Error generate() - During handling of the above exception, another exception occurred

[skorotkiewicz]

Tensorflow: 1.15.2
Python: 3.6.10
CUDA: 10.0.130_411.31_win10
cudNN: 10.0-windows10-x64-v7.6.5.32

generate.py

import gpt_2_simple as gpt2
import os
import requests

model_name = "124M"
file_name = "tekst.txt"
sess = gpt2.start_tf_sess()
gpt2.generate(sess)

Finetuning by the CPU is done without any problems, but I can't generate any text because the following errors are coming out:

(tf-cpu) C:\Users\sebas\Desktop\GPT-2\gpt-2-simple>python generate.py
WARNING:tensorflow:
The TensorFlow contrib module will not be included in TensorFlow 2.0.
For more information, please see:
  * https://github.com/tensorflow/community/blob/master/rfcs/20180907-contrib-sunset.md
  * https://github.com/tensorflow/addons
  * https://github.com/tensorflow/io (for I/O related ops)
If you depend on functionality not listed there, please file an issue.

2020-07-23 11:54:58.675088: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
Traceback (most recent call last):
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\client\session.py", line 1365, in _do_call
    return fn(*args)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\client\session.py", line 1350, in _run_fn         target_list, run_metadata)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\client\session.py", line 1443, in _call_tf_sessionrun
    run_metadata)
tensorflow.python.framework.errors_impl.FailedPreconditionError: Attempting to use uninitialized value model/h8/attn/c_attn/b
         [[{{node model/h8/attn/c_attn/b/read}}]]

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "generate.py", line 10, in <module>
    gpt2.generate(sess)
  File "C:\Users\sebas\Desktop\GPT-2\gpt-2-simple\gpt_2_simple\gpt_2.py", line 475, in generate
    out = sess.run(output)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\client\session.py", line 956, in run
    run_metadata_ptr)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\client\session.py", line 1180, in _run            feed_dict_tensor, options, run_metadata)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\client\session.py", line 1359, in _do_run         run_metadata)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\client\session.py", line 1384, in _do_call        raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.FailedPreconditionError: Attempting to use uninitialized value model/h8/attn/c_attn/b
         [[node model/h8/attn/c_attn/b/read (defined at C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\framework\ops.py:1748) ]]

Original stack trace for 'model/h8/attn/c_attn/b/read':
  File "generate.py", line 10, in <module>
    gpt2.generate(sess)
  File "C:\Users\sebas\Desktop\GPT-2\gpt-2-simple\gpt_2_simple\gpt_2.py", line 466, in generate
    temperature=temperature, top_k=top_k, top_p=top_p
  File "C:\Users\sebas\Desktop\GPT-2\gpt-2-simple\gpt_2_simple\src\sample.py", line 67, in sample_sequence
    context_output = step(hparams, context[:, :-1])
  File "C:\Users\sebas\Desktop\GPT-2\gpt-2-simple\gpt_2_simple\src\sample.py", line 52, in step
    past=past, reuse=tf.compat.v1.AUTO_REUSE)
  File "C:\Users\sebas\Desktop\GPT-2\gpt-2-simple\gpt_2_simple\src\model.py", line 197, in model
    h, present = block(h, 'h%d' % layer, past=past, hparams=hparams)
  File "C:\Users\sebas\Desktop\GPT-2\gpt-2-simple\gpt_2_simple\src\model.py", line 156, in block
    a, present = attn(norm(x, 'ln_1'), 'attn', nx, past=past, hparams=hparams)
  File "C:\Users\sebas\Desktop\GPT-2\gpt-2-simple\gpt_2_simple\src\model.py", line 132, in attn
    c = conv1d(x, 'c_attn', n_state*3)
  File "C:\Users\sebas\Desktop\GPT-2\gpt-2-simple\gpt_2_simple\src\model.py", line 84, in conv1d
    b = tf.compat.v1.get_variable('b', [nf], initializer=tf.compat.v1.constant_initializer(0))
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\ops\variable_scope.py", line 1500, in get_variable
    aggregation=aggregation)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\ops\variable_scope.py", line 1243, in get_variable
    aggregation=aggregation)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\ops\variable_scope.py", line 567, in get_variable
    aggregation=aggregation)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\ops\variable_scope.py", line 519, in _true_getter
    aggregation=aggregation)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\ops\variable_scope.py", line 933, in _get_single_variable
    aggregation=aggregation)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\ops\variables.py", line 258, in __call__          return cls._variable_v1_call(*args, **kwargs)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\ops\variables.py", line 219, in _variable_v1_call
    shape=shape)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\ops\variables.py", line 197, in <lambda>          previous_getter = lambda **kwargs: default_variable_creator(None, **kwargs)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\ops\variable_scope.py", line 2519, in default_variable_creator
    shape=shape)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\ops\variables.py", line 262, in __call__          return super(VariableMetaclass, cls).__call__(*args, **kwargs)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\ops\variables.py", line 1688, in __init__         shape=shape)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\ops\variables.py", line 1872, in _init_from_args
    self._snapshot = array_ops.identity(self._variable, name="read")
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\util\dispatch.py", line 180, in wrapper           return target(*args, **kwargs)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\ops\array_ops.py", line 203, in identity          ret = gen_array_ops.identity(input, name=name)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\ops\gen_array_ops.py", line 4238, in identity     "Identity", input=input, name=name)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\framework\op_def_library.py", line 794, in _apply_op_helper
    op_def=op_def)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\util\deprecation.py", line 507, in new_func       return func(*args, **kwargs)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\framework\ops.py", line 3357, in create_op        attrs, op_def, compute_device)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\framework\ops.py", line 3426, in _create_op_internal
    op_def=op_def)
  File "C:\Miniconda\envs\tf-cpu\lib\site-packages\tensorflow_core\python\framework\ops.py", line 1748, in __init__         self._traceback = tf_stack.extract_stack()


(tf-cpu) C:\Users\sebas\Desktop\GPT-2\gpt-2-simple>

[briansemrau]

This is solved by loading the model before generating text.

sess = gpt2.start_tf_sess()
gpt2.load_gpt2(sess, model_name=model_name)  # <- this line is necessary
gpt2.generate(sess, model_name=model_name)