cyclegan-keras、tensorflow

前提・実現したいこと

現在このqiitaのサイトで扱っているこのgithubを試しています。
images/
|→forest_mod
|→mtg_land_mod
フォルダを作成し、それぞれシマウマの画像5枚、普通の馬画像を5枚入れて
挙動を確認しています。

発生している問題・エラーメッセージ

(cycleganio) C:\Users\imcla\Desktop\500io\cycleGAN_keras>python train.py
Using TensorFlow backend.
2019-10-24 17:26:08.307816: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
Compile Discriminators
Compile Generatorその1
./images\forest_mod*.jpg
1
./images\mtg_land_mod*.jpg
1
C:\Users\imcla.conda\envs\cycleganio\lib\site-packages\keras\engine\training.py:297: UserWarning: Discrepancy between trainable weights and collected trainable weights, did you set `model.trainable` without calling `model.compile` after ?
  'Discrepancy between trainable weights and collected trainable'
C:\Users\imcla.conda\envs\cycleganio\lib\site-packages\keras\engine\training.py:297: UserWarning: Discrepancy between trainable weights and collected trainable weights, did you set `model.trainable` without calling `model.compile` after ?
  'Discrepancy between trainable weights and collected trainable'
epoch 0 0/1000 loss_d_x:0.8669 loss_d_y:0.6979 loss_g:19.5365 74.02秒
2
2
2019-10-24 17:27:28.245006: W tensorflow/core/common_runtime/base_collective_executor.cc:216] BaseCollectiveExecutor::StartAbort Failed precondition: Error while reading resource variable _AnonymousVar393 from Container: localhost. This could mean that the variable was uninitialized. Not found: Resource localhost/_AnonymousVar393/class tensorflow::Var does not exist.
         [[{{node mul_243/ReadVariableOp}}]]
Traceback (most recent call last):
  File "train.py", line 12, in <module>
    model.train()
  File "C:\Users\imcla\Desktop\500io\cycleGAN_keras\cycleGAN.py", line 88, in train
    G=inference_model_G, F=inference_model_F)
  File "C:\Users\imcla\Desktop\500io\cycleGAN_keras\cycleGAN.py", line 234, in training_iteration
    "G": batch_B, "F": batch_A})
  File "C:\Users\imcla.conda\envs\cycleganio\lib\site-packages\keras\engine\training.py", line 1514, in train_on_batch
    outputs = self.train_function(ins)
  File "C:\Users\imcla.conda\envs\cycleganio\lib\site-packages\tensorflow_core\python\keras\backend.py", line 3740, in __call__
    outputs = self._graph_fn(*converted_inputs)
  File "C:\Users\imcla.conda\envs\cycleganio\lib\site-packages\tensorflow_core\python\eager\function.py", line 1081, in __call__
    return self._call_impl(args, kwargs)
  File "C:\Users\imcla.conda\envs\cycleganio\lib\site-packages\tensorflow_core\python\eager\function.py", line 1121, in _call_impl
    return self._call_flat(args, self.captured_inputs, cancellation_manager)
  File "C:\Users\imcla.conda\envs\cycleganio\lib\site-packages\tensorflow_core\python\eager\function.py", line 1224, in _call_flat
    ctx, args, cancellation_manager=cancellation_manager)
  File "C:\Users\imcla.conda\envs\cycleganio\lib\site-packages\tensorflow_core\python\eager\function.py", line 511, in call
    ctx=ctx)
  File "C:\Users\imcla.conda\envs\cycleganio\lib\site-packages\tensorflow_core\python\eager\execute.py", line 67, in quick_execute
    six.raise_from(core._status_to_exception(e.code, message), None)
  File "<string>", line 3, in raise_from
tensorflow.python.framework.errors_impl.FailedPreconditionError:  Error while reading resource variable _AnonymousVar393 from Container: localhost. This could mean that the variable was uninitialized. Not found: Resource localhost/_AnonymousVar393/class tensorflow::Var does not exist.
         [[node mul_243/ReadVariableOp (defined at C:\Users\imcla.conda\envs\cycleganio\lib\site-packages\tensorflow_core\python\framework\ops.py:1751) ]] [Op:__inference_keras_scratch_graph_61310]

Function call stack:
keras_scratch_graph

なるエラーメッセージが出力されます。
File "C:\Users\imcla.conda\envs\cycleganio\lib\site-packages\tensorflow_core\python\eager\execute.py"
が何かしら原因がある？と思います。以下がそのコードです。
字数的に端折っています。

execute.py(一部抜粋)
1  except core._NotOkStatusException as e:
2    if name is not None:
3      message = e.message + " name: " + name
4    else:
5      message = e.message
6
7    six.raise_from(core._status_to_exception(e.code, message), None)#ここでエラー
8  except TypeError as e:
9    keras_symbolic_tensors = [
10        x for x in inputs if ops._is_keras_symbolic_tensor(x)
11    ]
12    if keras_symbolic_tensors:
13      raise core._SymbolicException(
14          "Inputs to eager execution function cannot be Keras symbolic "
15          "tensors, but found {}".format(keras_symbolic_tensors))
16    raise e
17  # pylint: enable=protected-access
18  return tensors
19

試したこと

エラー文で検索したところこちらのページがありましたが、変数初期値がまずいとのことですが、私は全くいじっていません。
tensorflowのバージョン等も原因があるのでしょうか？

詳しい方、ご教授お願い致します。

実行環境

windows10
Anaconda python3.6

conda
1# packages in environment at C:\Users\imcla.conda\envs\cycleganio:
2#
3# Name                    Version                   Build  Channel
4absl-py                   0.8.1                    pypi_0    pypi
5astor                     0.8.0                    pypi_0    pypi
6certifi                   2019.9.11                py36_0
7cycler                    0.10.0                   pypi_0    pypi
8decorator                 4.4.0                    pypi_0    pypi
9gast                      0.2.2                    pypi_0    pypi
10google-pasta              0.1.7                    pypi_0    pypi
11grpcio                    1.24.3                   pypi_0    pypi
12h5py                      2.10.0                   pypi_0    pypi
13imageio                   2.6.1                    pypi_0    pypi
14keras                     2.3.1                    pypi_0    pypi
15keras-applications        1.0.8                    pypi_0    pypi
16keras-preprocessing       1.1.0                    pypi_0    pypi
17kiwisolver                1.1.0                    pypi_0    pypi
18markdown                  3.1.1                    pypi_0    pypi
19matplotlib                3.1.1                    pypi_0    pypi
20networkx                  2.4                      pypi_0    pypi
21numpy                     1.17.3                   pypi_0    pypi
22opt-einsum                3.1.0                    pypi_0    pypi
23pandas                    0.25.2                   pypi_0    pypi
24pillow                    6.2.1                    pypi_0    pypi
25pip                       19.3.1                   py36_0
26protobuf                  3.10.0                   pypi_0    pypi
27pyparsing                 2.4.2                    pypi_0    pypi
28python                    3.6.9                h5500b2f_0
29python-dateutil           2.8.0                    pypi_0    pypi
30pytz                      2019.3                   pypi_0    pypi
31pywavelets                1.1.1                    pypi_0    pypi
32pyyaml                    5.1.2                    pypi_0    pypi
33scikit-image              0.16.2                   pypi_0    pypi
34scipy                     1.3.1                    pypi_0    pypi
35setuptools                41.4.0                   py36_0
36six                       1.12.0                   pypi_0    pypi
37sqlite                    3.30.0               he774522_0
38tensorboard               2.0.0                    pypi_0    pypi
39tensorflow                2.0.0                    pypi_0    pypi
40tensorflow-estimator      2.0.1                    pypi_0    pypi
41termcolor                 1.1.0                    pypi_0    pypi
42vc                        14.1                 h0510ff6_4
43vs2015_runtime            14.16.27012          hf0eaf9b_0
44werkzeug                  0.16.0                   pypi_0    pypi
45wheel                     0.33.6                   py36_0
46wincertstore              0.2              py36h7fe50ca_0
47wrapt                     1.11.2                   pypi_0    pypi