Python CNN学習時のエラー改善方法を教えていただきたいです。

Traceback (most recent call last):
  File "main2.py", line 198, in <module>
    'output2': Y_test2})
  File "C:\Users\k773412\AppData\Local\Continuum\anaconda3\envs\STGCN\lib\site-packages\tensorflow\python\keras\engine\training.py", line 108, in _method_wrapper
    return method(self, *args, **kwargs)
  File "C:\Users\k773412\AppData\Local\Continuum\anaconda3\envs\STGCN\lib\site-packages\tensorflow\python\keras\engine\training.py", line 1063, in fit
    steps_per_execution=self._steps_per_execution)
  File "C:\Users\k773412\AppData\Local\Continuum\anaconda3\envs\STGCN\lib\site-packages\tensorflow\python\keras\engine\data_adapter.py", line 1117, in __init__
    model=model)
  File "C:\Users\k773412\AppData\Local\Continuum\anaconda3\envs\STGCN\lib\site-packages\tensorflow\python\keras\engine\data_adapter.py", line 265, in __init__
    x, y, sample_weights = _process_tensorlike((x, y, sample_weights))
  File "C:\Users\k773412\AppData\Local\Continuum\anaconda3\envs\STGCN\lib\site-packages\tensorflow\python\keras\engine\data_adapter.py", line 1021, in _process_tensorlike
    inputs = nest.map_structure(_convert_numpy_and_scipy, inputs)
  File "C:\Users\k773412\AppData\Local\Continuum\anaconda3\envs\STGCN\lib\site-packages\tensorflow\python\util\nest.py", line 635, in map_structure
    structure[0], [func(*x) for x in entries],
  File "C:\Users\k773412\AppData\Local\Continuum\anaconda3\envs\STGCN\lib\site-packages\tensorflow\python\util\nest.py", line 635, in <listcomp>
    structure[0], [func(*x) for x in entries],
  File "C:\Users\k773412\AppData\Local\Continuum\anaconda3\envs\STGCN\lib\site-packages\tensorflow\python\keras\engine\data_adapter.py", line 1016, in _convert_numpy_and_scipy
    return ops.convert_to_tensor(x, dtype=dtype)
  File "C:\Users\k773412\AppData\Local\Continuum\anaconda3\envs\STGCN\lib\site-packages\tensorflow\python\framework\ops.py", line 1499, in convert_to_tensor
    ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
  File "C:\Users\k773412\AppData\Local\Continuum\anaconda3\envs\STGCN\lib\site-packages\tensorflow\python\framework\tensor_conversion_registry.py", line 52, in _default_conversion_function
    return constant_op.constant(value, dtype, name=name)
  File "C:\Users\k773412\AppData\Local\Continuum\anaconda3\envs\STGCN\lib\site-packages\tensorflow\python\framework\constant_op.py", line 264, in constant
    allow_broadcast=True)
  File "C:\Users\k773412\AppData\Local\Continuum\anaconda3\envs\STGCN\lib\site-packages\tensorflow\python\framework\constant_op.py", line 275, in _constant_impl
    return _constant_eager_impl(ctx, value, dtype, shape, verify_shape)
  File "C:\Users\k773412\AppData\Local\Continuum\anaconda3\envs\STGCN\lib\site-packages\tensorflow\python\framework\constant_op.py", line 300, in _constant_eager_impl
    t = convert_to_eager_tensor(value, ctx, dtype)
  File "C:\Users\k773412\AppData\Local\Continuum\anaconda3\envs\STGCN\lib\site-packages\tensorflow\python\framework\constant_op.py", line 98, in convert_to_eager_tensor
    return ops.EagerTensor(value, ctx.device_name, dtype)
ValueError: Failed to convert a NumPy array to a Tensor (Unsupported object type numpy.ndarray).

import numpy as np
import tensorflow as tf
import matplotlib.pyplot as plt
import keras
import cupy
import os
from PIL import Image
from sklearn.model_selection import train_test_split
from keras.utils import np_utils
from tensorflow.keras.models import Model, load_model
from keras.layers.core import Dense
from tensorflow.keras.layers import Dense, Dropout, Activation, Flatten, Input
from tensorflow.keras.layers import Dense, Flatten, Conv3D, MaxPooling3D, Dropout
from tensorflow.keras.callbacks import EarlyStopping, ModelCheckpoint
from tensorflow.keras.optimizers import Adam, RMSprop, SGD
#from keras.utils.np_utils import to_categorical
import copy
import glob
from tensorflow.keras.preprocessing.image import load_img,img_to_array
from tensorflow.keras import layers, models
from keras.utils import np_utils
from sklearn.metrics import mean_absolute_error

aInput = Input(shape = (160,120,120,3))

z = Conv3D(32, (20, 20, 20), padding='same', activation='relu')(aInput)
z = MaxPooling3D(pool_size=(10,10,10))(z)
z = Conv3D(64, (20, 20, 20), padding='same', activation='relu')(aInput)
z = MaxPooling3D(pool_size=(10,10,10))(z)
#z = Dropout(0.5)(z)
#z = Conv3D(64, (10, 10, 10), padding='same', activation='relu')(z)
#z = MaxPooling3D(pool_size=(2,2,2))(z)
#z = Dropout(0.5)(z)
z = Conv3D(128, (20, 20, 20), padding='same', activation='relu')(z)
z = MaxPooling3D(pool_size=(2,2,2))(z)
z = Conv3D(128, (20, 20, 20), padding='same', activation='relu')(z)
z = MaxPooling3D(pool_size=(2,2,2))(z)
z = Dropout(0.5)(z)
#z = Conv3D(128, (3, 3, 3), padding='same', activation='relu')(z)
#z = MaxPooling3D(pool_size=(2,2,2))(z)
z = Flatten()(z)
#z = Dropout(0.5)(z)
z = Dense(128, activation='relu')(z)

output1 = Dense(1, activation='sigmoid', name='output1')(z)
output2 = Dense(1, activation='sigmoid', name='output2')(z)

multiModel = Model(aInput, [output1, output2])

opt = tf.keras.optimizers.Adam(learning_rate = 0.01, decay = 1e-6)

multiModel.compile(loss = {'output1': 'mean_squared_error',
                           'output2': 'mean_squared_error'},
                   optimizer = opt,
                   metrics = ['mean_absolute_error'])

chkpt = os.path.join('./', 'MultiCifar10_.{epoch:02d}-{val_loss:.2f}-{val_output1_loss:.2f}-{val_output2_loss:.2f}-{val_output3_loss:.2f}-{val_output4_loss:.2f}.hdf5')
cp_cb = ModelCheckpoint(filepath = chkpt, monitor='val_loss', verbose=1, save_best_only=True, mode='auto')

X_train, X_test, Y_train1, Y_test1, Y_train2, Y_test2 = np.load('./dataset4.npy', allow_pickle = True)

print(X_train)
print(X_test)
print(Y_train1.shape)
print(Y_test1.shape)
print(Y_train2.shape)
print(Y_test2.shape)



model = multiModel.fit(X_train,
                  {'output1': Y_train1,
                   'output2': Y_train2},
                  epochs=50,
                  batch_size=32,
                  validation_data=(X_test,
                                   {'output1': Y_test1,
                                    'output2': Y_test2})
                  #callbacks = [es_cb, cp_cb])
                  )