ValueError: Variable chars/_char_embeddings already exists, disallowed. Did you mean to set reuse=True in VarScope? Originally defined at ner_model about sequence_tagging

[AutometricsInfi]

This issue is coming when we try to load second model with different word,char,tag files when first model is already loaded.

Also getting below error
File "/home/ubuntu/anaconda3/lib/python3.6/site-packages/tensorflow/python/training/saver.py", line 1562, in restore
{self.saver_def.filename_tensor_name: save_path})
File "/home/ubuntu/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 907, in run
run_metadata_ptr)
File "/home/ubuntu/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1144, in _run
feed_dict_tensor, options, run_metadata)
File "/home/ubuntu/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1372, in _do_run
options, run_metadata)
File "/home/ubuntu/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1391, in _do_call
raise type(e)(node_def, op, message)

tensorflow.python.framework.errors_impl.InvalidArgumentError: Assign requires shapes of both tensors to match. lhs shape= [1839,300] rhs shape= [1633,300]
[[Node: save/Assign_38 = Assign[T=DT_FLOAT, _class=["loc:@words/_word_embeddings"], use_locking=true, validate_shape=true, _device="/job:localhost/replica:0/task:0/cpu:0"](words/_word_embeddings, save/RestoreV2_38)]]

when trying to load multiple models. Even save_path for different models are correct and used different config files to load words,chars, tags files. But any model which is getting loaded considering words,tags,chars generated from last training session build_data.py script because of which above error is thrown from tensorflow.

Need help to understand this behavior and solution for the same.

[YogeshSakpal]

I am also facing the same issue. Could you please guide on below?

ValueError: Variable words/_word_embeddings already exists, disallowed. Did you mean to set reuse=True or reuse=tf.AUTO_REUSE in VarScope? Originally defined at:

File "C:\ProgramData\Anaconda3\lib\site-packages\tensorflow\python\framework\ops.py", line 1717, in init
self._traceback = tf_stack.extract_stack()
File "C:\ProgramData\Anaconda3\lib\site-packages\tensorflow\python\framework\ops.py", line 3155, in create_op
op_def=op_def)
File "C:\ProgramData\Anaconda3\lib\site-packages\tensorflow\python\util\deprecation.py", line 454, in new_func
return func(*args, **kwargs)

[guillaumegenthial]

If you change the size of your vocab etc. the shapes of your tensor change, and thus when you reload, you have shape mismatch!

[adamchuai]

If you change the size of your vocab etc. the shapes of your tensor change, and thus when you reload, you have shape mismatch!

Excuse me, but could you please explain more specifically about changing the size of vocab etc? And how does it cause the error? I am also facing the same problem, but I dont think I have changed anything. Thank you so much!