudnn5.0
我好不容易设置好环境，想跑一个例子试试，用的是官网的cnn_mnist.py，一行代码没动过，结果报错：
RuntimeError: Init operations did not make model ready. Init op: group_deps, init fn: None, local_init_op: name: "group_deps_1"
op: "NoOp"
input: "^init_2"
input: "^init_all_tables"
, error: Variables not initialized: global_step, conv2d/kernel, conv2d/bias, conv2d_1/kernel, conv2d_1/bias, dense/kernel, dense/bias, dense_1/kernel, dense_1/bias
这些variable怎么初始化啊？![[狂汗]](/static/emoticons/u72c2u6c57.png)
2018年01月30日 00点01分
1
我好不容易设置好环境，想跑一个例子试试，用的是官网的cnn_mnist.py，一行代码没动过，结果报错：
RuntimeError: Init operations did not make model ready. Init op: group_deps, init fn: None, local_init_op: name: "group_deps_1"
op: "NoOp"
input: "^init_2"
input: "^init_all_tables"
, error: Variables not initialized: global_step, conv2d/kernel, conv2d/bias, conv2d_1/kernel, conv2d_1/bias, dense/kernel, dense/bias, dense_1/kernel, dense_1/bias
这些variable怎么初始化啊？