tensorflow1.0基础操作

打印所有节点：

[n.name for n in tf.get_default_graph().as_graph_def().node]

save parameter to dict：

model_save = {}
variables = tf.get_collection('variables')
for var in variables:
  name = var.name
  name = name.split(":")[0]
  layer, parameter = name.split("/")
  if layer not in model_save.keys():
   	 model_save[layer] = dict()
  model_save[layer][parameter] = var.eval()

view graph on tensorboard:

writer = tf.summary.FileWriter("logs/", sess.graph)

start tensorboard

tensorboard --logdir logs --host 0.0.0.0 --port 6006