cc:184 Key generator/g_d1_bn/beta not found in checkpoint

[Henry658]

when I run pcgreat/zi2zi
do the instruction
have the error
I am not sure that i am get the train model in my Issues problem #59
I suppose that I am success get the model in the correct model_dir
but have the error info that
cc:184 Key generator/g_d1_bn/beta not found in checkpoint
How should I fixed this problem ?

the following text is my input instruction and the terminal log

Infer

PYTHONPATH=. python infer.py --model_dir=experiments/checkpoint/experiment_0
--batch_size=32
--source_obj=experiments/data/val.obj
--embedding_ids=0
--save_dir=save_dir/

(py36) sjshyu@sjshyu-titan:~/Desktop/360315$ sh test.sh
2019-03-29 14:10:03.478788: I tensorflow/core/platform/cpu_feature_guard.cc:140] Your CPU supports instructions that this TensorFlow binary was not compiled to use: SSE4.1 SSE4.2 AVX AVX2 FMA
2019-03-29 14:10:03.660390: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1356] Found device 0 with properties:
name: GeForce GTX TITAN X major: 5 minor: 2 memoryClockRate(GHz): 1.076
pciBusID: 0000:02:00.0
totalMemory: 11.92GiB freeMemory: 11.17GiB
2019-03-29 14:10:03.792179: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1356] Found device 1 with properties:
name: GeForce GTX TITAN X major: 5 minor: 2 memoryClockRate(GHz): 1.076
pciBusID: 0000:81:00.0
totalMemory: 11.92GiB freeMemory: 11.80GiB
2019-03-29 14:10:03.792283: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1435] Adding visible gpu devices: 0, 1
2019-03-29 14:10:04.332448: I tensorflow/core/common_runtime/gpu/gpu_device.cc:923] Device interconnect StreamExecutor with strength 1 edge matrix:
2019-03-29 14:10:04.332508: I tensorflow/core/common_runtime/gpu/gpu_device.cc:929] 0 1
2019-03-29 14:10:04.332519: I tensorflow/core/common_runtime/gpu/gpu_device.cc:942] 0: N N
2019-03-29 14:10:04.332524: I tensorflow/core/common_runtime/gpu/gpu_device.cc:942] 1: N N
2019-03-29 14:10:04.333112: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1053] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 10816 MB memory) -> physical GPU (device: 0, name: GeForce GTX TITAN X, pci bus id: 0000:02:00.0, compute capability: 5.2)
2019-03-29 14:10:04.333703: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1053] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:1 with 11430 MB memory) -> physical GPU (device: 1, name: GeForce GTX TITAN X, pci bus id: 0000:81:00.0, compute capability: 5.2)
unpickled total 946 examples
examples -> 946
2019-03-29 14:10:07.096577: W tensorflow/core/framework/op_kernel.cc:1318] OP_REQUIRES failed at save_restore_v2_ops.cc:184 : Not found: Key generator/g_d1_bn/beta not found in checkpoint
Traceback (most recent call last):
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1322, in _do_call
return fn(*args)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1307, in _run_fn
options, feed_dict, fetch_list, target_list, run_metadata)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1409, in _call_tf_sessionrun
run_metadata)
tensorflow.python.framework.errors_impl.NotFoundError: Key generator/g_d1_bn/beta not found in checkpoint
[[Node: save/RestoreV2 = RestoreV2[dtypes=[DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, ..., DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT], _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_save/Const_0_0, save/RestoreV2/tensor_names, save/RestoreV2/shape_and_slices)]]
[[Node: save/RestoreV2/_149 = _Recvclient_terminated=false, recv_device="/job:localhost/replica:0/task:0/device:GPU:0", send_device="/job:localhost/replica:0/task:0/device:CPU:0", send_device_incarnation=1, tensor_name="edge_154_save/RestoreV2", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:GPU:0"]]

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "infer.py", line 67, in
tf.app.run()
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/platform/app.py", line 126, in run
_sys.exit(main(argv))
File "infer.py", line 47, in main
save_dir=args.save_dir)
File "/home/sjshyu/Desktop/360315/model/unet.py", line 446, in infer
self.load_model(model_dir)
File "/home/sjshyu/Desktop/360315/model/unet.py", line 469, in load_model
self.restore_model(saver, model_dir)
File "/home/sjshyu/Desktop/360315/model/unet.py", line 357, in restore_model
saver.restore(self.sess, ckpt.model_checkpoint_path)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/training/saver.py", line 1802, in restore
{self.saver_def.filename_tensor_name: save_path})
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 900, in run
run_metadata_ptr)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1135, in _run
feed_dict_tensor, options, run_metadata)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1316, in _do_run
run_metadata)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1335, in _do_call
raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.NotFoundError: Key generator/g_d1_bn/beta not found in checkpoint
[[Node: save/RestoreV2 = RestoreV2[dtypes=[DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, ..., DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT], _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_save/Const_0_0, save/RestoreV2/tensor_names, save/RestoreV2/shape_and_slices)]]
[[Node: save/RestoreV2/_149 = _Recvclient_terminated=false, recv_device="/job:localhost/replica:0/task:0/device:GPU:0", send_device="/job:localhost/replica:0/task:0/device:CPU:0", send_device_incarnation=1, tensor_name="edge_154_save/RestoreV2", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:GPU:0"]]

Caused by op 'save/RestoreV2', defined at:
File "infer.py", line 67, in
tf.app.run()
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/platform/app.py", line 126, in run
_sys.exit(main(argv))
File "infer.py", line 47, in main
save_dir=args.save_dir)
File "/home/sjshyu/Desktop/360315/model/unet.py", line 446, in infer
self.load_model(model_dir)
File "/home/sjshyu/Desktop/360315/model/unet.py", line 468, in load_model
saver = tf.train.Saver(var_list=self.retrieve_generator_vars())
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/training/saver.py", line 1338, in init
self.build()
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/training/saver.py", line 1347, in build
self._build(self._filename, build_save=True, build_restore=True)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/training/saver.py", line 1384, in _build
build_save=build_save, build_restore=build_restore)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/training/saver.py", line 835, in _build_internal
restore_sequentially, reshape)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/training/saver.py", line 472, in _AddRestoreOps
restore_sequentially)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/training/saver.py", line 886, in bulk_restore
return io_ops.restore_v2(filename_tensor, names, slices, dtypes)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/ops/gen_io_ops.py", line 1463, in restore_v2
shape_and_slices=shape_and_slices, dtypes=dtypes, name=name)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py", line 787, in _apply_op_helper
op_def=op_def)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 3392, in create_op
op_def=op_def)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 1718, in init
self._traceback = self._graph._extract_stack() # pylint: disable=protected-access

NotFoundError (see above for traceback): Key generator/g_d1_bn/beta not found in checkpoint
[[Node: save/RestoreV2 = RestoreV2[dtypes=[DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, ..., DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT], _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_save/Const_0_0, save/RestoreV2/tensor_names, save/RestoreV2/shape_and_slices)]]
[[Node: save/RestoreV2/_149 = _Recvclient_terminated=false, recv_device="/job:localhost/replica:0/task:0/device:GPU:0", send_device="/job:localhost/replica:0/task:0/device:CPU:0", send_device_incarnation=1, tensor_name="edge_154_save/RestoreV2", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:GPU:0"]]

This is my contact information.
mail: henry19980520@gmail.com
I look forward to discussing with you.

[Henry658]

Hey guy I have some update for this issues
I used the tensorflow.python.tools inspect_chpoint to print the tensor in the checkpoint file
as following link

tensor.txt

Really have't the key g_d1_bn/beta
in fact its only esist the g_d3~d_d8
but why the infer.py or unet reload the checkpoint file still need this key
or have some error in my training or save model that leads to I without the key g_d1_bn

the same as past I and still try to solve this problem
This is my contact information.
mail: henry19980520@gmail.com
I look forward to discussing with you.

[Henry658]

when I run pcgreat/zi2zi
do the instruction
have the error
I am not sure that i am get the train model in my Issues problem #59
I suppose that I am success get the model in the correct model_dir
but have the error info that
cc:184 Key generator/g_d1_bn/beta not found in checkpoint
How should I fixed this problem ?

the following text is my input instruction and the terminal log

Infer

PYTHONPATH=. python infer.py --model_dir=experiments/checkpoint/experiment_0
--batch_size=32
--source_obj=experiments/data/val.obj
--embedding_ids=0
--save_dir=save_dir/

(py36) sjshyu@sjshyu-titan:~/Desktop/360315$ sh test.sh
2019-03-29 14:10:03.478788: I tensorflow/core/platform/cpu_feature_guard.cc:140] Your CPU supports instructions that this TensorFlow binary was not compiled to use: SSE4.1 SSE4.2 AVX AVX2 FMA
2019-03-29 14:10:03.660390: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1356] Found device 0 with properties:
name: GeForce GTX TITAN X major: 5 minor: 2 memoryClockRate(GHz): 1.076
pciBusID: 0000:02:00.0
totalMemory: 11.92GiB freeMemory: 11.17GiB
2019-03-29 14:10:03.792179: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1356] Found device 1 with properties:
name: GeForce GTX TITAN X major: 5 minor: 2 memoryClockRate(GHz): 1.076
pciBusID: 0000:81:00.0
totalMemory: 11.92GiB freeMemory: 11.80GiB
2019-03-29 14:10:03.792283: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1435] Adding visible gpu devices: 0, 1
2019-03-29 14:10:04.332448: I tensorflow/core/common_runtime/gpu/gpu_device.cc:923] Device interconnect StreamExecutor with strength 1 edge matrix:
2019-03-29 14:10:04.332508: I tensorflow/core/common_runtime/gpu/gpu_device.cc:929] 0 1
2019-03-29 14:10:04.332519: I tensorflow/core/common_runtime/gpu/gpu_device.cc:942] 0: N N
2019-03-29 14:10:04.332524: I tensorflow/core/common_runtime/gpu/gpu_device.cc:942] 1: N N
2019-03-29 14:10:04.333112: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1053] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 10816 MB memory) -> physical GPU (device: 0, name: GeForce GTX TITAN X, pci bus id: 0000:02:00.0, compute capability: 5.2)
2019-03-29 14:10:04.333703: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1053] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:1 with 11430 MB memory) -> physical GPU (device: 1, name: GeForce GTX TITAN X, pci bus id: 0000:81:00.0, compute capability: 5.2)
unpickled total 946 examples
examples -> 946
2019-03-29 14:10:07.096577: W tensorflow/core/framework/op_kernel.cc:1318] OP_REQUIRES failed at save_restore_v2_ops.cc:184 : Not found: Key generator/g_d1_bn/beta not found in checkpoint
Traceback (most recent call last):
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1322, in _do_call
return fn(*args)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1307, in _run_fn
options, feed_dict, fetch_list, target_list, run_metadata)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1409, in _call_tf_sessionrun
run_metadata)
tensorflow.python.framework.errors_impl.NotFoundError: Key generator/g_d1_bn/beta not found in checkpoint
[[Node: save/RestoreV2 = RestoreV2[dtypes=[DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, ..., DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT], _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_save/Const_0_0, save/RestoreV2/tensor_names, save/RestoreV2/shape_and_slices)]]
[[Node: save/RestoreV2/_149 = _Recvclient_terminated=false, recv_device="/job:localhost/replica:0/task:0/device:GPU:0", send_device="/job:localhost/replica:0/task:0/device:CPU:0", send_device_incarnation=1, tensor_name="edge_154_save/RestoreV2", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:GPU:0"]]

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "infer.py", line 67, in
tf.app.run()
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/platform/app.py", line 126, in run
_sys.exit(main(argv))
File "infer.py", line 47, in main
save_dir=args.save_dir)
File "/home/sjshyu/Desktop/360315/model/unet.py", line 446, in infer
self.load_model(model_dir)
File "/home/sjshyu/Desktop/360315/model/unet.py", line 469, in load_model
self.restore_model(saver, model_dir)
File "/home/sjshyu/Desktop/360315/model/unet.py", line 357, in restore_model
saver.restore(self.sess, ckpt.model_checkpoint_path)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/training/saver.py", line 1802, in restore
{self.saver_def.filename_tensor_name: save_path})
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 900, in run
run_metadata_ptr)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1135, in _run
feed_dict_tensor, options, run_metadata)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1316, in _do_run
run_metadata)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1335, in _do_call
raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.NotFoundError: Key generator/g_d1_bn/beta not found in checkpoint
[[Node: save/RestoreV2 = RestoreV2[dtypes=[DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, ..., DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT], _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_save/Const_0_0, save/RestoreV2/tensor_names, save/RestoreV2/shape_and_slices)]]
[[Node: save/RestoreV2/_149 = _Recvclient_terminated=false, recv_device="/job:localhost/replica:0/task:0/device:GPU:0", send_device="/job:localhost/replica:0/task:0/device:CPU:0", send_device_incarnation=1, tensor_name="edge_154_save/RestoreV2", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:GPU:0"]]

Caused by op 'save/RestoreV2', defined at:
File "infer.py", line 67, in
tf.app.run()
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/platform/app.py", line 126, in run
_sys.exit(main(argv))
File "infer.py", line 47, in main
save_dir=args.save_dir)
File "/home/sjshyu/Desktop/360315/model/unet.py", line 446, in infer
self.load_model(model_dir)
File "/home/sjshyu/Desktop/360315/model/unet.py", line 468, in load_model
saver = tf.train.Saver(var_list=self.retrieve_generator_vars())
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/training/saver.py", line 1338, in init
self.build()
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/training/saver.py", line 1347, in build
self._build(self._filename, build_save=True, build_restore=True)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/training/saver.py", line 1384, in _build
build_save=build_save, build_restore=build_restore)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/training/saver.py", line 835, in _build_internal
restore_sequentially, reshape)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/training/saver.py", line 472, in _AddRestoreOps
restore_sequentially)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/training/saver.py", line 886, in bulk_restore
return io_ops.restore_v2(filename_tensor, names, slices, dtypes)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/ops/gen_io_ops.py", line 1463, in restore_v2
shape_and_slices=shape_and_slices, dtypes=dtypes, name=name)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py", line 787, in _apply_op_helper
op_def=op_def)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 3392, in create_op
op_def=op_def)
File "/home/sjshyu/.conda/envs/py36/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 1718, in init
self._traceback = self._graph._extract_stack() # pylint: disable=protected-access

NotFoundError (see above for traceback): Key generator/g_d1_bn/beta not found in checkpoint
[[Node: save/RestoreV2 = RestoreV2[dtypes=[DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, ..., DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT], _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_save/Const_0_0, save/RestoreV2/tensor_names, save/RestoreV2/shape_and_slices)]]
[[Node: save/RestoreV2/_149 = _Recvclient_terminated=false, recv_device="/job:localhost/replica:0/task:0/device:GPU:0", send_device="/job:localhost/replica:0/task:0/device:CPU:0", send_device_incarnation=1, tensor_name="edge_154_save/RestoreV2", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:GPU:0"]]

This is my contact information.
mail: henry19980520@gmail.com
I look forward to discussing with you.

[jessiimay]

Hey guy I have some update for this issues
I used the tensorflow.python.tools inspect_chpoint to print the tensor in the checkpoint file
as following link

tensor.txt

Really have't the key g_d1_bn/beta
in fact its only esist the g_d3~d_d8
but why the infer.py or unet reload the checkpoint file still need this key
or have some error in my training or save model that leads to I without the key g_d1_bn

the same as past I and still try to solve this problem
This is my contact information.
mail: henry19980520@gmail.com
I look forward to discussing with you.

Hi, have you found any way to solve it?
I've ran into this error too, and I wonder if the "Key generator/g_d1/bn/beta" was not trained, or it was just not saved into checkpoint.

If you have any solution, please reply. I really need to get it work.
Thanks!

[Henry658]

Hi, have you found any way to solve it?
I've ran into this error too, and I wonder if the "Key generator/g_d1/bn/beta" was not trained, or it was just not saved into checkpoint.

If you have any solution, please reply. I really need to get it work.
Thanks!
This is because Unet insert with norm layer when the training but didn't insert when infer.
you can see that train.py parameter inst_norm default is 1.
but infer.py default is 0. and we should replace to 1.

maybe you will encounter others problems:

problem:
InvalidArgumentError (see above for traceback): Assign requires shapes of both tensors to match. lhs shape= [80,1,1,64] rhs shape= [185,1,1,128]

because Unet's embedding number ,size aren't match. you should ensure that the parameter of infer.py embedding_num and embedding_dim same with train.py

problem:
ValueError: Cannot feed value of shape (16, 256, 256, 2) for Tensor 'real_A_and_B_images:0', which has shape '(?, 128, 128, 2)'

because two document image_size are different should make sure that the tarin.py and infer.py are same

or you can see the original message someone answer me before

这是因为train.py和infer.py的inst_norm参数不一样导致的。

可以看到train.py的初始化参数inst_norm参数默认为1，而infer.py该参数默认为0，将其改为1即可，差别可见： model/unet.py#L114

除了这个问题以外，也许还会遇到：

InvalidArgumentError (see above for traceback): Assign requires shapes of both tensors to match. lhs shape= [80,1,1,64] rhs shape= [185,1,1,128]

因为UNet的embedding数量、大小不一样，对应train.py的embedding_num和embedding_dim参数，将其应用到infer.py就行，可见：/train.py#L24

以及这个问题：

ValueError: Cannot feed value of shape (16, 256, 256, 2) for Tensor 'real_A_and_B_images:0', which has shape '(?, 128, 128, 2)'

这是因为两个文件的image_size不一样，根据train.py的参数应用到infer.py就行了，可见：/train.py#L17