yolov3系列(二)-keras-yolo3训练自己的数据

训练自己的数据进行目标检测

0.建立相关的目录

在项目根目录下新建VOCdevkit\VOC2007\AnnotationsVOCdevkit\VOC2007\ImageSets\MainVOCdevkit\VOC2007\JPEGImageslogs\000四个目录

1. 使用标注工具labelimg标注数据

链接:https://pan.baidu.com/s/1SO4NqNSfXyKMNCGQA-4LVQ 提取码:ydqi

  • 标注数据
    open dir 选择需要标注的数据目录,change save dir 选择要保存的目录VOCdevkit\VOC2007\Annotations,use default label 勾选,填写一个标签名称,create Rectbox 标注数据,保存即可,如下图
    yolov3系列(二)-keras-yolo3训练自己的数据_第1张图片
    数据标注完成以后,会在VOCdevkit\VOC2007\Annotations目录下生成相关的.xml文件
2. 生成训练集测试集验证集
  • VOC2007目录下新建一个dataShape.py文件,目的是用来分割数据,运行此文件会在VOCdevkit\VOC2007\ImageSets\Main目录下生成test.txt train.txt trainval.txt val.txt四个文件,dataShape.py文件代码如下:
import os
import random

trainval_percent = 0.2
train_percent = 0.8
xmlfilepath = 'Annotations'
txtsavepath = 'ImageSets\Main'
total_xml = os.listdir(xmlfilepath)

num = len(total_xml)
list = range(num)
tv = int(num * trainval_percent)
tr = int(tv * train_percent)
trainval = random.sample(list, tv)
train = random.sample(trainval, tr)

ftrainval = open('ImageSets/Main/trainval.txt', 'w')
ftest = open('ImageSets/Main/test.txt', 'w')
ftrain = open('ImageSets/Main/train.txt', 'w')
fval = open('ImageSets/Main/val.txt', 'w')

for i in list:
   name = total_xml[i][:-4] + '\n'
   if i in trainval:
       ftrainval.write(name)
       if i in train:
           ftest.write(name)
       else:
           fval.write(name)
   else:
       ftrain.write(name)

ftrainval.close()
ftrain.close()
fval.close()
ftest.close()
3.生成yolo3所需的train.txt,val.txt,test.txt

生成的数据集不能供yolov3直接使用。需要运行voc_annotation.py ,classes以检测一个类为例(眼睛),在voc_annotation.py需改你的数据集为:

classes = ["eye"]

运行python voc_annotation.py会生成 2007_train.txt``2007_test.txt``2007_val.txt,把这三个txt文件分别改名为 train.txt``test.txt``val.txt

利用voc制作自己的数据集

4.修改参数文件yolo3.cfg

打开yolo3.cfg文件。搜索yolo(共出现三次),每次按下图都要修改

[convolutional]
size=1
stride=1
pad=1
# filters:3*(5+len(classes));<===> 3*(5+1)
filters=18
activation=linear


[yolo]
mask = 6,7,8
anchors = 10,13,  16,30,  33,23,  30,61,  62,45,  59,119,  116,90,  156,198,  373,326
#  classes: len(classes) = 1,这里是"eye"一类
classes=1
num=9
jitter=.3
ignore_thresh = .5
truth_thresh = 1
# random:改为0
random=0
5.修改model_data下的voc_classes.txt为自己训练的类别
eye
6.生成yolo_anchors.txt文件

运行 python kmeans.py,会在根目录下生成yolo_anchors.txt文件,剪切到 model_data目录下

7.修改train.py代码(用下面代码直接替换原来的代码)

因为 train.py会报错。本人电脑 win10家庭版.坑死人了

"""
Retrain the YOLO model for your own dataset.
"""
import numpy as np
import keras.backend as K
from keras.layers import Input, Lambda
from keras.models import Model
from keras.callbacks import TensorBoard, ModelCheckpoint, EarlyStopping
 
from yolo3.model import preprocess_true_boxes, yolo_body, tiny_yolo_body, yolo_loss
from yolo3.utils import get_random_data
 
 
def _main():
    annotation_path = '2007_train.txt'
    log_dir = 'logs/000/'
    classes_path = 'model_data/voc_classes.txt'
    anchors_path = 'model_data/yolo_anchors.txt'
    class_names = get_classes(classes_path)
    anchors = get_anchors(anchors_path)
    input_shape = (416,416) # multiple of 32, hw
    model = create_model(input_shape, anchors, len(class_names) )
    train(model, annotation_path, input_shape, anchors, len(class_names), log_dir=log_dir)
 
def train(model, annotation_path, input_shape, anchors, num_classes, log_dir='logs/'):
    model.compile(optimizer='adam', loss={
        'yolo_loss': lambda y_true, y_pred: y_pred})
    logging = TensorBoard(log_dir=log_dir)
    checkpoint = ModelCheckpoint(log_dir + "ep{epoch:03d}-loss{loss:.3f}-val_loss{val_loss:.3f}.h5",
        monitor='val_loss', save_weights_only=True, save_best_only=True, period=1)
    batch_size = 10
    val_split = 0.1
    with open(annotation_path) as f:
        lines = f.readlines()
    np.random.shuffle(lines)
    num_val = int(len(lines)*val_split)
    num_train = len(lines) - num_val
    print('Train on {} samples, val on {} samples, with batch size {}.'.format(num_train, num_val, batch_size))
 
    model.fit_generator(data_generator_wrap(lines[:num_train], batch_size, input_shape, anchors, num_classes),
            steps_per_epoch=max(1, num_train//batch_size),
            validation_data=data_generator_wrap(lines[num_train:], batch_size, input_shape, anchors, num_classes),
            validation_steps=max(1, num_val//batch_size),
            epochs=500,
            initial_epoch=0)
    model.save_weights(log_dir + 'trained_weights.h5')
 
def get_classes(classes_path):
    with open(classes_path) as f:
        class_names = f.readlines()
    class_names = [c.strip() for c in class_names]
    return class_names
 
def get_anchors(anchors_path):
    with open(anchors_path) as f:
        anchors = f.readline()
    anchors = [float(x) for x in anchors.split(',')]
    return np.array(anchors).reshape(-1, 2)
 
def create_model(input_shape, anchors, num_classes, load_pretrained=False, freeze_body=False,
            weights_path='model_data/yolo_weights.h5'):
    K.clear_session() # get a new session
    image_input = Input(shape=(None, None, 3))
    h, w = input_shape
    num_anchors = len(anchors)
    y_true = [Input(shape=(h//{0:32, 1:16, 2:8}[l], w//{0:32, 1:16, 2:8}[l], \
        num_anchors//3, num_classes+5)) for l in range(3)]
 
    model_body = yolo_body(image_input, num_anchors//3, num_classes)
    print('Create YOLOv3 model with {} anchors and {} classes.'.format(num_anchors, num_classes))
 
    if load_pretrained:
        model_body.load_weights(weights_path, by_name=True, skip_mismatch=True)
        print('Load weights {}.'.format(weights_path))
        if freeze_body:
            # Do not freeze 3 output layers.
            num = len(model_body.layers)-7
            for i in range(num): model_body.layers[i].trainable = False
            print('Freeze the first {} layers of total {} layers.'.format(num, len(model_body.layers)))
 
    model_loss = Lambda(yolo_loss, output_shape=(1,), name='yolo_loss',
        arguments={'anchors': anchors, 'num_classes': num_classes, 'ignore_thresh': 0.5})(
        [*model_body.output, *y_true])
    model = Model([model_body.input, *y_true], model_loss)
    return model
def data_generator(annotation_lines, batch_size, input_shape, anchors, num_classes):
    n = len(annotation_lines)
    np.random.shuffle(annotation_lines)
    i = 0
    while True:
        image_data = []
        box_data = []
        for b in range(batch_size):
            i %= n
            image, box = get_random_data(annotation_lines[i], input_shape, random=True)
            image_data.append(image)
            box_data.append(box)
            i += 1
        image_data = np.array(image_data)
        box_data = np.array(box_data)
        y_true = preprocess_true_boxes(box_data, input_shape, anchors, num_classes)
        yield [image_data, *y_true], np.zeros(batch_size)
 
def data_generator_wrap(annotation_lines, batch_size, input_shape, anchors, num_classes):
    n = len(annotation_lines)
    if n==0 or batch_size<=0: return None
    return data_generator(annotation_lines, batch_size, input_shape, anchors, num_classes)
 
if __name__ == '__main__':
    _main()
8.生成模型

运行python train.py,会在 logs\000下生成日志文件和trained_weights_stage_1.h5模型文件

9.测试训练效果

把生成的trained_weights_stage_1.h5模型文件,改为yolo.h5,放在 model_data目录下,运行 python yolo_video.py --image,输入图片路径,查看测试效果

使用google cloab训练自己的数据

把batch_size = 6,val_split = 0.1,epochs=500,下面是日志记录

Using TensorFlow backend.
/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/dtypes.py:516: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint8 = np.dtype([("qint8", np.int8, 1)])
/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/dtypes.py:517: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint8 = np.dtype([("quint8", np.uint8, 1)])
/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/dtypes.py:518: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint16 = np.dtype([("qint16", np.int16, 1)])
/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/dtypes.py:519: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint16 = np.dtype([("quint16", np.uint16, 1)])
/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/dtypes.py:520: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint32 = np.dtype([("qint32", np.int32, 1)])
/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/dtypes.py:525: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  np_resource = np.dtype([("resource", np.ubyte, 1)])
/usr/local/lib/python3.6/dist-packages/tensorboard/compat/tensorflow_stub/dtypes.py:541: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint8 = np.dtype([("qint8", np.int8, 1)])
/usr/local/lib/python3.6/dist-packages/tensorboard/compat/tensorflow_stub/dtypes.py:542: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint8 = np.dtype([("quint8", np.uint8, 1)])
/usr/local/lib/python3.6/dist-packages/tensorboard/compat/tensorflow_stub/dtypes.py:543: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint16 = np.dtype([("qint16", np.int16, 1)])
/usr/local/lib/python3.6/dist-packages/tensorboard/compat/tensorflow_stub/dtypes.py:544: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint16 = np.dtype([("quint16", np.uint16, 1)])
/usr/local/lib/python3.6/dist-packages/tensorboard/compat/tensorflow_stub/dtypes.py:545: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint32 = np.dtype([("qint32", np.int32, 1)])
/usr/local/lib/python3.6/dist-packages/tensorboard/compat/tensorflow_stub/dtypes.py:550: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  np_resource = np.dtype([("resource", np.ubyte, 1)])
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:88: The name tf.reset_default_graph is deprecated. Please use tf.compat.v1.reset_default_graph instead.

WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:91: The name tf.placeholder_with_default is deprecated. Please use tf.compat.v1.placeholder_with_default instead.

WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:95: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.

WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:504: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.

WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:3828: The name tf.random_uniform is deprecated. Please use tf.random.uniform instead.

2020-07-29 19:31:48.835165: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
2020-07-29 19:31:48.838145: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 2200000000 Hz
2020-07-29 19:31:48.838323: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x266ca00 executing computations on platform Host. Devices:
2020-07-29 19:31:48.838352: I tensorflow/compiler/xla/service/service.cc:175]   StreamExecutor device (0): , 
2020-07-29 19:31:48.854154: W tensorflow/compiler/jit/mark_for_compilation_pass.cc:1412] (One-time warning): Not using XLA:CPU for cluster because envvar TF_XLA_FLAGS=--tf_xla_cpu_global_jit was not set.  If you want XLA:CPU, either set that envvar, or use experimental_jit_scope to enable XLA:CPU.  To confirm that XLA is active, pass --vmodule=xla_compilation_cache=1 (as a proper command-line flag, not via TF_XLA_FLAGS) or set the envvar XLA_FLAGS=--xla_hlo_profile.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:1794: The name tf.nn.fused_batch_norm is deprecated. Please use tf.compat.v1.nn.fused_batch_norm instead.

WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:1937: The name tf.image.resize_nearest_neighbor is deprecated. Please use tf.compat.v1.image.resize_nearest_neighbor instead.

Create YOLOv3 model with 9 anchors and 3 classes.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:2833: add_dispatch_support..wrapper (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.where in 2.0, which has the same broadcast rule as np.where
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/optimizers.py:744: The name tf.train.Optimizer is deprecated. Please use tf.compat.v1.train.Optimizer instead.

Train on 18 samples, val on 2 samples, with batch size 6.
Epoch 1/500
2020-07-29 19:32:17.293855: E tensorflow/core/grappler/optimizers/meta_optimizer.cc:502] shape_optimizer failed: Invalid argument: Subshape must have computed start >= end since stride is negative, but is 0 and 2 (computed from start 0 and end 9223372036854775807 over shape with rank 2 and stride-1)
2020-07-29 19:32:20.108992: E tensorflow/core/grappler/optimizers/meta_optimizer.cc:502] shape_optimizer failed: Invalid argument: Subshape must have computed start >= end since stride is negative, but is 0 and 2 (computed from start 0 and end 9223372036854775807 over shape with rank 2 and stride-1)
3/3 [==============================] - 114s 38s/step - loss: 8446.7894 - val_loss: 328836546560.0000
Epoch 2/500
3/3 [==============================] - 96s 32s/step - loss: 4472.9715 - val_loss: 6751052288.0000
Epoch 3/500
3/3 [==============================] - 97s 32s/step - loss: 2980.1438 - val_loss: 3714627840.0000
Epoch 4/500
3/3 [==============================] - 96s 32s/step - loss: 1918.1794 - val_loss: 1800320896.0000
Epoch 5/500
3/3 [==============================] - 97s 32s/step - loss: 1456.8024 - val_loss: 60779916.0000
Epoch 6/500
3/3 [==============================] - 97s 32s/step - loss: 1127.4190 - val_loss: 140523376.0000
Epoch 7/500
3/3 [==============================] - 98s 33s/step - loss: 881.5944 - val_loss: 23782522.0000
Epoch 8/500
3/3 [==============================] - 97s 32s/step - loss: 718.2232 - val_loss: 3543072.5000
Epoch 9/500
3/3 [==============================] - 99s 33s/step - loss: 571.5178 - val_loss: 308097.9688
Epoch 10/500
3/3 [==============================] - 99s 33s/step - loss: 485.3580 - val_loss: 1016011.2500
Epoch 11/500
3/3 [==============================] - 98s 33s/step - loss: 428.7773 - val_loss: 705838.2500
Epoch 12/500
3/3 [==============================] - 98s 33s/step - loss: 392.4909 - val_loss: 12528.5156
Epoch 13/500
3/3 [==============================] - 98s 33s/step - loss: 333.6283 - val_loss: 22858.7109
Epoch 14/500
3/3 [==============================] - 99s 33s/step - loss: 298.4900 - val_loss: 407442.0312
Epoch 15/500
3/3 [==============================] - 99s 33s/step - loss: 277.3792 - val_loss: 5225.6387
Epoch 16/500
3/3 [==============================] - 98s 33s/step - loss: 272.6285 - val_loss: 10133.6934
Epoch 17/500
3/3 [==============================] - 98s 33s/step - loss: 236.0809 - val_loss: 5900.5854
Epoch 18/500
3/3 [==============================] - 98s 33s/step - loss: 223.6843 - val_loss: 4326.3848
Epoch 19/500
3/3 [==============================] - 99s 33s/step - loss: 222.1837 - val_loss: 1760.6387
Epoch 20/500
3/3 [==============================] - 99s 33s/step - loss: 200.2016 - val_loss: 1160.3622
Epoch 21/500
3/3 [==============================] - 99s 33s/step - loss: 188.6143 - val_loss: 22739.3555
Epoch 22/500
3/3 [==============================] - 99s 33s/step - loss: 184.1231 - val_loss: 2167.5916
Epoch 23/500
3/3 [==============================] - 99s 33s/step - loss: 175.4343 - val_loss: 1209.4757
Epoch 24/500
3/3 [==============================] - 98s 33s/step - loss: 169.6644 - val_loss: 1372.4573
Epoch 25/500
3/3 [==============================] - 98s 33s/step - loss: 166.9763 - val_loss: 1665.7690
Epoch 26/500
3/3 [==============================] - 98s 33s/step - loss: 160.6180 - val_loss: 1447.7512
Epoch 27/500
3/3 [==============================] - 99s 33s/step - loss: 160.6233 - val_loss: 1375.7462
Epoch 28/500
3/3 [==============================] - 100s 33s/step - loss: 152.4857 - val_loss: 433.5172
Epoch 29/500
3/3 [==============================] - 100s 33s/step - loss: 142.7151 - val_loss: 672.6494
Epoch 30/500
3/3 [==============================] - 100s 33s/step - loss: 145.6465 - val_loss: 559.5533
Epoch 31/500
3/3 [==============================] - 100s 33s/step - loss: 134.4635 - val_loss: 3666.5388
Epoch 32/500
3/3 [==============================] - 99s 33s/step - loss: 136.4725 - val_loss: 507.8727
Epoch 33/500
3/3 [==============================] - 100s 33s/step - loss: 130.5335 - val_loss: 241.9502
Epoch 34/500
3/3 [==============================] - 100s 33s/step - loss: 137.4070 - val_loss: 776.5807
Epoch 35/500
3/3 [==============================] - 100s 33s/step - loss: 131.9995 - val_loss: 183.0690
Epoch 36/500
3/3 [==============================] - 101s 34s/step - loss: 122.3113 - val_loss: 226.1042
Epoch 37/500
3/3 [==============================] - 100s 33s/step - loss: 133.1492 - val_loss: 184.4949
Epoch 38/500
3/3 [==============================] - 101s 34s/step - loss: 119.3009 - val_loss: 216.3880
Epoch 39/500
3/3 [==============================] - 101s 34s/step - loss: 121.0566 - val_loss: 165.3211
Epoch 40/500
3/3 [==============================] - 101s 34s/step - loss: 115.0365 - val_loss: 162.4400
Epoch 41/500
3/3 [==============================] - 101s 34s/step - loss: 117.7454 - val_loss: 160.9371
Epoch 42/500
3/3 [==============================] - 100s 33s/step - loss: 107.5088 - val_loss: 158.3717
Epoch 43/500
3/3 [==============================] - 99s 33s/step - loss: 110.1509 - val_loss: 144.3784
Epoch 44/500
3/3 [==============================] - 98s 33s/step - loss: 103.3773 - val_loss: 226.3996
Epoch 45/500
3/3 [==============================] - 96s 32s/step - loss: 106.3436 - val_loss: 127.7950
Epoch 46/500
3/3 [==============================] - 94s 31s/step - loss: 106.5829 - val_loss: 372.0866
Epoch 47/500
3/3 [==============================] - 95s 32s/step - loss: 113.7368 - val_loss: 151.2786
Epoch 48/500
3/3 [==============================] - 96s 32s/step - loss: 101.2595 - val_loss: 139.4087
Epoch 49/500
3/3 [==============================] - 95s 32s/step - loss: 106.9903 - val_loss: 116.3176
Epoch 50/500
3/3 [==============================] - 93s 31s/step - loss: 102.7912 - val_loss: 113.9480
Epoch 51/500
3/3 [==============================] - 96s 32s/step - loss: 97.7069 - val_loss: 117.8428
Epoch 52/500
3/3 [==============================] - 94s 31s/step - loss: 101.4496 - val_loss: 1598.7411
Epoch 53/500
3/3 [==============================] - 94s 31s/step - loss: 94.6171 - val_loss: 115.2314
Epoch 54/500
3/3 [==============================] - 94s 31s/step - loss: 97.0203 - val_loss: 130.4791
Epoch 55/500
3/3 [==============================] - 94s 31s/step - loss: 100.9805 - val_loss: 124.9349
Epoch 56/500
3/3 [==============================] - 93s 31s/step - loss: 124.1566 - val_loss: 1943.6835
Epoch 57/500
3/3 [==============================] - 93s 31s/step - loss: 95.4099 - val_loss: 268.2999
Epoch 58/500
3/3 [==============================] - 93s 31s/step - loss: 97.0702 - val_loss: 2665.6934
Epoch 59/500
3/3 [==============================] - 93s 31s/step - loss: 94.9395 - val_loss: 149.3464
Epoch 60/500
3/3 [==============================] - 93s 31s/step - loss: 96.6149 - val_loss: 106.7728
Epoch 61/500
3/3 [==============================] - 92s 31s/step - loss: 89.1285 - val_loss: 105.4032
Epoch 62/500
3/3 [==============================] - 91s 30s/step - loss: 89.5165 - val_loss: 110.5668
Epoch 63/500
3/3 [==============================] - 95s 32s/step - loss: 98.1173 - val_loss: 100.6487
Epoch 64/500
3/3 [==============================] - 94s 31s/step - loss: 86.8736 - val_loss: 236.2860
Epoch 65/500
3/3 [==============================] - 93s 31s/step - loss: 91.8959 - val_loss: 1611.6877
Epoch 66/500
3/3 [==============================] - 93s 31s/step - loss: 89.1248 - val_loss: 99.6845
Epoch 67/500
3/3 [==============================] - 94s 31s/step - loss: 85.5816 - val_loss: 104.5171
Epoch 68/500
3/3 [==============================] - 94s 31s/step - loss: 88.6607 - val_loss: 98.7154
Epoch 69/500
3/3 [==============================] - 92s 31s/step - loss: 85.6014 - val_loss: 99.9103
Epoch 70/500
3/3 [==============================] - 93s 31s/step - loss: 82.3052 - val_loss: 101.6201
Epoch 71/500
3/3 [==============================] - 94s 31s/step - loss: 82.6016 - val_loss: 92.8486
Epoch 72/500
3/3 [==============================] - 94s 31s/step - loss: 82.7450 - val_loss: 90.3581
Epoch 73/500
3/3 [==============================] - 95s 32s/step - loss: 81.3549 - val_loss: 88.9613
Epoch 74/500
3/3 [==============================] - 95s 32s/step - loss: 82.3145 - val_loss: 213.6332
Epoch 75/500
3/3 [==============================] - 94s 31s/step - loss: 79.2626 - val_loss: 86.4160
Epoch 76/500
3/3 [==============================] - 93s 31s/step - loss: 81.9035 - val_loss: 87.6611
Epoch 77/500
3/3 [==============================] - 93s 31s/step - loss: 81.4142 - val_loss: 134.0326
Epoch 78/500
3/3 [==============================] - 94s 31s/step - loss: 78.2896 - val_loss: 88.6898
Epoch 79/500
3/3 [==============================] - 94s 31s/step - loss: 77.4351 - val_loss: 85.9944
Epoch 80/500
3/3 [==============================] - 94s 31s/step - loss: 80.7055 - val_loss: 129.6742
Epoch 81/500
3/3 [==============================] - 95s 32s/step - loss: 83.0265 - val_loss: 79.0396
Epoch 82/500
3/3 [==============================] - 93s 31s/step - loss: 79.5868 - val_loss: 74.7130
Epoch 83/500
3/3 [==============================] - 93s 31s/step - loss: 78.5687 - val_loss: 103.1564
Epoch 84/500
3/3 [==============================] - 94s 31s/step - loss: 82.9866 - val_loss: 84.7129
Epoch 85/500
3/3 [==============================] - 95s 32s/step - loss: 81.0399 - val_loss: 84.6765
Epoch 86/500
3/3 [==============================] - 95s 32s/step - loss: 89.4371 - val_loss: 78.7141
Epoch 87/500
3/3 [==============================] - 96s 32s/step - loss: 77.3826 - val_loss: 85.1069
Epoch 88/500
3/3 [==============================] - 94s 31s/step - loss: 78.3961 - val_loss: 90.4308
Epoch 89/500
3/3 [==============================] - 94s 31s/step - loss: 76.3162 - val_loss: 83.5332
Epoch 90/500
3/3 [==============================] - 93s 31s/step - loss: 73.9612 - val_loss: 77.3753
Epoch 91/500
3/3 [==============================] - 93s 31s/step - loss: 76.5251 - val_loss: 77.7148
Epoch 92/500
3/3 [==============================] - 92s 31s/step - loss: 74.6306 - val_loss: 77.9972
Epoch 93/500
3/3 [==============================] - 92s 31s/step - loss: 74.3708 - val_loss: 84.3172
Epoch 94/500
3/3 [==============================] - 92s 31s/step - loss: 74.7169 - val_loss: 78.6280
Epoch 95/500
3/3 [==============================] - 92s 31s/step - loss: 82.1656 - val_loss: 76.0970
Epoch 96/500
3/3 [==============================] - 93s 31s/step - loss: 72.7696 - val_loss: 74.4637
Epoch 97/500
3/3 [==============================] - 93s 31s/step - loss: 72.3184 - val_loss: 75.5271
Epoch 98/500
3/3 [==============================] - 92s 31s/step - loss: 72.6673 - val_loss: 77.0445
Epoch 99/500
3/3 [==============================] - 92s 31s/step - loss: 84.6929 - val_loss: 70.5046
Epoch 100/500
3/3 [==============================] - 92s 31s/step - loss: 71.4358 - val_loss: 72.9689
Epoch 101/500
3/3 [==============================] - 92s 31s/step - loss: 72.5284 - val_loss: 72.9342
Epoch 102/500
3/3 [==============================] - 92s 31s/step - loss: 77.4157 - val_loss: 76.3590
Epoch 103/500
3/3 [==============================] - 91s 30s/step - loss: 72.1802 - val_loss: 72.9116
Epoch 104/500
3/3 [==============================] - 91s 30s/step - loss: 74.4906 - val_loss: 82.2748
Epoch 105/500
3/3 [==============================] - 91s 30s/step - loss: 70.0420 - val_loss: 77.5564
Epoch 106/500
3/3 [==============================] - 92s 31s/step - loss: 72.0894 - val_loss: 75.9684
Epoch 107/500
3/3 [==============================] - 91s 30s/step - loss: 68.5998 - val_loss: 71.8656
Epoch 108/500
3/3 [==============================] - 91s 30s/step - loss: 70.1323 - val_loss: 82.4274
Epoch 109/500
3/3 [==============================] - 91s 30s/step - loss: 67.9585 - val_loss: 70.6375
Epoch 110/500
3/3 [==============================] - 91s 30s/step - loss: 68.9471 - val_loss: 70.4525
Epoch 111/500
3/3 [==============================] - 90s 30s/step - loss: 67.6009 - val_loss: 74.1092
Epoch 112/500
3/3 [==============================] - 91s 30s/step - loss: 69.9344 - val_loss: 72.9113
Epoch 113/500
3/3 [==============================] - 91s 30s/step - loss: 70.3268 - val_loss: 70.1511
Epoch 114/500
3/3 [==============================] - 91s 30s/step - loss: 68.9224 - val_loss: 70.0101
Epoch 115/500
3/3 [==============================] - 91s 30s/step - loss: 71.2282 - val_loss: 70.6972
Epoch 116/500
3/3 [==============================] - 90s 30s/step - loss: 68.4256 - val_loss: 69.6087
Epoch 117/500
3/3 [==============================] - 91s 30s/step - loss: 66.9459 - val_loss: 67.1746
Epoch 118/500
3/3 [==============================] - 92s 31s/step - loss: 68.1223 - val_loss: 67.3573
Epoch 119/500
3/3 [==============================] - 91s 30s/step - loss: 67.0019 - val_loss: 63.1928
Epoch 120/500
3/3 [==============================] - 90s 30s/step - loss: 68.8135 - val_loss: 69.3240
Epoch 121/500
3/3 [==============================] - 90s 30s/step - loss: 68.0982 - val_loss: 70.9299
Epoch 122/500
3/3 [==============================] - 90s 30s/step - loss: 65.8677 - val_loss: 70.2080
Epoch 123/500
3/3 [==============================] - 91s 30s/step - loss: 65.3854 - val_loss: 67.3682
Epoch 124/500
3/3 [==============================] - 90s 30s/step - loss: 66.0033 - val_loss: 69.2807
Epoch 125/500
3/3 [==============================] - 90s 30s/step - loss: 65.6755 - val_loss: 97.9138
Epoch 126/500
3/3 [==============================] - 90s 30s/step - loss: 68.8470 - val_loss: 64.5845
Epoch 127/500
3/3 [==============================] - 90s 30s/step - loss: 65.8492 - val_loss: 67.1290
Epoch 128/500
3/3 [==============================] - 90s 30s/step - loss: 65.9662 - val_loss: 70.1799
Epoch 129/500
3/3 [==============================] - 90s 30s/step - loss: 65.9957 - val_loss: 67.1809
Epoch 130/500
3/3 [==============================] - 90s 30s/step - loss: 67.8415 - val_loss: 79.1722
Epoch 131/500
3/3 [==============================] - 90s 30s/step - loss: 67.1964 - val_loss: 100.1018
Epoch 132/500
3/3 [==============================] - 90s 30s/step - loss: 65.9952 - val_loss: 356.3392
Epoch 133/500
3/3 [==============================] - 89s 30s/step - loss: 67.8345 - val_loss: 66.3755
Epoch 134/500
3/3 [==============================] - 89s 30s/step - loss: 68.1466 - val_loss: 272.1732
Epoch 135/500
3/3 [==============================] - 92s 31s/step - loss: 66.4783 - val_loss: 67.1668
Epoch 136/500
3/3 [==============================] - 96s 32s/step - loss: 70.4941 - val_loss: 209.9447
Epoch 137/500
3/3 [==============================] - 95s 32s/step - loss: 68.0695 - val_loss: 93.2635
Epoch 138/500
3/3 [==============================] - 93s 31s/step - loss: 66.3705 - val_loss: 563.3969
Epoch 139/500
3/3 [==============================] - 93s 31s/step - loss: 66.3544 - val_loss: 197.4445
Epoch 140/500
3/3 [==============================] - 93s 31s/step - loss: 67.4917 - val_loss: 92.1586
Epoch 141/500
3/3 [==============================] - 93s 31s/step - loss: 65.6322 - val_loss: 68.0024
Epoch 142/500
3/3 [==============================] - 94s 31s/step - loss: 65.2297 - val_loss: 83.9944
Epoch 143/500
3/3 [==============================] - 93s 31s/step - loss: 69.0282 - val_loss: 93.9537
Epoch 144/500
3/3 [==============================] - 93s 31s/step - loss: 66.3855 - val_loss: 65.9286
Epoch 145/500
3/3 [==============================] - 95s 32s/step - loss: 74.9715 - val_loss: 67.1622
Epoch 146/500
3/3 [==============================] - 93s 31s/step - loss: 64.7021 - val_loss: 66.9411
Epoch 147/500
3/3 [==============================] - 93s 31s/step - loss: 64.8909 - val_loss: 68.1632
Epoch 148/500
3/3 [==============================] - 92s 31s/step - loss: 65.6112 - val_loss: 73.8619
Epoch 149/500
3/3 [==============================] - 94s 31s/step - loss: 68.6523 - val_loss: 67.7117
Epoch 150/500
3/3 [==============================] - 94s 31s/step - loss: 64.9112 - val_loss: 67.7232
Epoch 151/500
3/3 [==============================] - 94s 31s/step - loss: 64.4555 - val_loss: 115.2624
Epoch 152/500
3/3 [==============================] - 94s 31s/step - loss: 63.8525 - val_loss: 66.5625
Epoch 153/500
3/3 [==============================] - 97s 32s/step - loss: 65.4754 - val_loss: 66.6003
Epoch 154/500
3/3 [==============================] - 95s 32s/step - loss: 66.7031 - val_loss: 61.4111
Epoch 155/500
3/3 [==============================] - 95s 32s/step - loss: 72.1536 - val_loss: 67.0728
Epoch 156/500
3/3 [==============================] - 94s 31s/step - loss: 66.0222 - val_loss: 66.7473
Epoch 157/500
3/3 [==============================] - 94s 31s/step - loss: 65.2591 - val_loss: 64.7544
Epoch 158/500
3/3 [==============================] - 95s 32s/step - loss: 66.8519 - val_loss: 64.7245
Epoch 159/500
3/3 [==============================] - 94s 31s/step - loss: 70.4494 - val_loss: 65.1196
Epoch 160/500
3/3 [==============================] - 98s 33s/step - loss: 63.9866 - val_loss: 66.0846
Epoch 161/500
3/3 [==============================] - 96s 32s/step - loss: 68.5681 - val_loss: 70.9029
Epoch 162/500
3/3 [==============================] - 95s 32s/step - loss: 67.5553 - val_loss: 61.4139
Epoch 163/500
3/3 [==============================] - 94s 31s/step - loss: 63.5885 - val_loss: 72.5325
Epoch 164/500
3/3 [==============================] - 94s 31s/step - loss: 65.2117 - val_loss: 66.1309
Epoch 165/500
3/3 [==============================] - 95s 32s/step - loss: 63.7396 - val_loss: 61.8725
Epoch 166/500
3/3 [==============================] - 94s 31s/step - loss: 64.2908 - val_loss: 64.8733
Epoch 167/500
3/3 [==============================] - 93s 31s/step - loss: 62.2010 - val_loss: 61.0262
Epoch 168/500
3/3 [==============================] - 94s 31s/step - loss: 66.9426 - val_loss: 71.8234
Epoch 169/500
3/3 [==============================] - 94s 31s/step - loss: 61.8032 - val_loss: 75.7612
Epoch 170/500
3/3 [==============================] - 94s 31s/step - loss: 61.5885 - val_loss: 63.8740
Epoch 171/500
3/3 [==============================] - 94s 31s/step - loss: 63.5889 - val_loss: 62.7949
Epoch 172/500
3/3 [==============================] - 94s 31s/step - loss: 62.6449 - val_loss: 63.6134
Epoch 173/500
3/3 [==============================] - 94s 31s/step - loss: 64.1948 - val_loss: 62.7868
Epoch 174/500
3/3 [==============================] - 93s 31s/step - loss: 60.6847 - val_loss: 62.6806
Epoch 175/500
3/3 [==============================] - 94s 31s/step - loss: 62.6874 - val_loss: 63.2594
Epoch 176/500
3/3 [==============================] - 94s 31s/step - loss: 63.4824 - val_loss: 66.3612
Epoch 177/500
3/3 [==============================] - 93s 31s/step - loss: 63.5941 - val_loss: 64.1557
Epoch 178/500
3/3 [==============================] - 98s 33s/step - loss: 63.8656 - val_loss: 76.3857
Epoch 179/500
3/3 [==============================] - 98s 33s/step - loss: 62.4254 - val_loss: 96.7537
Epoch 180/500
3/3 [==============================] - 100s 33s/step - loss: 62.9986 - val_loss: 116.8363
Epoch 181/500
3/3 [==============================] - 99s 33s/step - loss: 61.3592 - val_loss: 62.1869
Epoch 182/500
3/3 [==============================] - 98s 33s/step - loss: 61.7200 - val_loss: 62.6008
Epoch 183/500
3/3 [==============================] - 95s 32s/step - loss: 62.2249 - val_loss: 64.5980
Epoch 184/500
3/3 [==============================] - 93s 31s/step - loss: 63.2679 - val_loss: 251.1574
Epoch 185/500
3/3 [==============================] - 97s 32s/step - loss: 64.9631 - val_loss: 61.8039
Epoch 186/500
3/3 [==============================] - 96s 32s/step - loss: 59.7994 - val_loss: 62.6701
Epoch 187/500
3/3 [==============================] - 94s 31s/step - loss: 62.1102 - val_loss: 62.2131
Epoch 188/500
3/3 [==============================] - 92s 31s/step - loss: 61.7272 - val_loss: 122.8638
Epoch 189/500
3/3 [==============================] - 91s 30s/step - loss: 63.9670 - val_loss: 61.9264
Epoch 190/500
3/3 [==============================] - 90s 30s/step - loss: 63.0687 - val_loss: 181.5622

把batch_size = 8,val_split = 0.1,epochs=1000,最终 - loss: 25.4310 - val_loss: 41.9392 下面是日志记录

Using TensorFlow backend.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:88: The name tf.reset_default_graph is deprecated. Please use tf.compat.v1.reset_default_graph instead.

WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:91: The name tf.placeholder_with_default is deprecated. Please use tf.compat.v1.placeholder_with_default instead.

WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:95: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.

WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:504: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.

WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:3828: The name tf.random_uniform is deprecated. Please use tf.random.uniform instead.

WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:166: The name tf.get_default_session is deprecated. Please use tf.compat.v1.get_default_session instead.

WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:171: The name tf.ConfigProto is deprecated. Please use tf.compat.v1.ConfigProto instead.

WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:176: The name tf.Session is deprecated. Please use tf.compat.v1.Session instead.

2020-08-05 21:36:57.458131: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 AVX512F FMA
2020-08-05 21:36:57.463220: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 2000165000 Hz
2020-08-05 21:36:57.463412: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x30a6bc0 initialized for platform Host (this does not guarantee that XLA will be used). Devices:
2020-08-05 21:36:57.463439: I tensorflow/compiler/xla/service/service.cc:176]   StreamExecutor device (0): Host, Default Version
2020-08-05 21:36:57.465354: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcuda.so.1
2020-08-05 21:36:57.566141: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:983] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-08-05 21:36:57.566853: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x30a6d80 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:
2020-08-05 21:36:57.566946: I tensorflow/compiler/xla/service/service.cc:176]   StreamExecutor device (0): Tesla P100-PCIE-16GB, Compute Capability 6.0
2020-08-05 21:36:57.567133: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:983] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-08-05 21:36:57.567695: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1618] Found device 0 with properties: 
name: Tesla P100-PCIE-16GB major: 6 minor: 0 memoryClockRate(GHz): 1.3285
pciBusID: 0000:00:04.0
2020-08-05 21:36:57.568048: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.0
2020-08-05 21:36:57.569244: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10.0
2020-08-05 21:36:57.570342: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcufft.so.10.0
2020-08-05 21:36:57.570705: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcurand.so.10.0
2020-08-05 21:36:57.571996: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusolver.so.10.0
2020-08-05 21:36:57.572961: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusparse.so.10.0
2020-08-05 21:36:57.575874: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7
2020-08-05 21:36:57.576019: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:983] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-08-05 21:36:57.576564: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:983] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-08-05 21:36:57.577072: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1746] Adding visible gpu devices: 0
2020-08-05 21:36:57.577130: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.0
2020-08-05 21:36:57.578108: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1159] Device interconnect StreamExecutor with strength 1 edge matrix:
2020-08-05 21:36:57.578134: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1165]      0 
2020-08-05 21:36:57.578146: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1178] 0:   N 
2020-08-05 21:36:57.578252: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:983] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-08-05 21:36:57.578794: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:983] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-08-05 21:36:57.579314: W tensorflow/core/common_runtime/gpu/gpu_bfc_allocator.cc:39] Overriding allow_growth setting because the TF_FORCE_GPU_ALLOW_GROWTH environment variable is set. Original config value was 0.
2020-08-05 21:36:57.579352: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1304] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 15216 MB memory) -> physical GPU (device: 0, name: Tesla P100-PCIE-16GB, pci bus id: 0000:00:04.0, compute capability: 6.0)
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:180: The name tf.global_variables is deprecated. Please use tf.compat.v1.global_variables instead.

WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:189: The name tf.is_variable_initialized is deprecated. Please use tf.compat.v1.is_variable_initialized instead.

WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:196: The name tf.variables_initializer is deprecated. Please use tf.compat.v1.variables_initializer instead.

WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:1794: The name tf.nn.fused_batch_norm is deprecated. Please use tf.compat.v1.nn.fused_batch_norm instead.

WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:1937: The name tf.image.resize_nearest_neighbor is deprecated. Please use tf.compat.v1.image.resize_nearest_neighbor instead.

Create YOLOv3 model with 9 anchors and 3 classes.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:1483: The name tf.log is deprecated. Please use tf.math.log instead.

WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:2833: where (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.where in 2.0, which has the same broadcast rule as np.where
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/optimizers.py:744: The name tf.train.Optimizer is deprecated. Please use tf.compat.v1.train.Optimizer instead.

Train on 18 samples, val on 2 samples, with batch size 8.
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:973: The name tf.assign_add is deprecated. Please use tf.compat.v1.assign_add instead.

WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:960: The name tf.assign is deprecated. Please use tf.compat.v1.assign instead.

Epoch 1/1000
2020-08-05 21:37:27.127073: E tensorflow/core/grappler/optimizers/meta_optimizer.cc:533] shape_optimizer failed: Invalid argument: Subshape must have computed start >= end since stride is negative, but is 0 and 2 (computed from start 0 and end 9223372036854775807 over shape with rank 2 and stride-1)
2020-08-05 21:37:28.466607: E tensorflow/core/grappler/optimizers/meta_optimizer.cc:533] layout failed: Invalid argument: Subshape must have computed start >= end since stride is negative, but is 0 and 2 (computed from start 0 and end 9223372036854775807 over shape with rank 2 and stride-1)
2020-08-05 21:37:29.392479: E tensorflow/core/grappler/optimizers/meta_optimizer.cc:533] shape_optimizer failed: Invalid argument: Subshape must have computed start >= end since stride is negative, but is 0 and 2 (computed from start 0 and end 9223372036854775807 over shape with rank 2 and stride-1)
2020-08-05 21:37:31.824659: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7
2020-08-05 21:37:36.162655: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10.0
1/2 [==============>...............] - ETA: 25s - loss: 7969.63332020-08-05 21:37:40.018111: E tensorflow/core/grappler/optimizers/meta_optimizer.cc:533] layout failed: Invalid argument: Subshape must have computed start >= end since stride is negative, but is 0 and 2 (computed from start 0 and end 9223372036854775807 over shape with rank 2 and stride-1)
2/2 [==============================] - 28s 14s/step - loss: 8186.4587 - val_loss: 9534952448.0000
Epoch 2/1000
2/2 [==============================] - 1s 449ms/step - loss: 5702.3152 - val_loss: 597999104.0000
Epoch 3/1000
2/2 [==============================] - 1s 468ms/step - loss: 3672.5740 - val_loss: 99250968.0000
Epoch 4/1000
2/2 [==============================] - 1s 449ms/step - loss: 2216.3652 - val_loss: 684495232.0000
Epoch 5/1000
2/2 [==============================] - 1s 440ms/step - loss: 1649.7612 - val_loss: 536749312.0000
Epoch 6/1000
2/2 [==============================] - 1s 449ms/step - loss: 1223.2501 - val_loss: 196764608.0000
Epoch 7/1000
2/2 [==============================] - 1s 455ms/step - loss: 979.0498 - val_loss: 139056080.0000
Epoch 8/1000
2/2 [==============================] - 1s 445ms/step - loss: 849.5832 - val_loss: 55919740.0000
Epoch 9/1000
2/2 [==============================] - 1s 444ms/step - loss: 734.8343 - val_loss: 35384504.0000
Epoch 10/1000
2/2 [==============================] - 1s 510ms/step - loss: 643.9365 - val_loss: 19303322.0000
Epoch 11/1000
2/2 [==============================] - 2s 808ms/step - loss: 570.7336 - val_loss: 4804616.0000
Epoch 12/1000
2/2 [==============================] - 2s 781ms/step - loss: 468.1741 - val_loss: 26623774.0000
Epoch 13/1000
2/2 [==============================] - 1s 711ms/step - loss: 446.2462 - val_loss: 26038038.0000
Epoch 14/1000
2/2 [==============================] - 1s 689ms/step - loss: 420.7665 - val_loss: 36308384.0000
Epoch 15/1000
2/2 [==============================] - 1s 697ms/step - loss: 375.3553 - val_loss: 708763.4375
Epoch 16/1000
2/2 [==============================] - 1s 713ms/step - loss: 340.0721 - val_loss: 928544.2500
Epoch 17/1000
2/2 [==============================] - 1s 695ms/step - loss: 318.7858 - val_loss: 278976.7812
Epoch 18/1000
2/2 [==============================] - 1s 694ms/step - loss: 300.6269 - val_loss: 126673.4375
Epoch 19/1000
2/2 [==============================] - 1s 703ms/step - loss: 283.8869 - val_loss: 50764.3086
Epoch 20/1000
2/2 [==============================] - 1s 721ms/step - loss: 271.6725 - val_loss: 9123.7217
Epoch 21/1000
2/2 [==============================] - 1s 721ms/step - loss: 253.4683 - val_loss: 12756.3857
Epoch 22/1000
2/2 [==============================] - 1s 701ms/step - loss: 239.0525 - val_loss: 3118.6313
Epoch 23/1000
2/2 [==============================] - 1s 698ms/step - loss: 232.6324 - val_loss: 7248.1943
Epoch 24/1000
2/2 [==============================] - 1s 698ms/step - loss: 222.7768 - val_loss: 2424.2925
Epoch 25/1000
2/2 [==============================] - 1s 688ms/step - loss: 224.5528 - val_loss: 1614.3586
Epoch 26/1000
2/2 [==============================] - 1s 700ms/step - loss: 205.3776 - val_loss: 1558.3833
Epoch 27/1000
2/2 [==============================] - 1s 707ms/step - loss: 204.4061 - val_loss: 914.1429
Epoch 28/1000
2/2 [==============================] - 1s 717ms/step - loss: 196.4786 - val_loss: 474.5502
Epoch 29/1000
2/2 [==============================] - 1s 706ms/step - loss: 197.7677 - val_loss: 518.0853
Epoch 30/1000
2/2 [==============================] - 1s 739ms/step - loss: 182.0629 - val_loss: 390.2279
Epoch 31/1000
2/2 [==============================] - 1s 704ms/step - loss: 176.1717 - val_loss: 288.8252
Epoch 32/1000
2/2 [==============================] - 1s 693ms/step - loss: 170.6895 - val_loss: 288.8852
Epoch 33/1000
2/2 [==============================] - 1s 746ms/step - loss: 168.1491 - val_loss: 267.1690
Epoch 34/1000
2/2 [==============================] - 1s 692ms/step - loss: 161.4012 - val_loss: 251.2336
Epoch 35/1000
2/2 [==============================] - 1s 702ms/step - loss: 157.9353 - val_loss: 219.1266
Epoch 36/1000
2/2 [==============================] - 1s 709ms/step - loss: 172.8404 - val_loss: 1249.4637
Epoch 37/1000
2/2 [==============================] - 1s 702ms/step - loss: 151.7123 - val_loss: 8604.3398
Epoch 38/1000
2/2 [==============================] - 1s 711ms/step - loss: 151.2136 - val_loss: 20819.5625
Epoch 39/1000
2/2 [==============================] - 1s 721ms/step - loss: 155.8932 - val_loss: 23781.6289
Epoch 40/1000
2/2 [==============================] - 1s 717ms/step - loss: 147.5706 - val_loss: 24381.8828
Epoch 41/1000
2/2 [==============================] - 1s 713ms/step - loss: 142.8667 - val_loss: 16587.2383
Epoch 42/1000
2/2 [==============================] - 2s 753ms/step - loss: 141.3436 - val_loss: 19190.6465
Epoch 43/1000
2/2 [==============================] - 1s 722ms/step - loss: 143.7613 - val_loss: 16876.6777
Epoch 44/1000
2/2 [==============================] - 1s 695ms/step - loss: 144.3512 - val_loss: 64942.7227
Epoch 45/1000
2/2 [==============================] - 1s 685ms/step - loss: 135.1258 - val_loss: 20435.7539
Epoch 46/1000
2/2 [==============================] - 1s 728ms/step - loss: 131.0828 - val_loss: 15262.2695
Epoch 47/1000
2/2 [==============================] - 1s 708ms/step - loss: 128.2111 - val_loss: 5933.7202
Epoch 48/1000
2/2 [==============================] - 1s 710ms/step - loss: 128.4950 - val_loss: 2783.9941
Epoch 49/1000
2/2 [==============================] - 1s 704ms/step - loss: 124.4590 - val_loss: 3383.4919
Epoch 50/1000
2/2 [==============================] - 1s 729ms/step - loss: 124.6741 - val_loss: 1555.5051
Epoch 51/1000
2/2 [==============================] - 1s 736ms/step - loss: 118.6213 - val_loss: 2725.6091
Epoch 52/1000
2/2 [==============================] - 1s 715ms/step - loss: 118.0993 - val_loss: 7863.4556
Epoch 53/1000
2/2 [==============================] - 1s 705ms/step - loss: 115.9478 - val_loss: 9228.0283
Epoch 54/1000
2/2 [==============================] - 1s 737ms/step - loss: 114.4714 - val_loss: 2254.4880
Epoch 55/1000
2/2 [==============================] - 1s 722ms/step - loss: 115.2809 - val_loss: 4547.3027
Epoch 56/1000
2/2 [==============================] - 1s 719ms/step - loss: 112.6010 - val_loss: 803.7517
Epoch 57/1000
2/2 [==============================] - 1s 719ms/step - loss: 109.3934 - val_loss: 3072.1636
Epoch 58/1000
2/2 [==============================] - 1s 685ms/step - loss: 108.6409 - val_loss: 2237.6448
Epoch 59/1000
2/2 [==============================] - 1s 693ms/step - loss: 105.4296 - val_loss: 766.6730
Epoch 60/1000
2/2 [==============================] - 1s 723ms/step - loss: 105.3314 - val_loss: 1921.3658
Epoch 61/1000
2/2 [==============================] - 1s 691ms/step - loss: 103.5984 - val_loss: 503.1118
Epoch 62/1000
2/2 [==============================] - 1s 695ms/step - loss: 104.0284 - val_loss: 880.3438
Epoch 63/1000
2/2 [==============================] - 1s 735ms/step - loss: 102.5103 - val_loss: 310.0865
Epoch 64/1000
2/2 [==============================] - 1s 693ms/step - loss: 99.9862 - val_loss: 339.5742
Epoch 65/1000
2/2 [==============================] - 1s 701ms/step - loss: 100.3449 - val_loss: 544.6689
Epoch 66/1000
2/2 [==============================] - 1s 696ms/step - loss: 98.8916 - val_loss: 659.7013
Epoch 67/1000
2/2 [==============================] - 1s 685ms/step - loss: 97.9051 - val_loss: 298.4399
Epoch 68/1000
2/2 [==============================] - 1s 694ms/step - loss: 98.4892 - val_loss: 354.9012
Epoch 69/1000
2/2 [==============================] - 1s 729ms/step - loss: 96.5579 - val_loss: 317.0739
Epoch 70/1000
2/2 [==============================] - 1s 698ms/step - loss: 96.8138 - val_loss: 284.7998
Epoch 71/1000
2/2 [==============================] - 1s 672ms/step - loss: 99.2281 - val_loss: 238.4481
Epoch 72/1000
2/2 [==============================] - 1s 723ms/step - loss: 90.4005 - val_loss: 183.3866
Epoch 73/1000
2/2 [==============================] - 1s 704ms/step - loss: 89.4818 - val_loss: 108.3206
Epoch 74/1000
2/2 [==============================] - 1s 729ms/step - loss: 91.9687 - val_loss: 112.7051
Epoch 75/1000
2/2 [==============================] - 1s 700ms/step - loss: 89.1224 - val_loss: 117.0770
Epoch 76/1000
2/2 [==============================] - 1s 674ms/step - loss: 89.4737 - val_loss: 120.0273
Epoch 77/1000
2/2 [==============================] - 1s 740ms/step - loss: 89.7691 - val_loss: 114.9730
Epoch 78/1000
2/2 [==============================] - 1s 691ms/step - loss: 87.6388 - val_loss: 144.8410
Epoch 79/1000
2/2 [==============================] - 1s 693ms/step - loss: 86.1340 - val_loss: 125.3056
Epoch 80/1000
2/2 [==============================] - 1s 699ms/step - loss: 87.3309 - val_loss: 111.5640
Epoch 81/1000
2/2 [==============================] - 1s 685ms/step - loss: 86.7782 - val_loss: 102.8090
Epoch 82/1000
2/2 [==============================] - 1s 720ms/step - loss: 84.4730 - val_loss: 105.0989
Epoch 83/1000
2/2 [==============================] - 1s 699ms/step - loss: 85.6442 - val_loss: 113.2263
Epoch 84/1000
2/2 [==============================] - 1s 706ms/step - loss: 86.4309 - val_loss: 103.5953
Epoch 85/1000
2/2 [==============================] - 1s 717ms/step - loss: 85.1090 - val_loss: 101.2634
Epoch 86/1000
2/2 [==============================] - 1s 711ms/step - loss: 82.6330 - val_loss: 106.6024
Epoch 87/1000
2/2 [==============================] - 1s 718ms/step - loss: 82.1430 - val_loss: 96.0940
Epoch 88/1000
2/2 [==============================] - 1s 699ms/step - loss: 86.2355 - val_loss: 431.7940
Epoch 89/1000
2/2 [==============================] - 1s 710ms/step - loss: 82.4486 - val_loss: 5900.9771
Epoch 90/1000
2/2 [==============================] - 1s 700ms/step - loss: 83.2474 - val_loss: 3927.3020
Epoch 91/1000
2/2 [==============================] - 1s 711ms/step - loss: 82.4303 - val_loss: 14808.2012
Epoch 92/1000
2/2 [==============================] - 1s 704ms/step - loss: 81.2858 - val_loss: 25279.6191
Epoch 93/1000
2/2 [==============================] - 1s 680ms/step - loss: 81.8035 - val_loss: 22490.1660
Epoch 94/1000
2/2 [==============================] - 1s 709ms/step - loss: 84.3035 - val_loss: 14307.6260
Epoch 95/1000
2/2 [==============================] - 1s 698ms/step - loss: 80.2626 - val_loss: 45136.9961
Epoch 96/1000
2/2 [==============================] - 1s 689ms/step - loss: 79.8200 - val_loss: 219.0539
Epoch 97/1000
2/2 [==============================] - 1s 690ms/step - loss: 77.5758 - val_loss: 19271.7383
Epoch 98/1000
2/2 [==============================] - 1s 700ms/step - loss: 82.2283 - val_loss: 7148.6919
Epoch 99/1000
2/2 [==============================] - 1s 718ms/step - loss: 77.0820 - val_loss: 29114.7246
Epoch 100/1000
2/2 [==============================] - 1s 738ms/step - loss: 77.9773 - val_loss: 4779.7168
Epoch 101/1000
2/2 [==============================] - 1s 718ms/step - loss: 78.4817 - val_loss: 1758.8824
Epoch 102/1000
2/2 [==============================] - 1s 703ms/step - loss: 76.6394 - val_loss: 2086.5754
Epoch 103/1000
2/2 [==============================] - 1s 721ms/step - loss: 76.8728 - val_loss: 1025.0924
Epoch 104/1000
2/2 [==============================] - 1s 689ms/step - loss: 75.1968 - val_loss: 723.8922
Epoch 105/1000
2/2 [==============================] - 1s 733ms/step - loss: 77.3439 - val_loss: 266.9564
Epoch 106/1000
2/2 [==============================] - 1s 700ms/step - loss: 76.3785 - val_loss: 98.1485
Epoch 107/1000
2/2 [==============================] - 1s 707ms/step - loss: 76.3574 - val_loss: 82.4798
Epoch 108/1000
2/2 [==============================] - 1s 714ms/step - loss: 77.1764 - val_loss: 2312.3916
Epoch 109/1000
2/2 [==============================] - 1s 720ms/step - loss: 77.1834 - val_loss: 7767.4282
Epoch 110/1000
2/2 [==============================] - 1s 733ms/step - loss: 78.0006 - val_loss: 2437.9241
Epoch 111/1000
2/2 [==============================] - 1s 740ms/step - loss: 75.9033 - val_loss: 270.6180
Epoch 112/1000
2/2 [==============================] - 1s 718ms/step - loss: 75.0501 - val_loss: 206.5292
Epoch 113/1000
2/2 [==============================] - 1s 724ms/step - loss: 74.8012 - val_loss: 264.1358
Epoch 114/1000
2/2 [==============================] - 1s 727ms/step - loss: 79.1943 - val_loss: 133.2687
Epoch 115/1000
2/2 [==============================] - 1s 702ms/step - loss: 76.8605 - val_loss: 175.3550
Epoch 116/1000
2/2 [==============================] - 1s 712ms/step - loss: 76.8013 - val_loss: 276.8629
Epoch 117/1000
2/2 [==============================] - 1s 691ms/step - loss: 73.1108 - val_loss: 341.7842
Epoch 118/1000
2/2 [==============================] - 1s 727ms/step - loss: 75.1429 - val_loss: 242.8858
Epoch 119/1000
2/2 [==============================] - 1s 709ms/step - loss: 73.6428 - val_loss: 192.3394
Epoch 120/1000
2/2 [==============================] - 1s 686ms/step - loss: 74.8524 - val_loss: 116.0671
Epoch 121/1000
2/2 [==============================] - 1s 711ms/step - loss: 71.8896 - val_loss: 100.4062
Epoch 122/1000
2/2 [==============================] - 1s 712ms/step - loss: 77.1144 - val_loss: 99.0121
Epoch 123/1000
2/2 [==============================] - 1s 694ms/step - loss: 72.3892 - val_loss: 110.7798
Epoch 124/1000
2/2 [==============================] - 1s 702ms/step - loss: 70.8538 - val_loss: 87.8876
Epoch 125/1000
2/2 [==============================] - 1s 722ms/step - loss: 71.9307 - val_loss: 93.1757
Epoch 126/1000
2/2 [==============================] - 1s 701ms/step - loss: 73.0780 - val_loss: 94.5864
Epoch 127/1000
2/2 [==============================] - 1s 722ms/step - loss: 69.1555 - val_loss: 74.2723
Epoch 128/1000
2/2 [==============================] - 1s 701ms/step - loss: 70.0682 - val_loss: 87.9805
Epoch 129/1000
2/2 [==============================] - 1s 712ms/step - loss: 68.2077 - val_loss: 81.3937
Epoch 130/1000
2/2 [==============================] - 1s 719ms/step - loss: 67.7026 - val_loss: 81.9447
Epoch 131/1000
2/2 [==============================] - 1s 698ms/step - loss: 69.7414 - val_loss: 77.6003
Epoch 132/1000
2/2 [==============================] - 1s 700ms/step - loss: 68.8834 - val_loss: 77.4561
Epoch 133/1000
2/2 [==============================] - 1s 695ms/step - loss: 71.0164 - val_loss: 89.4072
Epoch 134/1000
2/2 [==============================] - 1s 709ms/step - loss: 66.7966 - val_loss: 85.3141
Epoch 135/1000
2/2 [==============================] - 1s 715ms/step - loss: 67.9731 - val_loss: 80.6536
Epoch 136/1000
2/2 [==============================] - 1s 741ms/step - loss: 67.4988 - val_loss: 83.9438
Epoch 137/1000
2/2 [==============================] - 1s 682ms/step - loss: 68.2809 - val_loss: 76.8331
Epoch 138/1000
2/2 [==============================] - 1s 705ms/step - loss: 66.0533 - val_loss: 73.8218
Epoch 139/1000
2/2 [==============================] - 1s 715ms/step - loss: 68.1797 - val_loss: 81.4507
Epoch 140/1000
2/2 [==============================] - 1s 698ms/step - loss: 64.2035 - val_loss: 80.0064
Epoch 141/1000
2/2 [==============================] - 1s 710ms/step - loss: 65.7754 - val_loss: 76.4766
Epoch 142/1000
2/2 [==============================] - 1s 699ms/step - loss: 68.4295 - val_loss: 75.2618
Epoch 143/1000
2/2 [==============================] - 1s 691ms/step - loss: 69.3593 - val_loss: 72.8651
Epoch 144/1000
2/2 [==============================] - 1s 692ms/step - loss: 65.4228 - val_loss: 72.6723
Epoch 145/1000
2/2 [==============================] - 1s 707ms/step - loss: 66.1651 - val_loss: 77.3613
Epoch 146/1000
2/2 [==============================] - 1s 692ms/step - loss: 62.0737 - val_loss: 69.9946
Epoch 147/1000
2/2 [==============================] - 1s 708ms/step - loss: 64.4143 - val_loss: 75.8825
Epoch 148/1000
2/2 [==============================] - 1s 701ms/step - loss: 66.4123 - val_loss: 76.6905
Epoch 149/1000
2/2 [==============================] - 1s 686ms/step - loss: 63.2083 - val_loss: 80.5138
Epoch 150/1000
2/2 [==============================] - 1s 732ms/step - loss: 65.5307 - val_loss: 77.5580
Epoch 151/1000
2/2 [==============================] - 1s 691ms/step - loss: 61.5735 - val_loss: 74.5618
Epoch 152/1000
2/2 [==============================] - 1s 710ms/step - loss: 63.8837 - val_loss: 72.1070
Epoch 153/1000
2/2 [==============================] - 1s 737ms/step - loss: 63.1886 - val_loss: 79.2590
Epoch 154/1000
2/2 [==============================] - 1s 674ms/step - loss: 62.3140 - val_loss: 69.1958
Epoch 155/1000
2/2 [==============================] - 1s 713ms/step - loss: 62.5876 - val_loss: 68.6972
Epoch 156/1000
2/2 [==============================] - 1s 706ms/step - loss: 65.4333 - val_loss: 69.1168
Epoch 157/1000
2/2 [==============================] - 1s 712ms/step - loss: 63.7094 - val_loss: 73.9093
Epoch 158/1000
2/2 [==============================] - 1s 721ms/step - loss: 63.0607 - val_loss: 62.2182
Epoch 159/1000
2/2 [==============================] - 1s 710ms/step - loss: 64.8581 - val_loss: 66.9144
Epoch 160/1000
2/2 [==============================] - 1s 690ms/step - loss: 68.1950 - val_loss: 75.4925
Epoch 161/1000
2/2 [==============================] - 1s 708ms/step - loss: 63.6817 - val_loss: 73.7491
Epoch 162/1000
2/2 [==============================] - 1s 701ms/step - loss: 62.7927 - val_loss: 81.1486
Epoch 163/1000
2/2 [==============================] - 1s 691ms/step - loss: 62.6768 - val_loss: 70.0830
Epoch 164/1000
2/2 [==============================] - 1s 708ms/step - loss: 58.8765 - val_loss: 69.8806
Epoch 165/1000
2/2 [==============================] - 1s 727ms/step - loss: 61.0019 - val_loss: 63.3081
Epoch 166/1000
2/2 [==============================] - 1s 679ms/step - loss: 62.0177 - val_loss: 73.4355
Epoch 167/1000
2/2 [==============================] - 1s 698ms/step - loss: 59.2929 - val_loss: 71.7453
Epoch 168/1000
2/2 [==============================] - 1s 703ms/step - loss: 59.0190 - val_loss: 68.3251
Epoch 169/1000
2/2 [==============================] - 1s 728ms/step - loss: 59.0588 - val_loss: 67.9200
Epoch 170/1000
2/2 [==============================] - 1s 696ms/step - loss: 57.3374 - val_loss: 76.8149
Epoch 171/1000
2/2 [==============================] - 1s 679ms/step - loss: 59.7184 - val_loss: 101.1476
Epoch 172/1000
2/2 [==============================] - 1s 693ms/step - loss: 59.7741 - val_loss: 62.3451
Epoch 173/1000
2/2 [==============================] - 1s 717ms/step - loss: 60.8730 - val_loss: 73.7490
Epoch 174/1000
2/2 [==============================] - 1s 701ms/step - loss: 60.5413 - val_loss: 82.7301
Epoch 175/1000
2/2 [==============================] - 1s 705ms/step - loss: 61.8753 - val_loss: 63.3600
Epoch 176/1000
2/2 [==============================] - 1s 700ms/step - loss: 59.3824 - val_loss: 72.7750
Epoch 177/1000
2/2 [==============================] - 1s 712ms/step - loss: 58.7562 - val_loss: 79.5509
Epoch 178/1000
2/2 [==============================] - 1s 670ms/step - loss: 88.6343 - val_loss: 83.2231
Epoch 179/1000
2/2 [==============================] - 1s 711ms/step - loss: 61.7254 - val_loss: 89.9067
Epoch 180/1000
2/2 [==============================] - 1s 704ms/step - loss: 60.2445 - val_loss: 87.9576
Epoch 181/1000
2/2 [==============================] - 1s 701ms/step - loss: 63.7083 - val_loss: 89.0508
Epoch 182/1000
2/2 [==============================] - 1s 710ms/step - loss: 60.8309 - val_loss: 108.2797
Epoch 183/1000
2/2 [==============================] - 1s 683ms/step - loss: 60.5114 - val_loss: 88.0969
Epoch 184/1000
2/2 [==============================] - 1s 690ms/step - loss: 61.3794 - val_loss: 98.1869
Epoch 185/1000
2/2 [==============================] - 1s 690ms/step - loss: 60.6632 - val_loss: 73.6376
Epoch 186/1000
2/2 [==============================] - 1s 706ms/step - loss: 65.2189 - val_loss: 114.6802
Epoch 187/1000
2/2 [==============================] - 1s 705ms/step - loss: 60.5351 - val_loss: 149.3211
Epoch 188/1000
2/2 [==============================] - 1s 678ms/step - loss: 62.4969 - val_loss: 402.2162
Epoch 189/1000
2/2 [==============================] - 2s 750ms/step - loss: 66.4559 - val_loss: 84.9707
Epoch 190/1000
2/2 [==============================] - 1s 695ms/step - loss: 61.6406 - val_loss: 104.4164
Epoch 191/1000
2/2 [==============================] - 1s 669ms/step - loss: 63.0275 - val_loss: 528.4590
Epoch 192/1000
2/2 [==============================] - 1s 736ms/step - loss: 56.8533 - val_loss: 256.3582
Epoch 193/1000
2/2 [==============================] - 1s 665ms/step - loss: 59.6143 - val_loss: 403.5498
Epoch 194/1000
2/2 [==============================] - 1s 686ms/step - loss: 69.4881 - val_loss: 78.1647
Epoch 195/1000
2/2 [==============================] - 1s 698ms/step - loss: 60.2627 - val_loss: 97.6945
Epoch 196/1000
2/2 [==============================] - 1s 693ms/step - loss: 64.9644 - val_loss: 292.2778
Epoch 197/1000
2/2 [==============================] - 1s 703ms/step - loss: 58.8772 - val_loss: 398.9035
Epoch 198/1000
2/2 [==============================] - 1s 672ms/step - loss: 57.9382 - val_loss: 243.6344
Epoch 199/1000
2/2 [==============================] - 1s 715ms/step - loss: 74.7384 - val_loss: 105.3438
Epoch 200/1000
2/2 [==============================] - 1s 706ms/step - loss: 59.9442 - val_loss: 128.1539
Epoch 201/1000
2/2 [==============================] - 1s 694ms/step - loss: 62.1294 - val_loss: 3552.6963
Epoch 202/1000
2/2 [==============================] - 1s 687ms/step - loss: 76.4816 - val_loss: 5201.3730
Epoch 203/1000
2/2 [==============================] - 1s 705ms/step - loss: 65.5638 - val_loss: 114.1406
Epoch 204/1000
2/2 [==============================] - 1s 694ms/step - loss: 60.3018 - val_loss: 250.0608
Epoch 205/1000
2/2 [==============================] - 1s 690ms/step - loss: 62.8755 - val_loss: 181.4681
Epoch 206/1000
2/2 [==============================] - 1s 696ms/step - loss: 61.2556 - val_loss: 169.6718
Epoch 207/1000
2/2 [==============================] - 1s 711ms/step - loss: 57.4117 - val_loss: 149.8103
Epoch 208/1000
2/2 [==============================] - 1s 743ms/step - loss: 58.2789 - val_loss: 221.2277
Epoch 209/1000
2/2 [==============================] - 1s 686ms/step - loss: 58.3614 - val_loss: 172.3034
Epoch 210/1000
2/2 [==============================] - 1s 707ms/step - loss: 58.4972 - val_loss: 117.7860
Epoch 211/1000
2/2 [==============================] - 1s 690ms/step - loss: 61.8050 - val_loss: 107.3047
Epoch 212/1000
2/2 [==============================] - 1s 734ms/step - loss: 57.7628 - val_loss: 196.1572
Epoch 213/1000
2/2 [==============================] - 1s 741ms/step - loss: 60.6579 - val_loss: 1948.8420
Epoch 214/1000
2/2 [==============================] - 1s 709ms/step - loss: 79.4055 - val_loss: 12504.8789
Epoch 215/1000
2/2 [==============================] - 1s 685ms/step - loss: 62.8874 - val_loss: 7257.0649
Epoch 216/1000
2/2 [==============================] - 1s 716ms/step - loss: 59.7865 - val_loss: 23739.9551
Epoch 217/1000
2/2 [==============================] - 1s 709ms/step - loss: 61.7270 - val_loss: 61060.4609
Epoch 218/1000
2/2 [==============================] - 1s 699ms/step - loss: 63.2554 - val_loss: 11329.4082
Epoch 219/1000
2/2 [==============================] - 1s 692ms/step - loss: 59.5646 - val_loss: 17769.6699
Epoch 220/1000
2/2 [==============================] - 1s 714ms/step - loss: 60.1251 - val_loss: 23452.6250
Epoch 221/1000
2/2 [==============================] - 1s 695ms/step - loss: 63.4970 - val_loss: 15721.3350
Epoch 222/1000
2/2 [==============================] - 1s 691ms/step - loss: 55.3300 - val_loss: 51875.7109
Epoch 223/1000
2/2 [==============================] - 1s 703ms/step - loss: 55.7939 - val_loss: 325.8849
Epoch 224/1000
2/2 [==============================] - 1s 697ms/step - loss: 61.0418 - val_loss: 21709.6016
Epoch 225/1000
2/2 [==============================] - 1s 711ms/step - loss: 56.6303 - val_loss: 7023.8618
Epoch 226/1000
2/2 [==============================] - 1s 711ms/step - loss: 59.4141 - val_loss: 27701.3535
Epoch 227/1000
2/2 [==============================] - 1s 719ms/step - loss: 55.2657 - val_loss: 4767.0205
Epoch 228/1000
2/2 [==============================] - 1s 705ms/step - loss: 55.3487 - val_loss: 8603.0254
Epoch 229/1000
2/2 [==============================] - 1s 688ms/step - loss: 59.7111 - val_loss: 3367.8643
Epoch 230/1000
2/2 [==============================] - 1s 694ms/step - loss: 54.9894 - val_loss: 2993.4731
Epoch 231/1000
2/2 [==============================] - 1s 716ms/step - loss: 57.5861 - val_loss: 1207.6735
Epoch 232/1000
2/2 [==============================] - 1s 711ms/step - loss: 59.4704 - val_loss: 368.2351
Epoch 233/1000
2/2 [==============================] - 1s 689ms/step - loss: 59.3151 - val_loss: 254.8345
Epoch 234/1000
2/2 [==============================] - 1s 702ms/step - loss: 56.5996 - val_loss: 192.3918
Epoch 235/1000
2/2 [==============================] - 1s 683ms/step - loss: 60.5428 - val_loss: 416.6565
Epoch 236/1000
2/2 [==============================] - 1s 709ms/step - loss: 54.6318 - val_loss: 974.1688
Epoch 237/1000
2/2 [==============================] - 1s 702ms/step - loss: 57.6328 - val_loss: 111.1767
Epoch 238/1000
2/2 [==============================] - 1s 688ms/step - loss: 54.5258 - val_loss: 522.5461
Epoch 239/1000
2/2 [==============================] - 1s 709ms/step - loss: 58.4756 - val_loss: 298.7509
Epoch 240/1000
2/2 [==============================] - 1s 691ms/step - loss: 54.9827 - val_loss: 635.8406
Epoch 241/1000
2/2 [==============================] - 1s 696ms/step - loss: 57.2686 - val_loss: 241.7990
Epoch 242/1000
2/2 [==============================] - 1s 694ms/step - loss: 53.6040 - val_loss: 154.0756
Epoch 243/1000
2/2 [==============================] - 1s 698ms/step - loss: 52.6420 - val_loss: 81.3568
Epoch 244/1000
2/2 [==============================] - 1s 720ms/step - loss: 58.0524 - val_loss: 68.1273
Epoch 245/1000
2/2 [==============================] - 1s 687ms/step - loss: 56.5608 - val_loss: 90.6462
Epoch 246/1000
2/2 [==============================] - 1s 703ms/step - loss: 56.1809 - val_loss: 116.5314
Epoch 247/1000
2/2 [==============================] - 1s 685ms/step - loss: 53.2724 - val_loss: 104.8160
Epoch 248/1000
2/2 [==============================] - 1s 699ms/step - loss: 52.4322 - val_loss: 109.3930
Epoch 249/1000
2/2 [==============================] - 1s 704ms/step - loss: 52.9678 - val_loss: 91.6722
Epoch 250/1000
2/2 [==============================] - 1s 691ms/step - loss: 56.1327 - val_loss: 70.4001
Epoch 251/1000
2/2 [==============================] - 1s 693ms/step - loss: 51.4660 - val_loss: 109.3429
Epoch 252/1000
2/2 [==============================] - 1s 708ms/step - loss: 56.5912 - val_loss: 132.7803
Epoch 253/1000
2/2 [==============================] - 1s 688ms/step - loss: 54.3512 - val_loss: 136.4538
Epoch 254/1000
2/2 [==============================] - 1s 692ms/step - loss: 56.0544 - val_loss: 117.7168
Epoch 255/1000
2/2 [==============================] - 1s 741ms/step - loss: 52.0592 - val_loss: 61.3104
Epoch 256/1000
2/2 [==============================] - 1s 710ms/step - loss: 52.5562 - val_loss: 75.5092
Epoch 257/1000
2/2 [==============================] - 1s 696ms/step - loss: 51.4375 - val_loss: 112.9265
Epoch 258/1000
2/2 [==============================] - 1s 689ms/step - loss: 52.8557 - val_loss: 59.3239
Epoch 259/1000
2/2 [==============================] - 1s 693ms/step - loss: 53.4036 - val_loss: 57.8893
Epoch 260/1000
2/2 [==============================] - 1s 726ms/step - loss: 49.7026 - val_loss: 82.3798
Epoch 261/1000
2/2 [==============================] - 1s 667ms/step - loss: 56.8266 - val_loss: 89.1497
Epoch 262/1000
2/2 [==============================] - 1s 683ms/step - loss: 53.3663 - val_loss: 74.0474
Epoch 263/1000
2/2 [==============================] - 1s 707ms/step - loss: 52.5214 - val_loss: 80.5810
Epoch 264/1000
2/2 [==============================] - 1s 686ms/step - loss: 49.1434 - val_loss: 57.4133
Epoch 265/1000
2/2 [==============================] - 1s 694ms/step - loss: 52.9569 - val_loss: 59.9578
Epoch 266/1000
2/2 [==============================] - 1s 689ms/step - loss: 55.9096 - val_loss: 68.8102
Epoch 267/1000
2/2 [==============================] - 1s 682ms/step - loss: 52.1110 - val_loss: 59.2936
Epoch 268/1000
2/2 [==============================] - 1s 674ms/step - loss: 50.3630 - val_loss: 56.4216
Epoch 269/1000
2/2 [==============================] - 1s 699ms/step - loss: 53.9528 - val_loss: 60.1277
Epoch 270/1000
2/2 [==============================] - 1s 693ms/step - loss: 53.1277 - val_loss: 54.1851
Epoch 271/1000
2/2 [==============================] - 1s 705ms/step - loss: 55.0275 - val_loss: 58.0377
Epoch 272/1000
2/2 [==============================] - 1s 728ms/step - loss: 49.9111 - val_loss: 54.3251
Epoch 273/1000
2/2 [==============================] - 1s 697ms/step - loss: 53.3347 - val_loss: 52.2859
Epoch 274/1000
2/2 [==============================] - 1s 684ms/step - loss: 54.2701 - val_loss: 56.8513
Epoch 275/1000
2/2 [==============================] - 1s 741ms/step - loss: 50.8474 - val_loss: 57.1234
Epoch 276/1000
2/2 [==============================] - 1s 688ms/step - loss: 52.1873 - val_loss: 61.4195
Epoch 277/1000
2/2 [==============================] - 1s 720ms/step - loss: 49.8746 - val_loss: 58.1829
Epoch 278/1000
2/2 [==============================] - 1s 690ms/step - loss: 52.1085 - val_loss: 58.3174
Epoch 279/1000
2/2 [==============================] - 1s 686ms/step - loss: 53.9259 - val_loss: 69.7742
Epoch 280/1000
2/2 [==============================] - 1s 684ms/step - loss: 54.1144 - val_loss: 59.4125
Epoch 281/1000
2/2 [==============================] - 1s 684ms/step - loss: 51.2623 - val_loss: 60.1210
Epoch 282/1000
2/2 [==============================] - 1s 717ms/step - loss: 52.9054 - val_loss: 60.5561
Epoch 283/1000
2/2 [==============================] - 1s 686ms/step - loss: 48.6429 - val_loss: 52.7351
Epoch 284/1000
2/2 [==============================] - 1s 743ms/step - loss: 48.7632 - val_loss: 57.0283
Epoch 285/1000
2/2 [==============================] - 1s 695ms/step - loss: 48.1715 - val_loss: 53.7158
Epoch 286/1000
2/2 [==============================] - 1s 697ms/step - loss: 47.9568 - val_loss: 52.0483
Epoch 287/1000
2/2 [==============================] - 1s 668ms/step - loss: 51.1057 - val_loss: 60.4001
Epoch 288/1000
2/2 [==============================] - 1s 721ms/step - loss: 51.1609 - val_loss: 51.3995
Epoch 289/1000
2/2 [==============================] - 1s 688ms/step - loss: 51.6477 - val_loss: 54.2394
Epoch 290/1000
2/2 [==============================] - 1s 711ms/step - loss: 51.0582 - val_loss: 55.1861
Epoch 291/1000
2/2 [==============================] - 1s 709ms/step - loss: 47.2263 - val_loss: 55.8537
Epoch 292/1000
2/2 [==============================] - 1s 702ms/step - loss: 52.6189 - val_loss: 54.8989
Epoch 293/1000
2/2 [==============================] - 1s 715ms/step - loss: 50.5812 - val_loss: 56.3028
Epoch 294/1000
2/2 [==============================] - 1s 704ms/step - loss: 50.7975 - val_loss: 59.0524
Epoch 295/1000
2/2 [==============================] - 1s 691ms/step - loss: 50.9159 - val_loss: 53.9619
Epoch 296/1000
2/2 [==============================] - 1s 698ms/step - loss: 50.6242 - val_loss: 55.0031
Epoch 297/1000
2/2 [==============================] - 1s 692ms/step - loss: 51.5263 - val_loss: 57.2521
Epoch 298/1000
2/2 [==============================] - 1s 688ms/step - loss: 49.8356 - val_loss: 56.3303
Epoch 299/1000
2/2 [==============================] - 1s 723ms/step - loss: 52.3118 - val_loss: 54.4299
Epoch 300/1000
2/2 [==============================] - 1s 700ms/step - loss: 46.8096 - val_loss: 58.0258
Epoch 301/1000
2/2 [==============================] - 1s 690ms/step - loss: 48.9000 - val_loss: 53.8135
Epoch 302/1000
2/2 [==============================] - 1s 722ms/step - loss: 51.5692 - val_loss: 57.0236
Epoch 303/1000
2/2 [==============================] - 1s 700ms/step - loss: 52.4983 - val_loss: 54.5679
Epoch 304/1000
2/2 [==============================] - 1s 695ms/step - loss: 49.6317 - val_loss: 56.8486
Epoch 305/1000
2/2 [==============================] - 1s 703ms/step - loss: 50.7997 - val_loss: 53.5918
Epoch 306/1000
2/2 [==============================] - 1s 702ms/step - loss: 47.1099 - val_loss: 56.9564
Epoch 307/1000
2/2 [==============================] - 1s 676ms/step - loss: 51.8578 - val_loss: 55.6393
Epoch 308/1000
2/2 [==============================] - 1s 680ms/step - loss: 48.1095 - val_loss: 57.5407
Epoch 309/1000
2/2 [==============================] - 1s 719ms/step - loss: 49.6183 - val_loss: 54.0441
Epoch 310/1000
2/2 [==============================] - 1s 691ms/step - loss: 48.2078 - val_loss: 55.3195
Epoch 311/1000
2/2 [==============================] - 1s 678ms/step - loss: 53.9805 - val_loss: 52.1699
Epoch 312/1000
2/2 [==============================] - 1s 672ms/step - loss: 51.8439 - val_loss: 52.3915
Epoch 313/1000
2/2 [==============================] - 1s 702ms/step - loss: 49.6062 - val_loss: 58.5782
Epoch 314/1000
2/2 [==============================] - 1s 696ms/step - loss: 49.7764 - val_loss: 53.1823
Epoch 315/1000
2/2 [==============================] - 1s 688ms/step - loss: 49.8202 - val_loss: 57.0323
Epoch 316/1000
2/2 [==============================] - 1s 702ms/step - loss: 48.6666 - val_loss: 57.2278
Epoch 317/1000
2/2 [==============================] - 1s 674ms/step - loss: 50.7374 - val_loss: 54.4665
Epoch 318/1000
2/2 [==============================] - 1s 680ms/step - loss: 49.8838 - val_loss: 57.6369
Epoch 319/1000
2/2 [==============================] - 1s 683ms/step - loss: 49.5271 - val_loss: 59.7637
Epoch 320/1000
2/2 [==============================] - 1s 660ms/step - loss: 52.8834 - val_loss: 59.4877
Epoch 321/1000
2/2 [==============================] - 1s 681ms/step - loss: 53.8509 - val_loss: 57.7161
Epoch 322/1000
2/2 [==============================] - 1s 676ms/step - loss: 50.1003 - val_loss: 59.1504
Epoch 323/1000
2/2 [==============================] - 1s 716ms/step - loss: 47.8362 - val_loss: 58.7599
Epoch 324/1000
2/2 [==============================] - 1s 671ms/step - loss: 54.0127 - val_loss: 59.4073
Epoch 325/1000
2/2 [==============================] - 1s 694ms/step - loss: 47.1544 - val_loss: 54.9719
Epoch 326/1000
2/2 [==============================] - 1s 675ms/step - loss: 49.4163 - val_loss: 56.5511
Epoch 327/1000
2/2 [==============================] - 1s 716ms/step - loss: 48.3715 - val_loss: 59.6960
Epoch 328/1000
2/2 [==============================] - 1s 704ms/step - loss: 50.4742 - val_loss: 58.4887
Epoch 329/1000
2/2 [==============================] - 1s 690ms/step - loss: 48.5780 - val_loss: 57.6500
Epoch 330/1000
2/2 [==============================] - 1s 695ms/step - loss: 50.5480 - val_loss: 58.8726
Epoch 331/1000
2/2 [==============================] - 1s 687ms/step - loss: 47.7796 - val_loss: 57.3859
Epoch 332/1000
2/2 [==============================] - 1s 697ms/step - loss: 49.8573 - val_loss: 57.1050
Epoch 333/1000
2/2 [==============================] - 1s 724ms/step - loss: 47.3477 - val_loss: 58.3561
Epoch 334/1000
2/2 [==============================] - 1s 683ms/step - loss: 51.8071 - val_loss: 56.5254
Epoch 335/1000
2/2 [==============================] - 1s 684ms/step - loss: 45.9009 - val_loss: 54.2488
Epoch 336/1000
2/2 [==============================] - 1s 680ms/step - loss: 52.4549 - val_loss: 55.4710
Epoch 337/1000
2/2 [==============================] - 1s 706ms/step - loss: 48.1972 - val_loss: 53.7606
Epoch 338/1000
2/2 [==============================] - 1s 693ms/step - loss: 51.2362 - val_loss: 53.2250
Epoch 339/1000
2/2 [==============================] - 1s 677ms/step - loss: 48.3991 - val_loss: 53.8296
Epoch 340/1000
2/2 [==============================] - 1s 710ms/step - loss: 50.8009 - val_loss: 52.9559
Epoch 341/1000
2/2 [==============================] - 1s 716ms/step - loss: 48.7019 - val_loss: 58.3466
Epoch 342/1000
2/2 [==============================] - 1s 703ms/step - loss: 47.1880 - val_loss: 51.9359
Epoch 343/1000
2/2 [==============================] - 1s 686ms/step - loss: 46.1623 - val_loss: 52.3608
Epoch 344/1000
2/2 [==============================] - 1s 720ms/step - loss: 45.6359 - val_loss: 60.2318
Epoch 345/1000
2/2 [==============================] - 1s 675ms/step - loss: 45.9361 - val_loss: 54.6231
Epoch 346/1000
2/2 [==============================] - 1s 697ms/step - loss: 45.1348 - val_loss: 51.5880
Epoch 347/1000
2/2 [==============================] - 1s 681ms/step - loss: 46.5780 - val_loss: 51.0613
Epoch 348/1000
2/2 [==============================] - 1s 686ms/step - loss: 48.3601 - val_loss: 56.2566
Epoch 349/1000
2/2 [==============================] - 1s 713ms/step - loss: 47.5172 - val_loss: 54.2813
Epoch 350/1000
2/2 [==============================] - 1s 669ms/step - loss: 49.2896 - val_loss: 61.6554
Epoch 351/1000
2/2 [==============================] - 1s 718ms/step - loss: 46.1563 - val_loss: 53.0881
Epoch 352/1000
2/2 [==============================] - 1s 679ms/step - loss: 49.8480 - val_loss: 54.1130
Epoch 353/1000
2/2 [==============================] - 1s 669ms/step - loss: 47.5613 - val_loss: 56.8230
Epoch 354/1000
2/2 [==============================] - 1s 692ms/step - loss: 48.5034 - val_loss: 52.4317
Epoch 355/1000
2/2 [==============================] - 1s 700ms/step - loss: 46.5847 - val_loss: 50.8114
Epoch 356/1000
2/2 [==============================] - 1s 699ms/step - loss: 48.8825 - val_loss: 53.1471
Epoch 357/1000
2/2 [==============================] - 1s 671ms/step - loss: 45.3614 - val_loss: 53.6789
Epoch 358/1000
2/2 [==============================] - 1s 690ms/step - loss: 47.3564 - val_loss: 47.5798
Epoch 359/1000
2/2 [==============================] - 1s 701ms/step - loss: 50.8374 - val_loss: 55.4264
Epoch 360/1000
2/2 [==============================] - 1s 702ms/step - loss: 46.3671 - val_loss: 51.8793
Epoch 361/1000
2/2 [==============================] - 1s 730ms/step - loss: 48.7494 - val_loss: 55.6850
Epoch 362/1000
2/2 [==============================] - 1s 684ms/step - loss: 46.7247 - val_loss: 54.2583
Epoch 363/1000
2/2 [==============================] - 1s 698ms/step - loss: 47.9700 - val_loss: 52.8345
Epoch 364/1000
2/2 [==============================] - 1s 657ms/step - loss: 45.4318 - val_loss: 51.1504
Epoch 365/1000
2/2 [==============================] - 1s 703ms/step - loss: 44.2894 - val_loss: 51.0256
Epoch 366/1000
2/2 [==============================] - 1s 664ms/step - loss: 44.0876 - val_loss: 54.1137
Epoch 367/1000
2/2 [==============================] - 1s 682ms/step - loss: 44.7409 - val_loss: 54.5598
Epoch 368/1000
2/2 [==============================] - 1s 695ms/step - loss: 44.9804 - val_loss: 57.1743
Epoch 369/1000
2/2 [==============================] - 1s 676ms/step - loss: 47.4558 - val_loss: 54.5004
Epoch 370/1000
2/2 [==============================] - 1s 697ms/step - loss: 43.2216 - val_loss: 53.7530
Epoch 371/1000
2/2 [==============================] - 1s 695ms/step - loss: 43.1975 - val_loss: 50.1553
Epoch 372/1000
2/2 [==============================] - 1s 691ms/step - loss: 42.0838 - val_loss: 55.1354
Epoch 373/1000
2/2 [==============================] - 1s 691ms/step - loss: 43.1547 - val_loss: 55.4836
Epoch 374/1000
2/2 [==============================] - 1s 678ms/step - loss: 48.2671 - val_loss: 54.2077
Epoch 375/1000
2/2 [==============================] - 1s 676ms/step - loss: 43.4116 - val_loss: 57.7872
Epoch 376/1000
2/2 [==============================] - 1s 682ms/step - loss: 43.7885 - val_loss: 52.2729
Epoch 377/1000
2/2 [==============================] - 1s 679ms/step - loss: 46.8950 - val_loss: 55.9999
Epoch 378/1000
2/2 [==============================] - 1s 686ms/step - loss: 50.8122 - val_loss: 51.2072
Epoch 379/1000
2/2 [==============================] - 1s 708ms/step - loss: 45.0814 - val_loss: 56.6879
Epoch 380/1000
2/2 [==============================] - 1s 715ms/step - loss: 47.9717 - val_loss: 55.7220
Epoch 381/1000
2/2 [==============================] - 1s 689ms/step - loss: 46.5620 - val_loss: 57.1321
Epoch 382/1000
2/2 [==============================] - 1s 683ms/step - loss: 47.7367 - val_loss: 53.9666
Epoch 383/1000
2/2 [==============================] - 1s 679ms/step - loss: 47.7409 - val_loss: 52.5368
Epoch 384/1000
2/2 [==============================] - 1s 680ms/step - loss: 43.6771 - val_loss: 55.7053
Epoch 385/1000
2/2 [==============================] - 1s 722ms/step - loss: 44.8171 - val_loss: 50.3364
Epoch 386/1000
2/2 [==============================] - 1s 655ms/step - loss: 45.7791 - val_loss: 51.8528
Epoch 387/1000
2/2 [==============================] - 1s 715ms/step - loss: 45.3418 - val_loss: 57.7935
Epoch 388/1000
2/2 [==============================] - 1s 693ms/step - loss: 45.4774 - val_loss: 53.1141
Epoch 389/1000
2/2 [==============================] - 1s 695ms/step - loss: 41.7331 - val_loss: 53.8495
Epoch 390/1000
2/2 [==============================] - 1s 681ms/step - loss: 48.5116 - val_loss: 52.9551
Epoch 391/1000
2/2 [==============================] - 1s 698ms/step - loss: 45.7480 - val_loss: 55.5822
Epoch 392/1000
2/2 [==============================] - 1s 707ms/step - loss: 41.6466 - val_loss: 56.1475
Epoch 393/1000
2/2 [==============================] - 1s 669ms/step - loss: 47.4867 - val_loss: 54.8596
Epoch 394/1000
2/2 [==============================] - 1s 684ms/step - loss: 44.8001 - val_loss: 54.9624
Epoch 395/1000
2/2 [==============================] - 1s 726ms/step - loss: 40.3447 - val_loss: 53.3772
Epoch 396/1000
2/2 [==============================] - 1s 682ms/step - loss: 44.6380 - val_loss: 54.4784
Epoch 397/1000
2/2 [==============================] - 1s 666ms/step - loss: 43.8974 - val_loss: 50.5411
Epoch 398/1000
2/2 [==============================] - 1s 707ms/step - loss: 46.3774 - val_loss: 50.9153
Epoch 399/1000
2/2 [==============================] - 1s 680ms/step - loss: 44.8756 - val_loss: 55.1611
Epoch 400/1000
2/2 [==============================] - 1s 712ms/step - loss: 45.6362 - val_loss: 54.5463
Epoch 401/1000
2/2 [==============================] - 1s 686ms/step - loss: 46.8470 - val_loss: 53.7373
Epoch 402/1000
2/2 [==============================] - 1s 681ms/step - loss: 47.1026 - val_loss: 48.7956
Epoch 403/1000
2/2 [==============================] - 1s 653ms/step - loss: 47.2564 - val_loss: 51.3838
Epoch 404/1000
2/2 [==============================] - 1s 709ms/step - loss: 46.2076 - val_loss: 49.3888
Epoch 405/1000
2/2 [==============================] - 1s 702ms/step - loss: 44.2578 - val_loss: 52.2157
Epoch 406/1000
2/2 [==============================] - 1s 691ms/step - loss: 43.9020 - val_loss: 52.8849
Epoch 407/1000
2/2 [==============================] - 1s 686ms/step - loss: 50.7434 - val_loss: 50.8852
Epoch 408/1000
2/2 [==============================] - 1s 699ms/step - loss: 42.8908 - val_loss: 51.2977
Epoch 409/1000
2/2 [==============================] - 1s 686ms/step - loss: 41.4731 - val_loss: 56.8227
Epoch 410/1000
2/2 [==============================] - 1s 688ms/step - loss: 47.2221 - val_loss: 56.5785
Epoch 411/1000
2/2 [==============================] - 1s 687ms/step - loss: 43.1729 - val_loss: 52.1136
Epoch 412/1000
2/2 [==============================] - 1s 713ms/step - loss: 44.8202 - val_loss: 55.4007
Epoch 413/1000
2/2 [==============================] - 1s 690ms/step - loss: 42.3773 - val_loss: 47.5705
Epoch 414/1000
2/2 [==============================] - 1s 702ms/step - loss: 44.7139 - val_loss: 44.9783
Epoch 415/1000
2/2 [==============================] - 1s 701ms/step - loss: 44.2456 - val_loss: 60.2501
Epoch 416/1000
2/2 [==============================] - 1s 688ms/step - loss: 40.7747 - val_loss: 49.3095
Epoch 417/1000
2/2 [==============================] - 1s 686ms/step - loss: 43.7354 - val_loss: 54.7731
Epoch 418/1000
2/2 [==============================] - 1s 716ms/step - loss: 42.0471 - val_loss: 55.3689
Epoch 419/1000
2/2 [==============================] - 1s 673ms/step - loss: 42.7402 - val_loss: 52.3465
Epoch 420/1000
2/2 [==============================] - 1s 666ms/step - loss: 45.9687 - val_loss: 49.3432
Epoch 421/1000
2/2 [==============================] - 1s 681ms/step - loss: 42.8463 - val_loss: 49.9990
Epoch 422/1000
2/2 [==============================] - 1s 666ms/step - loss: 46.9708 - val_loss: 58.1547
Epoch 423/1000
2/2 [==============================] - 1s 670ms/step - loss: 42.0595 - val_loss: 51.3352
Epoch 424/1000
2/2 [==============================] - 1s 713ms/step - loss: 42.1335 - val_loss: 55.4496
Epoch 425/1000
2/2 [==============================] - 1s 698ms/step - loss: 43.4115 - val_loss: 51.4946
Epoch 426/1000
2/2 [==============================] - 1s 678ms/step - loss: 40.1357 - val_loss: 58.2421
Epoch 427/1000
2/2 [==============================] - 1s 696ms/step - loss: 46.3034 - val_loss: 54.8397
Epoch 428/1000
2/2 [==============================] - 1s 723ms/step - loss: 37.6315 - val_loss: 51.8335
Epoch 429/1000
2/2 [==============================] - 1s 663ms/step - loss: 47.4493 - val_loss: 47.1941
Epoch 430/1000
2/2 [==============================] - 1s 702ms/step - loss: 45.8689 - val_loss: 46.6865
Epoch 431/1000
2/2 [==============================] - 1s 658ms/step - loss: 42.0644 - val_loss: 46.5267
Epoch 432/1000
2/2 [==============================] - 1s 656ms/step - loss: 44.6724 - val_loss: 51.5658
Epoch 433/1000
2/2 [==============================] - 1s 679ms/step - loss: 43.5221 - val_loss: 56.9247
Epoch 434/1000
2/2 [==============================] - 1s 677ms/step - loss: 42.1624 - val_loss: 53.4914
Epoch 435/1000
2/2 [==============================] - 1s 681ms/step - loss: 41.4022 - val_loss: 50.0241
Epoch 436/1000
2/2 [==============================] - 1s 695ms/step - loss: 45.8576 - val_loss: 56.3042
Epoch 437/1000
2/2 [==============================] - 1s 687ms/step - loss: 44.8952 - val_loss: 54.8909
Epoch 438/1000
2/2 [==============================] - 1s 666ms/step - loss: 45.5130 - val_loss: 48.8161
Epoch 439/1000
2/2 [==============================] - 1s 678ms/step - loss: 41.9149 - val_loss: 59.5367
Epoch 440/1000
2/2 [==============================] - 1s 686ms/step - loss: 43.1573 - val_loss: 51.9587
Epoch 441/1000
2/2 [==============================] - 1s 699ms/step - loss: 45.2810 - val_loss: 56.0813
Epoch 442/1000
2/2 [==============================] - 1s 702ms/step - loss: 43.8960 - val_loss: 52.8776
Epoch 443/1000
2/2 [==============================] - 1s 688ms/step - loss: 40.3606 - val_loss: 56.1284
Epoch 444/1000
2/2 [==============================] - 1s 696ms/step - loss: 42.3231 - val_loss: 51.3984
Epoch 445/1000
2/2 [==============================] - 1s 677ms/step - loss: 41.6964 - val_loss: 60.9919
Epoch 446/1000
2/2 [==============================] - 1s 692ms/step - loss: 41.1567 - val_loss: 62.0165
Epoch 447/1000
2/2 [==============================] - 1s 667ms/step - loss: 40.3131 - val_loss: 59.1302
Epoch 448/1000
2/2 [==============================] - 1s 673ms/step - loss: 40.1935 - val_loss: 52.0091
Epoch 449/1000
2/2 [==============================] - 1s 678ms/step - loss: 44.8293 - val_loss: 57.5943
Epoch 450/1000
2/2 [==============================] - 1s 704ms/step - loss: 44.0889 - val_loss: 49.9251
Epoch 451/1000
2/2 [==============================] - 1s 699ms/step - loss: 43.7645 - val_loss: 52.2732
Epoch 452/1000
2/2 [==============================] - 1s 679ms/step - loss: 42.9228 - val_loss: 53.6828
Epoch 453/1000
2/2 [==============================] - 1s 677ms/step - loss: 42.7482 - val_loss: 49.2607
Epoch 454/1000
2/2 [==============================] - 1s 706ms/step - loss: 40.7342 - val_loss: 53.8881
Epoch 455/1000
2/2 [==============================] - 1s 695ms/step - loss: 40.7345 - val_loss: 48.0677
Epoch 456/1000
2/2 [==============================] - 1s 680ms/step - loss: 41.6022 - val_loss: 49.3977
Epoch 457/1000
2/2 [==============================] - 1s 674ms/step - loss: 40.2518 - val_loss: 50.7591
Epoch 458/1000
2/2 [==============================] - 1s 693ms/step - loss: 41.5903 - val_loss: 49.4567
Epoch 459/1000
2/2 [==============================] - 1s 700ms/step - loss: 39.6693 - val_loss: 48.5292
Epoch 460/1000
2/2 [==============================] - 1s 688ms/step - loss: 43.0216 - val_loss: 51.5665
Epoch 461/1000
2/2 [==============================] - 1s 686ms/step - loss: 39.8130 - val_loss: 46.5828
Epoch 462/1000
2/2 [==============================] - 1s 719ms/step - loss: 41.5893 - val_loss: 51.6068
Epoch 463/1000
2/2 [==============================] - 1s 686ms/step - loss: 39.1208 - val_loss: 51.6566
Epoch 464/1000
2/2 [==============================] - 1s 696ms/step - loss: 37.8145 - val_loss: 51.8526
Epoch 465/1000
2/2 [==============================] - 1s 660ms/step - loss: 37.8272 - val_loss: 59.6309
Epoch 466/1000
2/2 [==============================] - 1s 720ms/step - loss: 41.3560 - val_loss: 57.3544
Epoch 467/1000
2/2 [==============================] - 1s 715ms/step - loss: 39.2753 - val_loss: 50.4395
Epoch 468/1000
2/2 [==============================] - 1s 675ms/step - loss: 39.3742 - val_loss: 51.2484
Epoch 469/1000
2/2 [==============================] - 1s 678ms/step - loss: 38.0770 - val_loss: 48.0660
Epoch 470/1000
2/2 [==============================] - 1s 687ms/step - loss: 37.9763 - val_loss: 53.7813
Epoch 471/1000
2/2 [==============================] - 1s 697ms/step - loss: 38.4766 - val_loss: 52.9159
Epoch 472/1000
2/2 [==============================] - 1s 733ms/step - loss: 41.7668 - val_loss: 51.8154
Epoch 473/1000
2/2 [==============================] - 1s 672ms/step - loss: 41.8332 - val_loss: 48.1136
Epoch 474/1000
2/2 [==============================] - 1s 699ms/step - loss: 35.4069 - val_loss: 53.5075
Epoch 475/1000
2/2 [==============================] - 1s 670ms/step - loss: 35.7083 - val_loss: 49.8751
Epoch 476/1000
2/2 [==============================] - 1s 692ms/step - loss: 40.0931 - val_loss: 47.3973
Epoch 477/1000
2/2 [==============================] - 1s 691ms/step - loss: 39.5437 - val_loss: 48.2688
Epoch 478/1000
2/2 [==============================] - 1s 680ms/step - loss: 38.1044 - val_loss: 47.7504
Epoch 479/1000
2/2 [==============================] - 1s 692ms/step - loss: 38.1685 - val_loss: 46.5336
Epoch 480/1000
2/2 [==============================] - 1s 670ms/step - loss: 35.7607 - val_loss: 48.7737
Epoch 481/1000
2/2 [==============================] - 1s 686ms/step - loss: 37.2051 - val_loss: 52.8284
Epoch 482/1000
2/2 [==============================] - 1s 681ms/step - loss: 39.2557 - val_loss: 45.4980
Epoch 483/1000
2/2 [==============================] - 1s 687ms/step - loss: 38.6313 - val_loss: 46.3936
Epoch 484/1000
2/2 [==============================] - 1s 689ms/step - loss: 36.9187 - val_loss: 48.6597
Epoch 485/1000
2/2 [==============================] - 1s 732ms/step - loss: 39.0153 - val_loss: 48.3393
Epoch 486/1000
2/2 [==============================] - 1s 670ms/step - loss: 44.7889 - val_loss: 46.5220
Epoch 487/1000
2/2 [==============================] - 1s 696ms/step - loss: 43.1694 - val_loss: 53.0441
Epoch 488/1000
2/2 [==============================] - 1s 676ms/step - loss: 38.5157 - val_loss: 48.0706
Epoch 489/1000
2/2 [==============================] - 1s 693ms/step - loss: 38.1375 - val_loss: 49.3065
Epoch 490/1000
2/2 [==============================] - 1s 654ms/step - loss: 38.1864 - val_loss: 46.7871
Epoch 491/1000
2/2 [==============================] - 1s 699ms/step - loss: 38.9397 - val_loss: 49.3081
Epoch 492/1000
2/2 [==============================] - 1s 709ms/step - loss: 36.1494 - val_loss: 48.2080
Epoch 493/1000
2/2 [==============================] - 1s 664ms/step - loss: 39.8549 - val_loss: 50.0019
Epoch 494/1000
2/2 [==============================] - 1s 693ms/step - loss: 35.8085 - val_loss: 50.6910
Epoch 495/1000
2/2 [==============================] - 1s 673ms/step - loss: 36.5182 - val_loss: 49.9071
Epoch 496/1000
2/2 [==============================] - 1s 726ms/step - loss: 39.0902 - val_loss: 51.8748
Epoch 497/1000
2/2 [==============================] - 1s 699ms/step - loss: 38.2428 - val_loss: 50.2728
Epoch 498/1000
2/2 [==============================] - 1s 678ms/step - loss: 42.0463 - val_loss: 44.8220
Epoch 499/1000
2/2 [==============================] - 1s 675ms/step - loss: 39.1546 - val_loss: 47.9653
Epoch 500/1000
2/2 [==============================] - 1s 667ms/step - loss: 40.0544 - val_loss: 46.1104
Epoch 501/1000
2/2 [==============================] - 1s 704ms/step - loss: 37.4243 - val_loss: 46.0463
Epoch 502/1000
2/2 [==============================] - 1s 682ms/step - loss: 45.0485 - val_loss: 49.6297
Epoch 503/1000
2/2 [==============================] - 1s 708ms/step - loss: 36.1906 - val_loss: 47.9873
Epoch 504/1000
2/2 [==============================] - 1s 705ms/step - loss: 38.4932 - val_loss: 48.3210
Epoch 505/1000
2/2 [==============================] - 1s 702ms/step - loss: 43.2418 - val_loss: 52.7294
Epoch 506/1000
2/2 [==============================] - 1s 688ms/step - loss: 35.7126 - val_loss: 59.0199
Epoch 507/1000
2/2 [==============================] - 1s 691ms/step - loss: 34.8668 - val_loss: 59.4352
Epoch 508/1000
2/2 [==============================] - 1s 712ms/step - loss: 37.9326 - val_loss: 51.7351
Epoch 509/1000
2/2 [==============================] - 1s 668ms/step - loss: 40.0281 - val_loss: 58.1157
Epoch 510/1000
2/2 [==============================] - 1s 713ms/step - loss: 38.9676 - val_loss: 56.8955
Epoch 511/1000
2/2 [==============================] - 1s 684ms/step - loss: 38.5099 - val_loss: 50.8544
Epoch 512/1000
2/2 [==============================] - 1s 700ms/step - loss: 34.4537 - val_loss: 51.1837
Epoch 513/1000
2/2 [==============================] - 1s 696ms/step - loss: 37.2780 - val_loss: 47.0678
Epoch 514/1000
2/2 [==============================] - 1s 698ms/step - loss: 36.8927 - val_loss: 46.5135
Epoch 515/1000
2/2 [==============================] - 1s 719ms/step - loss: 35.7765 - val_loss: 50.1114
Epoch 516/1000
2/2 [==============================] - 1s 712ms/step - loss: 38.6742 - val_loss: 47.5188
Epoch 517/1000
2/2 [==============================] - 1s 684ms/step - loss: 37.3726 - val_loss: 51.2360
Epoch 518/1000
2/2 [==============================] - 1s 693ms/step - loss: 37.6826 - val_loss: 51.5218
Epoch 519/1000
2/2 [==============================] - 1s 682ms/step - loss: 33.6008 - val_loss: 48.8102
Epoch 520/1000
2/2 [==============================] - 1s 677ms/step - loss: 33.6523 - val_loss: 54.5394
Epoch 521/1000
2/2 [==============================] - 1s 678ms/step - loss: 39.2950 - val_loss: 52.1814
Epoch 522/1000
2/2 [==============================] - 1s 714ms/step - loss: 33.3672 - val_loss: 52.2608
Epoch 523/1000
2/2 [==============================] - 1s 668ms/step - loss: 39.4467 - val_loss: 53.8641
Epoch 524/1000
2/2 [==============================] - 1s 689ms/step - loss: 36.3641 - val_loss: 45.1890
Epoch 525/1000
2/2 [==============================] - 1s 655ms/step - loss: 36.8110 - val_loss: 46.3900
Epoch 526/1000
2/2 [==============================] - 1s 692ms/step - loss: 43.6871 - val_loss: 45.8233
Epoch 527/1000
2/2 [==============================] - 1s 673ms/step - loss: 38.2474 - val_loss: 52.4575
Epoch 528/1000
2/2 [==============================] - 1s 691ms/step - loss: 38.8468 - val_loss: 46.0994
Epoch 529/1000
2/2 [==============================] - 1s 663ms/step - loss: 40.0486 - val_loss: 46.7390
Epoch 530/1000
2/2 [==============================] - 1s 692ms/step - loss: 34.5802 - val_loss: 51.2739
Epoch 531/1000
2/2 [==============================] - 1s 666ms/step - loss: 43.9822 - val_loss: 46.8019
Epoch 532/1000
2/2 [==============================] - 1s 682ms/step - loss: 35.5644 - val_loss: 51.0220
Epoch 533/1000
2/2 [==============================] - 1s 681ms/step - loss: 34.2730 - val_loss: 47.4280
Epoch 534/1000
2/2 [==============================] - 1s 669ms/step - loss: 37.5558 - val_loss: 45.8784
Epoch 535/1000
2/2 [==============================] - 1s 690ms/step - loss: 39.1082 - val_loss: 40.2572
Epoch 536/1000
2/2 [==============================] - 1s 693ms/step - loss: 37.8281 - val_loss: 43.0711
Epoch 537/1000
2/2 [==============================] - 1s 706ms/step - loss: 33.5548 - val_loss: 45.2469
Epoch 538/1000
2/2 [==============================] - 1s 669ms/step - loss: 40.6005 - val_loss: 45.9331
Epoch 539/1000
2/2 [==============================] - 1s 679ms/step - loss: 39.8344 - val_loss: 46.9692
Epoch 540/1000
2/2 [==============================] - 1s 689ms/step - loss: 40.3130 - val_loss: 45.5495
Epoch 541/1000
2/2 [==============================] - 1s 695ms/step - loss: 35.6860 - val_loss: 46.1704
Epoch 542/1000
2/2 [==============================] - 1s 701ms/step - loss: 40.3441 - val_loss: 45.9533
Epoch 543/1000
2/2 [==============================] - 1s 680ms/step - loss: 42.0399 - val_loss: 46.7131
Epoch 544/1000
2/2 [==============================] - 1s 696ms/step - loss: 34.7655 - val_loss: 48.3146
Epoch 545/1000
2/2 [==============================] - 1s 657ms/step - loss: 37.7199 - val_loss: 46.8845
Epoch 546/1000
2/2 [==============================] - 1s 669ms/step - loss: 37.8103 - val_loss: 48.7448
Epoch 547/1000
2/2 [==============================] - 1s 695ms/step - loss: 36.0161 - val_loss: 47.0784
Epoch 548/1000
2/2 [==============================] - 1s 708ms/step - loss: 37.2980 - val_loss: 43.8409
Epoch 549/1000
2/2 [==============================] - 1s 692ms/step - loss: 37.5406 - val_loss: 46.8668
Epoch 550/1000
2/2 [==============================] - 1s 712ms/step - loss: 36.4812 - val_loss: 45.6274
Epoch 551/1000
2/2 [==============================] - 1s 666ms/step - loss: 37.8119 - val_loss: 50.3204
Epoch 552/1000
2/2 [==============================] - 1s 692ms/step - loss: 41.6157 - val_loss: 54.8975
Epoch 553/1000
2/2 [==============================] - 1s 671ms/step - loss: 41.5941 - val_loss: 46.2809
Epoch 554/1000
2/2 [==============================] - 1s 680ms/step - loss: 35.0598 - val_loss: 49.6529
Epoch 555/1000
2/2 [==============================] - 1s 705ms/step - loss: 35.8253 - val_loss: 49.7709
Epoch 556/1000
2/2 [==============================] - 1s 673ms/step - loss: 39.8184 - val_loss: 52.2072
Epoch 557/1000
2/2 [==============================] - 1s 668ms/step - loss: 33.9134 - val_loss: 45.0838
Epoch 558/1000
2/2 [==============================] - 1s 696ms/step - loss: 37.9047 - val_loss: 46.2854
Epoch 559/1000
2/2 [==============================] - 1s 700ms/step - loss: 37.0513 - val_loss: 53.3577
Epoch 560/1000
2/2 [==============================] - 1s 683ms/step - loss: 32.2747 - val_loss: 54.3857
Epoch 561/1000
2/2 [==============================] - 1s 676ms/step - loss: 37.1896 - val_loss: 51.0243
Epoch 562/1000
2/2 [==============================] - 1s 698ms/step - loss: 37.5837 - val_loss: 54.6323
Epoch 563/1000
2/2 [==============================] - 1s 676ms/step - loss: 37.1565 - val_loss: 43.7987
Epoch 564/1000
2/2 [==============================] - 1s 686ms/step - loss: 32.7695 - val_loss: 51.9940
Epoch 565/1000
2/2 [==============================] - 1s 676ms/step - loss: 34.1376 - val_loss: 42.0889
Epoch 566/1000
2/2 [==============================] - 1s 665ms/step - loss: 38.8335 - val_loss: 40.9004
Epoch 567/1000
2/2 [==============================] - 1s 681ms/step - loss: 36.4484 - val_loss: 45.3237
Epoch 568/1000
2/2 [==============================] - 1s 651ms/step - loss: 39.7150 - val_loss: 48.8906
Epoch 569/1000
2/2 [==============================] - 1s 690ms/step - loss: 34.7086 - val_loss: 47.9237
Epoch 570/1000
2/2 [==============================] - 1s 685ms/step - loss: 35.6202 - val_loss: 46.6155
Epoch 571/1000
2/2 [==============================] - 1s 689ms/step - loss: 38.6348 - val_loss: 45.3803
Epoch 572/1000
2/2 [==============================] - 1s 678ms/step - loss: 38.2410 - val_loss: 55.3526
Epoch 573/1000
2/2 [==============================] - 1s 671ms/step - loss: 35.8911 - val_loss: 52.6790
Epoch 574/1000
2/2 [==============================] - 1s 695ms/step - loss: 35.2054 - val_loss: 50.1599
Epoch 575/1000
2/2 [==============================] - 1s 685ms/step - loss: 34.5771 - val_loss: 48.0486
Epoch 576/1000
2/2 [==============================] - 1s 687ms/step - loss: 35.3723 - val_loss: 45.9123
Epoch 577/1000
2/2 [==============================] - 1s 693ms/step - loss: 37.5408 - val_loss: 52.1078
Epoch 578/1000
2/2 [==============================] - 1s 720ms/step - loss: 33.9236 - val_loss: 47.1653
Epoch 579/1000
2/2 [==============================] - 1s 713ms/step - loss: 33.1154 - val_loss: 50.7849
Epoch 580/1000
2/2 [==============================] - 1s 690ms/step - loss: 37.1781 - val_loss: 56.3022
Epoch 581/1000
2/2 [==============================] - 1s 695ms/step - loss: 34.9853 - val_loss: 45.2264
Epoch 582/1000
2/2 [==============================] - 1s 673ms/step - loss: 36.7230 - val_loss: 53.3255
Epoch 583/1000
2/2 [==============================] - 1s 665ms/step - loss: 36.7673 - val_loss: 52.2955
Epoch 584/1000
2/2 [==============================] - 1s 711ms/step - loss: 33.9963 - val_loss: 51.2539
Epoch 585/1000
2/2 [==============================] - 1s 698ms/step - loss: 35.2726 - val_loss: 49.5592
Epoch 586/1000
2/2 [==============================] - 1s 707ms/step - loss: 38.6609 - val_loss: 45.6424
Epoch 587/1000
2/2 [==============================] - 1s 666ms/step - loss: 35.5455 - val_loss: 51.1136
Epoch 588/1000
2/2 [==============================] - 1s 684ms/step - loss: 35.3728 - val_loss: 47.2203
Epoch 589/1000
2/2 [==============================] - 1s 685ms/step - loss: 33.2928 - val_loss: 48.3399
Epoch 590/1000
2/2 [==============================] - 1s 688ms/step - loss: 36.8566 - val_loss: 43.1677
Epoch 591/1000
2/2 [==============================] - 1s 646ms/step - loss: 37.5659 - val_loss: 45.4194
Epoch 592/1000
2/2 [==============================] - 1s 694ms/step - loss: 36.0483 - val_loss: 40.3804
Epoch 593/1000
2/2 [==============================] - 1s 687ms/step - loss: 38.2540 - val_loss: 42.2066
Epoch 594/1000
2/2 [==============================] - 1s 697ms/step - loss: 33.6577 - val_loss: 46.6832
Epoch 595/1000
2/2 [==============================] - 1s 702ms/step - loss: 35.4459 - val_loss: 47.6134
Epoch 596/1000
2/2 [==============================] - 1s 689ms/step - loss: 35.3875 - val_loss: 48.2164
Epoch 597/1000
2/2 [==============================] - 1s 668ms/step - loss: 34.7686 - val_loss: 48.8942
Epoch 598/1000
2/2 [==============================] - 1s 680ms/step - loss: 31.3067 - val_loss: 55.8987
Epoch 599/1000
2/2 [==============================] - 1s 720ms/step - loss: 40.0416 - val_loss: 46.4193
Epoch 600/1000
2/2 [==============================] - 1s 698ms/step - loss: 31.7959 - val_loss: 42.4261
Epoch 601/1000
2/2 [==============================] - 1s 650ms/step - loss: 36.6444 - val_loss: 45.4792
Epoch 602/1000
2/2 [==============================] - 1s 722ms/step - loss: 32.8213 - val_loss: 46.4275
Epoch 603/1000
2/2 [==============================] - 1s 714ms/step - loss: 33.2713 - val_loss: 38.4951
Epoch 604/1000
2/2 [==============================] - 1s 691ms/step - loss: 31.3246 - val_loss: 48.6067
Epoch 605/1000
2/2 [==============================] - 1s 690ms/step - loss: 34.2481 - val_loss: 47.5895
Epoch 606/1000
2/2 [==============================] - 1s 684ms/step - loss: 33.7054 - val_loss: 49.1366
Epoch 607/1000
2/2 [==============================] - 1s 722ms/step - loss: 34.4718 - val_loss: 46.6946
Epoch 608/1000
2/2 [==============================] - 1s 703ms/step - loss: 33.5503 - val_loss: 52.8735
Epoch 609/1000
2/2 [==============================] - 1s 684ms/step - loss: 33.6312 - val_loss: 53.5881
Epoch 610/1000
2/2 [==============================] - 1s 684ms/step - loss: 35.1347 - val_loss: 43.7584
Epoch 611/1000
2/2 [==============================] - 1s 683ms/step - loss: 32.3203 - val_loss: 47.4924
Epoch 612/1000
2/2 [==============================] - 1s 695ms/step - loss: 34.0328 - val_loss: 43.0905
Epoch 613/1000
2/2 [==============================] - 1s 685ms/step - loss: 32.0430 - val_loss: 44.2894
Epoch 614/1000
2/2 [==============================] - 1s 695ms/step - loss: 32.5368 - val_loss: 50.8637
Epoch 615/1000
2/2 [==============================] - 1s 678ms/step - loss: 36.7926 - val_loss: 42.3458
Epoch 616/1000
2/2 [==============================] - 1s 695ms/step - loss: 32.5463 - val_loss: 39.1585
Epoch 617/1000
2/2 [==============================] - 1s 691ms/step - loss: 36.5281 - val_loss: 40.6674
Epoch 618/1000
2/2 [==============================] - 1s 692ms/step - loss: 35.3740 - val_loss: 41.7192
Epoch 619/1000
2/2 [==============================] - 1s 696ms/step - loss: 34.1579 - val_loss: 48.0161
Epoch 620/1000
2/2 [==============================] - 1s 666ms/step - loss: 38.7538 - val_loss: 41.9786
Epoch 621/1000
2/2 [==============================] - 1s 700ms/step - loss: 36.0298 - val_loss: 43.5290
Epoch 622/1000
2/2 [==============================] - 1s 682ms/step - loss: 36.2059 - val_loss: 42.6956
Epoch 623/1000
2/2 [==============================] - 1s 719ms/step - loss: 33.8769 - val_loss: 42.4440
Epoch 624/1000
2/2 [==============================] - 1s 711ms/step - loss: 35.2602 - val_loss: 47.4006
Epoch 625/1000
2/2 [==============================] - 1s 700ms/step - loss: 34.3096 - val_loss: 45.9258
Epoch 626/1000
2/2 [==============================] - 1s 680ms/step - loss: 32.0355 - val_loss: 37.8181
Epoch 627/1000
2/2 [==============================] - 1s 691ms/step - loss: 30.6872 - val_loss: 40.9036
Epoch 628/1000
2/2 [==============================] - 1s 687ms/step - loss: 34.4675 - val_loss: 45.0151
Epoch 629/1000
2/2 [==============================] - 1s 700ms/step - loss: 33.6664 - val_loss: 46.5704
Epoch 630/1000
2/2 [==============================] - 1s 682ms/step - loss: 35.3561 - val_loss: 45.5545
Epoch 631/1000
2/2 [==============================] - 1s 684ms/step - loss: 30.3408 - val_loss: 40.8984
Epoch 632/1000
2/2 [==============================] - 1s 681ms/step - loss: 31.7281 - val_loss: 43.1092
Epoch 633/1000
2/2 [==============================] - 1s 697ms/step - loss: 40.3131 - val_loss: 43.0506
Epoch 634/1000
2/2 [==============================] - 1s 719ms/step - loss: 30.7412 - val_loss: 43.4934
Epoch 635/1000
2/2 [==============================] - 1s 705ms/step - loss: 35.9726 - val_loss: 50.2398
Epoch 636/1000
2/2 [==============================] - 1s 705ms/step - loss: 31.3832 - val_loss: 48.1825
Epoch 637/1000
2/2 [==============================] - 1s 687ms/step - loss: 36.1252 - val_loss: 45.9165
Epoch 638/1000
2/2 [==============================] - 1s 693ms/step - loss: 35.2608 - val_loss: 46.8245
Epoch 639/1000
2/2 [==============================] - 1s 695ms/step - loss: 32.7199 - val_loss: 52.1948
Epoch 640/1000
2/2 [==============================] - 1s 707ms/step - loss: 32.1611 - val_loss: 47.8398
Epoch 641/1000
2/2 [==============================] - 1s 711ms/step - loss: 29.8492 - val_loss: 44.7605
Epoch 642/1000
2/2 [==============================] - 1s 696ms/step - loss: 34.1642 - val_loss: 45.4986
Epoch 643/1000
2/2 [==============================] - 1s 705ms/step - loss: 33.8683 - val_loss: 49.6359
Epoch 644/1000
2/2 [==============================] - 1s 685ms/step - loss: 34.8632 - val_loss: 44.1732
Epoch 645/1000
2/2 [==============================] - 1s 712ms/step - loss: 30.8442 - val_loss: 40.0918
Epoch 646/1000
2/2 [==============================] - 1s 722ms/step - loss: 38.0228 - val_loss: 43.2014
Epoch 647/1000
2/2 [==============================] - 1s 680ms/step - loss: 33.2534 - val_loss: 45.9154
Epoch 648/1000
2/2 [==============================] - 1s 706ms/step - loss: 31.7534 - val_loss: 46.7879
Epoch 649/1000
2/2 [==============================] - 1s 694ms/step - loss: 31.2246 - val_loss: 39.2816
Epoch 650/1000
2/2 [==============================] - 1s 686ms/step - loss: 30.9284 - val_loss: 43.6734
Epoch 651/1000
2/2 [==============================] - 1s 675ms/step - loss: 32.8071 - val_loss: 46.0496
Epoch 652/1000
2/2 [==============================] - 1s 662ms/step - loss: 34.6655 - val_loss: 52.0561
Epoch 653/1000
2/2 [==============================] - 1s 663ms/step - loss: 34.2062 - val_loss: 50.8351
Epoch 654/1000
2/2 [==============================] - 1s 681ms/step - loss: 29.7228 - val_loss: 42.8612
Epoch 655/1000
2/2 [==============================] - 1s 692ms/step - loss: 30.2601 - val_loss: 49.8683
Epoch 656/1000
2/2 [==============================] - 1s 682ms/step - loss: 31.0844 - val_loss: 42.8555
Epoch 657/1000
2/2 [==============================] - 1s 687ms/step - loss: 33.0810 - val_loss: 44.9957
Epoch 658/1000
2/2 [==============================] - 1s 689ms/step - loss: 36.2986 - val_loss: 43.1183
Epoch 659/1000
2/2 [==============================] - 1s 690ms/step - loss: 34.2161 - val_loss: 40.8816
Epoch 660/1000
2/2 [==============================] - 1s 665ms/step - loss: 36.1142 - val_loss: 46.1456
Epoch 661/1000
2/2 [==============================] - 1s 690ms/step - loss: 33.2869 - val_loss: 46.9438
Epoch 662/1000
2/2 [==============================] - 1s 651ms/step - loss: 31.8876 - val_loss: 40.3115
Epoch 663/1000
2/2 [==============================] - 1s 692ms/step - loss: 31.2387 - val_loss: 43.2756
Epoch 664/1000
2/2 [==============================] - 1s 674ms/step - loss: 34.6307 - val_loss: 47.0692
Epoch 665/1000
2/2 [==============================] - 1s 666ms/step - loss: 31.3754 - val_loss: 44.0641
Epoch 666/1000
2/2 [==============================] - 1s 700ms/step - loss: 31.9631 - val_loss: 49.7731
Epoch 667/1000
2/2 [==============================] - 1s 703ms/step - loss: 31.0102 - val_loss: 47.3629
Epoch 668/1000
2/2 [==============================] - 1s 699ms/step - loss: 28.5740 - val_loss: 48.3040
Epoch 669/1000
2/2 [==============================] - 1s 693ms/step - loss: 31.6062 - val_loss: 47.7728
Epoch 670/1000
2/2 [==============================] - 1s 704ms/step - loss: 36.0070 - val_loss: 44.2961
Epoch 671/1000
2/2 [==============================] - 1s 700ms/step - loss: 33.8138 - val_loss: 45.1436
Epoch 672/1000
2/2 [==============================] - 1s 687ms/step - loss: 30.0938 - val_loss: 42.1995
Epoch 673/1000
2/2 [==============================] - 1s 714ms/step - loss: 32.3423 - val_loss: 44.3092
Epoch 674/1000
2/2 [==============================] - 1s 681ms/step - loss: 29.7448 - val_loss: 47.8112
Epoch 675/1000
2/2 [==============================] - 1s 671ms/step - loss: 31.9186 - val_loss: 51.9435
Epoch 676/1000
2/2 [==============================] - 1s 686ms/step - loss: 29.2424 - val_loss: 51.1462
Epoch 677/1000
2/2 [==============================] - 1s 668ms/step - loss: 29.6073 - val_loss: 45.9276
Epoch 678/1000
2/2 [==============================] - 1s 694ms/step - loss: 30.5544 - val_loss: 52.2804
Epoch 679/1000
2/2 [==============================] - 1s 677ms/step - loss: 29.6195 - val_loss: 56.7024
Epoch 680/1000
2/2 [==============================] - 1s 684ms/step - loss: 31.7219 - val_loss: 49.2342
Epoch 681/1000
2/2 [==============================] - 1s 685ms/step - loss: 33.5669 - val_loss: 53.2818
Epoch 682/1000
2/2 [==============================] - 1s 692ms/step - loss: 33.4813 - val_loss: 44.3998
Epoch 683/1000
2/2 [==============================] - 1s 684ms/step - loss: 32.4058 - val_loss: 45.5278
Epoch 684/1000
2/2 [==============================] - 1s 692ms/step - loss: 35.9742 - val_loss: 48.8750
Epoch 685/1000
2/2 [==============================] - 1s 688ms/step - loss: 33.0473 - val_loss: 42.8505
Epoch 686/1000
2/2 [==============================] - 1s 707ms/step - loss: 32.4484 - val_loss: 43.4810
Epoch 687/1000
2/2 [==============================] - 1s 692ms/step - loss: 31.6658 - val_loss: 42.0430
Epoch 688/1000
2/2 [==============================] - 1s 666ms/step - loss: 32.6279 - val_loss: 49.4810
Epoch 689/1000
2/2 [==============================] - 1s 688ms/step - loss: 36.0326 - val_loss: 44.3564
Epoch 690/1000
2/2 [==============================] - 1s 734ms/step - loss: 32.0976 - val_loss: 46.2913
Epoch 691/1000
2/2 [==============================] - 1s 718ms/step - loss: 35.7098 - val_loss: 42.7401
Epoch 692/1000
2/2 [==============================] - 1s 655ms/step - loss: 36.8835 - val_loss: 42.0969
Epoch 693/1000
2/2 [==============================] - 1s 717ms/step - loss: 33.1746 - val_loss: 42.8010
Epoch 694/1000
2/2 [==============================] - 1s 693ms/step - loss: 30.7674 - val_loss: 46.5780
Epoch 695/1000
2/2 [==============================] - 1s 698ms/step - loss: 29.8354 - val_loss: 41.6936
Epoch 696/1000
2/2 [==============================] - 1s 681ms/step - loss: 30.6600 - val_loss: 40.1046
Epoch 697/1000
2/2 [==============================] - 1s 699ms/step - loss: 28.7097 - val_loss: 44.2366
Epoch 698/1000
2/2 [==============================] - 1s 685ms/step - loss: 28.1175 - val_loss: 48.8665
Epoch 699/1000
2/2 [==============================] - 1s 689ms/step - loss: 30.7875 - val_loss: 45.4667
Epoch 700/1000
2/2 [==============================] - 1s 696ms/step - loss: 29.4950 - val_loss: 46.3765
Epoch 701/1000
2/2 [==============================] - 1s 671ms/step - loss: 27.8763 - val_loss: 45.1216
Epoch 702/1000
2/2 [==============================] - 1s 683ms/step - loss: 32.4320 - val_loss: 42.9609
Epoch 703/1000
2/2 [==============================] - 1s 690ms/step - loss: 28.1909 - val_loss: 40.1497
Epoch 704/1000
2/2 [==============================] - 1s 683ms/step - loss: 30.0920 - val_loss: 43.1573
Epoch 705/1000
2/2 [==============================] - 1s 671ms/step - loss: 31.2919 - val_loss: 40.2759
Epoch 706/1000
2/2 [==============================] - 1s 717ms/step - loss: 29.9737 - val_loss: 40.4786
Epoch 707/1000
2/2 [==============================] - 1s 699ms/step - loss: 29.1717 - val_loss: 45.8648
Epoch 708/1000
2/2 [==============================] - 1s 670ms/step - loss: 26.6337 - val_loss: 36.9907
Epoch 709/1000
2/2 [==============================] - 1s 705ms/step - loss: 28.3121 - val_loss: 44.4344
Epoch 710/1000
2/2 [==============================] - 1s 696ms/step - loss: 31.0011 - val_loss: 38.3463
Epoch 711/1000
2/2 [==============================] - 1s 666ms/step - loss: 31.8956 - val_loss: 44.3572
Epoch 712/1000
2/2 [==============================] - 1s 698ms/step - loss: 27.6157 - val_loss: 44.9659
Epoch 713/1000
2/2 [==============================] - 1s 696ms/step - loss: 29.0599 - val_loss: 43.8186
Epoch 714/1000
2/2 [==============================] - 1s 701ms/step - loss: 27.0294 - val_loss: 42.5014
Epoch 715/1000
2/2 [==============================] - 1s 705ms/step - loss: 31.1712 - val_loss: 45.4947
Epoch 716/1000
2/2 [==============================] - 1s 693ms/step - loss: 32.2314 - val_loss: 46.5021
Epoch 717/1000
2/2 [==============================] - 1s 701ms/step - loss: 30.6807 - val_loss: 43.0174
Epoch 718/1000
2/2 [==============================] - 1s 695ms/step - loss: 32.9190 - val_loss: 45.7878
Epoch 719/1000
2/2 [==============================] - 1s 704ms/step - loss: 29.2005 - val_loss: 41.1135
Epoch 720/1000
2/2 [==============================] - 1s 686ms/step - loss: 31.9488 - val_loss: 47.1480
Epoch 721/1000
2/2 [==============================] - 1s 700ms/step - loss: 29.1095 - val_loss: 46.8317
Epoch 722/1000
2/2 [==============================] - 1s 701ms/step - loss: 31.2070 - val_loss: 40.3331
Epoch 723/1000
2/2 [==============================] - 1s 734ms/step - loss: 29.0325 - val_loss: 43.4289
Epoch 724/1000
2/2 [==============================] - 1s 699ms/step - loss: 29.0649 - val_loss: 43.3311
Epoch 725/1000
2/2 [==============================] - 1s 693ms/step - loss: 31.4502 - val_loss: 44.4229
Epoch 726/1000
2/2 [==============================] - 1s 715ms/step - loss: 30.1041 - val_loss: 40.8046
Epoch 727/1000
2/2 [==============================] - 1s 709ms/step - loss: 31.0756 - val_loss: 46.9802
Epoch 728/1000
2/2 [==============================] - 1s 715ms/step - loss: 29.2459 - val_loss: 38.3546
Epoch 729/1000
2/2 [==============================] - 1s 727ms/step - loss: 28.2762 - val_loss: 40.6436
Epoch 730/1000
2/2 [==============================] - 1s 698ms/step - loss: 32.4723 - val_loss: 38.3211
Epoch 731/1000
2/2 [==============================] - 1s 701ms/step - loss: 29.2893 - val_loss: 39.6601
Epoch 732/1000
2/2 [==============================] - 1s 697ms/step - loss: 31.5680 - val_loss: 37.0637
Epoch 733/1000
2/2 [==============================] - 1s 712ms/step - loss: 29.0516 - val_loss: 38.4257
Epoch 734/1000
2/2 [==============================] - 1s 701ms/step - loss: 30.2407 - val_loss: 42.6218
Epoch 735/1000
2/2 [==============================] - 1s 664ms/step - loss: 31.7266 - val_loss: 44.3272
Epoch 736/1000
2/2 [==============================] - 1s 720ms/step - loss: 28.4929 - val_loss: 41.3844
Epoch 737/1000
2/2 [==============================] - 1s 665ms/step - loss: 29.9835 - val_loss: 44.8237
Epoch 738/1000
2/2 [==============================] - 1s 676ms/step - loss: 28.9603 - val_loss: 44.1838
Epoch 739/1000
2/2 [==============================] - 1s 701ms/step - loss: 28.3438 - val_loss: 45.4834
Epoch 740/1000
2/2 [==============================] - 1s 660ms/step - loss: 29.5695 - val_loss: 42.8671
Epoch 741/1000
2/2 [==============================] - 1s 726ms/step - loss: 28.4767 - val_loss: 48.1958
Epoch 742/1000
2/2 [==============================] - 1s 706ms/step - loss: 31.2414 - val_loss: 48.4264
Epoch 743/1000
2/2 [==============================] - 1s 695ms/step - loss: 27.5082 - val_loss: 54.5266
Epoch 744/1000
2/2 [==============================] - 1s 686ms/step - loss: 32.6860 - val_loss: 52.0117
Epoch 745/1000
2/2 [==============================] - 1s 676ms/step - loss: 31.8796 - val_loss: 51.9983
Epoch 746/1000
2/2 [==============================] - 1s 681ms/step - loss: 28.8156 - val_loss: 46.6259
Epoch 747/1000
2/2 [==============================] - 1s 695ms/step - loss: 33.9749 - val_loss: 56.8239
Epoch 748/1000
2/2 [==============================] - 1s 681ms/step - loss: 33.9931 - val_loss: 45.0227
Epoch 749/1000
2/2 [==============================] - 1s 688ms/step - loss: 29.9119 - val_loss: 39.3667
Epoch 750/1000
2/2 [==============================] - 1s 663ms/step - loss: 28.8715 - val_loss: 43.2403
Epoch 751/1000
2/2 [==============================] - 1s 698ms/step - loss: 29.2242 - val_loss: 37.0113
Epoch 752/1000
2/2 [==============================] - 1s 690ms/step - loss: 31.5128 - val_loss: 40.0524
Epoch 753/1000
2/2 [==============================] - 1s 710ms/step - loss: 30.1997 - val_loss: 39.2152
Epoch 754/1000
2/2 [==============================] - 1s 676ms/step - loss: 31.7371 - val_loss: 35.8455
Epoch 755/1000
2/2 [==============================] - 1s 696ms/step - loss: 32.1462 - val_loss: 42.3486
Epoch 756/1000
2/2 [==============================] - 1s 681ms/step - loss: 30.5663 - val_loss: 41.3869
Epoch 757/1000
2/2 [==============================] - 1s 690ms/step - loss: 29.2509 - val_loss: 40.5433
Epoch 758/1000
2/2 [==============================] - 1s 687ms/step - loss: 31.6788 - val_loss: 42.0980
Epoch 759/1000
2/2 [==============================] - 1s 700ms/step - loss: 31.2647 - val_loss: 43.5244
Epoch 760/1000
2/2 [==============================] - 1s 705ms/step - loss: 29.5435 - val_loss: 39.3862
Epoch 761/1000
2/2 [==============================] - 1s 684ms/step - loss: 33.4285 - val_loss: 36.0696
Epoch 762/1000
2/2 [==============================] - 1s 691ms/step - loss: 31.7567 - val_loss: 43.2857
Epoch 763/1000
2/2 [==============================] - 1s 688ms/step - loss: 30.0649 - val_loss: 39.8769
Epoch 764/1000
2/2 [==============================] - 1s 706ms/step - loss: 28.2743 - val_loss: 43.9678
Epoch 765/1000
2/2 [==============================] - 1s 678ms/step - loss: 30.0274 - val_loss: 40.0398
Epoch 766/1000
2/2 [==============================] - 1s 719ms/step - loss: 31.7305 - val_loss: 48.7012
Epoch 767/1000
2/2 [==============================] - 1s 697ms/step - loss: 28.9716 - val_loss: 40.8125
Epoch 768/1000
2/2 [==============================] - 1s 715ms/step - loss: 28.8855 - val_loss: 37.7843
Epoch 769/1000
2/2 [==============================] - 1s 696ms/step - loss: 30.8764 - val_loss: 42.1410
Epoch 770/1000
2/2 [==============================] - 1s 701ms/step - loss: 29.0211 - val_loss: 39.5429
Epoch 771/1000
2/2 [==============================] - 1s 704ms/step - loss: 30.1399 - val_loss: 42.2818
Epoch 772/1000
2/2 [==============================] - 1s 708ms/step - loss: 28.3727 - val_loss: 40.4175
Epoch 773/1000
2/2 [==============================] - 1s 701ms/step - loss: 30.5539 - val_loss: 41.0016
Epoch 774/1000
2/2 [==============================] - 1s 707ms/step - loss: 33.5891 - val_loss: 45.9277
Epoch 775/1000
2/2 [==============================] - 1s 697ms/step - loss: 34.3365 - val_loss: 42.0180
Epoch 776/1000
2/2 [==============================] - 1s 706ms/step - loss: 29.5092 - val_loss: 40.7533
Epoch 777/1000
2/2 [==============================] - 1s 696ms/step - loss: 29.7114 - val_loss: 40.2760
Epoch 778/1000
2/2 [==============================] - 1s 677ms/step - loss: 26.6183 - val_loss: 48.3798
Epoch 779/1000
2/2 [==============================] - 1s 680ms/step - loss: 30.9731 - val_loss: 39.5137
Epoch 780/1000
2/2 [==============================] - 1s 688ms/step - loss: 27.0275 - val_loss: 42.3147
Epoch 781/1000
2/2 [==============================] - 1s 692ms/step - loss: 27.6266 - val_loss: 42.4086
Epoch 782/1000
2/2 [==============================] - 1s 686ms/step - loss: 27.0056 - val_loss: 39.1328
Epoch 783/1000
2/2 [==============================] - 1s 688ms/step - loss: 29.6837 - val_loss: 39.0390
Epoch 784/1000
2/2 [==============================] - 1s 669ms/step - loss: 30.1448 - val_loss: 46.8866
Epoch 785/1000
2/2 [==============================] - 1s 701ms/step - loss: 28.7871 - val_loss: 48.2540
Epoch 786/1000
2/2 [==============================] - 1s 683ms/step - loss: 28.6108 - val_loss: 37.0412
Epoch 787/1000
2/2 [==============================] - 1s 693ms/step - loss: 31.3929 - val_loss: 41.8200
Epoch 788/1000
2/2 [==============================] - 1s 679ms/step - loss: 29.5959 - val_loss: 39.2087
Epoch 789/1000
2/2 [==============================] - 1s 691ms/step - loss: 26.5902 - val_loss: 47.2058
Epoch 790/1000
2/2 [==============================] - 1s 684ms/step - loss: 28.3548 - val_loss: 41.3357
Epoch 791/1000
2/2 [==============================] - 1s 695ms/step - loss: 26.8667 - val_loss: 40.2221
Epoch 792/1000
2/2 [==============================] - 1s 685ms/step - loss: 34.3268 - val_loss: 36.5073
Epoch 793/1000
2/2 [==============================] - 1s 678ms/step - loss: 28.3714 - val_loss: 47.4812
Epoch 794/1000
2/2 [==============================] - 1s 698ms/step - loss: 25.3407 - val_loss: 42.6943
Epoch 795/1000
2/2 [==============================] - 1s 694ms/step - loss: 29.9932 - val_loss: 41.9824
Epoch 796/1000
2/2 [==============================] - 1s 686ms/step - loss: 28.4018 - val_loss: 39.8411
Epoch 797/1000
2/2 [==============================] - 1s 683ms/step - loss: 30.8703 - val_loss: 39.0850
Epoch 798/1000
2/2 [==============================] - 1s 676ms/step - loss: 28.8596 - val_loss: 49.7752
Epoch 799/1000
2/2 [==============================] - 1s 687ms/step - loss: 30.4799 - val_loss: 45.0435
Epoch 800/1000
2/2 [==============================] - 1s 692ms/step - loss: 28.6880 - val_loss: 46.7910
Epoch 801/1000
2/2 [==============================] - 1s 680ms/step - loss: 25.0962 - val_loss: 45.8799
Epoch 802/1000
2/2 [==============================] - 1s 690ms/step - loss: 25.6701 - val_loss: 49.1145
Epoch 803/1000
2/2 [==============================] - 1s 655ms/step - loss: 31.0509 - val_loss: 42.4195
Epoch 804/1000
2/2 [==============================] - 1s 701ms/step - loss: 32.6225 - val_loss: 42.6592
Epoch 805/1000
2/2 [==============================] - 1s 700ms/step - loss: 27.3884 - val_loss: 40.4563
Epoch 806/1000
2/2 [==============================] - 1s 686ms/step - loss: 36.5876 - val_loss: 47.1634
Epoch 807/1000
2/2 [==============================] - 1s 706ms/step - loss: 26.8056 - val_loss: 39.7587
Epoch 808/1000
2/2 [==============================] - 1s 698ms/step - loss: 27.1343 - val_loss: 45.5709
Epoch 809/1000
2/2 [==============================] - 1s 678ms/step - loss: 31.9769 - val_loss: 48.0324
Epoch 810/1000
2/2 [==============================] - 1s 698ms/step - loss: 34.1627 - val_loss: 51.0324
Epoch 811/1000
2/2 [==============================] - 1s 673ms/step - loss: 29.7767 - val_loss: 46.0994
Epoch 812/1000
2/2 [==============================] - 1s 695ms/step - loss: 27.4867 - val_loss: 55.0387
Epoch 813/1000
2/2 [==============================] - 1s 688ms/step - loss: 29.0594 - val_loss: 50.4182
Epoch 814/1000
2/2 [==============================] - 1s 696ms/step - loss: 29.2442 - val_loss: 53.4359
Epoch 815/1000
2/2 [==============================] - 1s 680ms/step - loss: 32.4910 - val_loss: 58.5694
Epoch 816/1000
2/2 [==============================] - 1s 692ms/step - loss: 29.9500 - val_loss: 45.6441
Epoch 817/1000
2/2 [==============================] - 1s 697ms/step - loss: 28.7924 - val_loss: 53.6616
Epoch 818/1000
2/2 [==============================] - 1s 667ms/step - loss: 26.1860 - val_loss: 56.4352
Epoch 819/1000
2/2 [==============================] - 1s 702ms/step - loss: 28.1401 - val_loss: 47.8154
Epoch 820/1000
2/2 [==============================] - 1s 670ms/step - loss: 28.9219 - val_loss: 43.0942
Epoch 821/1000
2/2 [==============================] - 1s 706ms/step - loss: 29.3566 - val_loss: 42.1640
Epoch 822/1000
2/2 [==============================] - 1s 683ms/step - loss: 27.0476 - val_loss: 39.0209
Epoch 823/1000
2/2 [==============================] - 1s 695ms/step - loss: 26.7523 - val_loss: 36.7071
Epoch 824/1000
2/2 [==============================] - 1s 678ms/step - loss: 31.0817 - val_loss: 37.4235
Epoch 825/1000
2/2 [==============================] - 1s 692ms/step - loss: 27.8418 - val_loss: 37.2163
Epoch 826/1000
2/2 [==============================] - 1s 679ms/step - loss: 26.7588 - val_loss: 43.2407
Epoch 827/1000
2/2 [==============================] - 1s 707ms/step - loss: 27.6319 - val_loss: 47.4357
Epoch 828/1000
2/2 [==============================] - 1s 693ms/step - loss: 29.8004 - val_loss: 46.2374
Epoch 829/1000
2/2 [==============================] - 1s 675ms/step - loss: 29.3901 - val_loss: 42.9368
Epoch 830/1000
2/2 [==============================] - 1s 684ms/step - loss: 27.1519 - val_loss: 45.8122
Epoch 831/1000
2/2 [==============================] - 1s 690ms/step - loss: 29.6572 - val_loss: 45.2152
Epoch 832/1000
2/2 [==============================] - 1s 655ms/step - loss: 30.6505 - val_loss: 34.0605
Epoch 833/1000
2/2 [==============================] - 1s 669ms/step - loss: 26.9858 - val_loss: 38.0602
Epoch 834/1000
2/2 [==============================] - 1s 682ms/step - loss: 27.6287 - val_loss: 43.5597
Epoch 835/1000
2/2 [==============================] - 1s 689ms/step - loss: 28.0732 - val_loss: 35.4373
Epoch 836/1000
2/2 [==============================] - 1s 674ms/step - loss: 30.9231 - val_loss: 41.2961
Epoch 837/1000
2/2 [==============================] - 1s 719ms/step - loss: 30.9979 - val_loss: 44.2574
Epoch 838/1000
2/2 [==============================] - 1s 664ms/step - loss: 26.4859 - val_loss: 45.3612
Epoch 839/1000
2/2 [==============================] - 1s 685ms/step - loss: 28.4636 - val_loss: 45.4517
Epoch 840/1000
2/2 [==============================] - 1s 712ms/step - loss: 27.5798 - val_loss: 43.4013
Epoch 841/1000
2/2 [==============================] - 1s 676ms/step - loss: 27.4877 - val_loss: 46.6920
Epoch 842/1000
2/2 [==============================] - 1s 691ms/step - loss: 31.0397 - val_loss: 43.7919
Epoch 843/1000
2/2 [==============================] - 1s 687ms/step - loss: 30.4286 - val_loss: 43.2969
Epoch 844/1000
2/2 [==============================] - 1s 649ms/step - loss: 34.1212 - val_loss: 42.6919
Epoch 845/1000
2/2 [==============================] - 1s 671ms/step - loss: 29.6378 - val_loss: 42.8588
Epoch 846/1000
2/2 [==============================] - 1s 699ms/step - loss: 30.3503 - val_loss: 45.5784
Epoch 847/1000
2/2 [==============================] - 1s 695ms/step - loss: 28.4277 - val_loss: 42.6635
Epoch 848/1000
2/2 [==============================] - 1s 677ms/step - loss: 28.3798 - val_loss: 40.1475
Epoch 849/1000
2/2 [==============================] - 1s 707ms/step - loss: 29.1490 - val_loss: 44.0912
Epoch 850/1000
2/2 [==============================] - 1s 691ms/step - loss: 32.1440 - val_loss: 39.7091
Epoch 851/1000
2/2 [==============================] - 1s 665ms/step - loss: 30.4432 - val_loss: 40.4515
Epoch 852/1000
2/2 [==============================] - 1s 700ms/step - loss: 29.6387 - val_loss: 38.2209
Epoch 853/1000
2/2 [==============================] - 1s 677ms/step - loss: 29.8221 - val_loss: 44.0481
Epoch 854/1000
2/2 [==============================] - 1s 658ms/step - loss: 30.6730 - val_loss: 44.6473
Epoch 855/1000
2/2 [==============================] - 1s 662ms/step - loss: 27.9772 - val_loss: 42.4222
Epoch 856/1000
2/2 [==============================] - 1s 670ms/step - loss: 28.1711 - val_loss: 35.7361
Epoch 857/1000
2/2 [==============================] - 1s 692ms/step - loss: 26.6970 - val_loss: 38.1355
Epoch 858/1000
2/2 [==============================] - 1s 680ms/step - loss: 29.6530 - val_loss: 38.5222
Epoch 859/1000
2/2 [==============================] - 1s 691ms/step - loss: 28.1254 - val_loss: 41.7841
Epoch 860/1000
2/2 [==============================] - 1s 695ms/step - loss: 25.9286 - val_loss: 47.2036
Epoch 861/1000
2/2 [==============================] - 1s 683ms/step - loss: 25.6029 - val_loss: 50.8103
Epoch 862/1000
2/2 [==============================] - 1s 687ms/step - loss: 26.3886 - val_loss: 48.0723
Epoch 863/1000
2/2 [==============================] - 1s 684ms/step - loss: 30.8915 - val_loss: 41.9841
Epoch 864/1000
2/2 [==============================] - 1s 675ms/step - loss: 28.7440 - val_loss: 47.0927
Epoch 865/1000
2/2 [==============================] - 1s 718ms/step - loss: 27.7029 - val_loss: 49.7381
Epoch 866/1000
2/2 [==============================] - 1s 687ms/step - loss: 28.3206 - val_loss: 45.0214
Epoch 867/1000
2/2 [==============================] - 1s 713ms/step - loss: 28.2700 - val_loss: 49.5574
Epoch 868/1000
2/2 [==============================] - 1s 710ms/step - loss: 27.0644 - val_loss: 56.0733
Epoch 869/1000
2/2 [==============================] - 1s 691ms/step - loss: 26.0810 - val_loss: 46.3118
Epoch 870/1000
2/2 [==============================] - 1s 720ms/step - loss: 26.6829 - val_loss: 46.1308
Epoch 871/1000
2/2 [==============================] - 1s 708ms/step - loss: 30.7707 - val_loss: 44.1327
Epoch 872/1000
2/2 [==============================] - 1s 720ms/step - loss: 28.6211 - val_loss: 47.7186
Epoch 873/1000
2/2 [==============================] - 1s 712ms/step - loss: 29.9035 - val_loss: 50.0863
Epoch 874/1000
2/2 [==============================] - 1s 673ms/step - loss: 28.8753 - val_loss: 44.4601
Epoch 875/1000
2/2 [==============================] - 1s 665ms/step - loss: 29.9943 - val_loss: 43.1384
Epoch 876/1000
2/2 [==============================] - 1s 679ms/step - loss: 29.3173 - val_loss: 44.1569
Epoch 877/1000
2/2 [==============================] - 1s 672ms/step - loss: 28.0956 - val_loss: 41.7220
Epoch 878/1000
2/2 [==============================] - 1s 689ms/step - loss: 25.8687 - val_loss: 49.0650
Epoch 879/1000
2/2 [==============================] - 1s 697ms/step - loss: 25.8007 - val_loss: 46.4414
Epoch 880/1000
2/2 [==============================] - 1s 681ms/step - loss: 29.1241 - val_loss: 46.5951
Epoch 881/1000
2/2 [==============================] - 1s 659ms/step - loss: 29.9233 - val_loss: 43.0760
Epoch 882/1000
2/2 [==============================] - 1s 687ms/step - loss: 28.8068 - val_loss: 38.7811
Epoch 883/1000
2/2 [==============================] - 1s 684ms/step - loss: 29.1017 - val_loss: 38.0423
Epoch 884/1000
2/2 [==============================] - 1s 683ms/step - loss: 26.7774 - val_loss: 39.2857
Epoch 885/1000
2/2 [==============================] - 1s 677ms/step - loss: 26.9854 - val_loss: 39.3785
Epoch 886/1000
2/2 [==============================] - 1s 672ms/step - loss: 28.9176 - val_loss: 35.5048
Epoch 887/1000
2/2 [==============================] - 1s 693ms/step - loss: 26.8241 - val_loss: 38.7749
Epoch 888/1000
2/2 [==============================] - 1s 699ms/step - loss: 27.1526 - val_loss: 42.0893
Epoch 889/1000
2/2 [==============================] - 1s 678ms/step - loss: 29.4670 - val_loss: 42.2539
Epoch 890/1000
2/2 [==============================] - 1s 682ms/step - loss: 32.4938 - val_loss: 42.3958
Epoch 891/1000
2/2 [==============================] - 1s 689ms/step - loss: 27.8139 - val_loss: 37.0356
Epoch 892/1000
2/2 [==============================] - 1s 650ms/step - loss: 29.1820 - val_loss: 43.8752
Epoch 893/1000
2/2 [==============================] - 1s 682ms/step - loss: 26.0616 - val_loss: 48.3419
Epoch 894/1000
2/2 [==============================] - 1s 679ms/step - loss: 27.8328 - val_loss: 45.4732
Epoch 895/1000
2/2 [==============================] - 1s 695ms/step - loss: 28.7152 - val_loss: 41.5383
Epoch 896/1000
2/2 [==============================] - 1s 674ms/step - loss: 29.7643 - val_loss: 40.0750
Epoch 897/1000
2/2 [==============================] - 1s 675ms/step - loss: 28.5398 - val_loss: 42.3744
Epoch 898/1000
2/2 [==============================] - 1s 693ms/step - loss: 27.9339 - val_loss: 33.9499
Epoch 899/1000
2/2 [==============================] - 1s 682ms/step - loss: 27.6951 - val_loss: 41.0452
Epoch 900/1000
2/2 [==============================] - 1s 666ms/step - loss: 29.3537 - val_loss: 41.4077
Epoch 901/1000
2/2 [==============================] - 1s 655ms/step - loss: 31.5854 - val_loss: 44.6186
Epoch 902/1000
2/2 [==============================] - 1s 686ms/step - loss: 30.0967 - val_loss: 43.6805
Epoch 903/1000
2/2 [==============================] - 1s 673ms/step - loss: 27.3583 - val_loss: 48.5465
Epoch 904/1000
2/2 [==============================] - 1s 702ms/step - loss: 27.7645 - val_loss: 41.7374
Epoch 905/1000
2/2 [==============================] - 1s 673ms/step - loss: 26.4171 - val_loss: 40.4036
Epoch 906/1000
2/2 [==============================] - 1s 644ms/step - loss: 29.0391 - val_loss: 40.8176
Epoch 907/1000
2/2 [==============================] - 1s 709ms/step - loss: 28.0415 - val_loss: 43.8244
Epoch 908/1000
2/2 [==============================] - 1s 689ms/step - loss: 30.5693 - val_loss: 43.8526
Epoch 909/1000
2/2 [==============================] - 1s 676ms/step - loss: 25.9677 - val_loss: 40.3344
Epoch 910/1000
2/2 [==============================] - 1s 669ms/step - loss: 28.4437 - val_loss: 38.6447
Epoch 911/1000
2/2 [==============================] - 1s 685ms/step - loss: 25.8081 - val_loss: 41.4012
Epoch 912/1000
2/2 [==============================] - 1s 659ms/step - loss: 27.4798 - val_loss: 40.1864
Epoch 913/1000
2/2 [==============================] - 1s 672ms/step - loss: 27.5017 - val_loss: 44.9905
Epoch 914/1000
2/2 [==============================] - 1s 685ms/step - loss: 30.0692 - val_loss: 39.7244
Epoch 915/1000
2/2 [==============================] - 1s 701ms/step - loss: 27.6583 - val_loss: 44.0580
Epoch 916/1000
2/2 [==============================] - 1s 669ms/step - loss: 26.2522 - val_loss: 41.5874
Epoch 917/1000
2/2 [==============================] - 1s 678ms/step - loss: 30.3738 - val_loss: 48.0301
Epoch 918/1000
2/2 [==============================] - 1s 689ms/step - loss: 31.4550 - val_loss: 43.0540
Epoch 919/1000
2/2 [==============================] - 1s 689ms/step - loss: 28.6554 - val_loss: 42.5740
Epoch 920/1000
2/2 [==============================] - 1s 691ms/step - loss: 28.1361 - val_loss: 41.5858
Epoch 921/1000
2/2 [==============================] - 1s 717ms/step - loss: 27.3141 - val_loss: 48.0219
Epoch 922/1000
2/2 [==============================] - 1s 687ms/step - loss: 27.2643 - val_loss: 47.5143
Epoch 923/1000
2/2 [==============================] - 1s 681ms/step - loss: 30.4947 - val_loss: 38.4766
Epoch 924/1000
2/2 [==============================] - 1s 694ms/step - loss: 25.8093 - val_loss: 39.1880
Epoch 925/1000
2/2 [==============================] - 1s 683ms/step - loss: 25.4669 - val_loss: 38.8028
Epoch 926/1000
2/2 [==============================] - 1s 700ms/step - loss: 28.4791 - val_loss: 37.5433
Epoch 927/1000
2/2 [==============================] - 1s 674ms/step - loss: 26.3481 - val_loss: 35.4374
Epoch 928/1000
2/2 [==============================] - 1s 700ms/step - loss: 27.5912 - val_loss: 36.5585
Epoch 929/1000
2/2 [==============================] - 1s 701ms/step - loss: 25.7922 - val_loss: 35.6523
Epoch 930/1000
2/2 [==============================] - 1s 701ms/step - loss: 28.3641 - val_loss: 38.6921
Epoch 931/1000
2/2 [==============================] - 1s 704ms/step - loss: 24.2192 - val_loss: 37.9325
Epoch 932/1000
2/2 [==============================] - 1s 699ms/step - loss: 29.3473 - val_loss: 41.7065
Epoch 933/1000
2/2 [==============================] - 1s 688ms/step - loss: 27.4438 - val_loss: 44.2319
Epoch 934/1000
2/2 [==============================] - 1s 674ms/step - loss: 26.6104 - val_loss: 41.1721
Epoch 935/1000
2/2 [==============================] - 1s 695ms/step - loss: 25.6384 - val_loss: 38.1668
Epoch 936/1000
2/2 [==============================] - 1s 695ms/step - loss: 25.1551 - val_loss: 41.8818
Epoch 937/1000
2/2 [==============================] - 1s 653ms/step - loss: 28.5088 - val_loss: 40.3398
Epoch 938/1000
2/2 [==============================] - 1s 688ms/step - loss: 27.7384 - val_loss: 40.1709
Epoch 939/1000
2/2 [==============================] - 1s 661ms/step - loss: 27.3032 - val_loss: 42.1984
Epoch 940/1000
2/2 [==============================] - 1s 684ms/step - loss: 26.4706 - val_loss: 41.2957
Epoch 941/1000
2/2 [==============================] - 1s 692ms/step - loss: 27.4714 - val_loss: 42.1667
Epoch 942/1000
2/2 [==============================] - 1s 706ms/step - loss: 24.8465 - val_loss: 40.1129
Epoch 943/1000
2/2 [==============================] - 1s 694ms/step - loss: 27.0154 - val_loss: 43.2371
Epoch 944/1000
2/2 [==============================] - 1s 697ms/step - loss: 23.1423 - val_loss: 45.4148
Epoch 945/1000
2/2 [==============================] - 1s 679ms/step - loss: 23.9419 - val_loss: 40.9840
Epoch 946/1000
2/2 [==============================] - 1s 689ms/step - loss: 28.2605 - val_loss: 45.2223
Epoch 947/1000
2/2 [==============================] - 1s 733ms/step - loss: 25.0737 - val_loss: 48.4394
Epoch 948/1000
2/2 [==============================] - 1s 694ms/step - loss: 29.2103 - val_loss: 43.4900
Epoch 949/1000
2/2 [==============================] - 1s 707ms/step - loss: 26.9660 - val_loss: 44.2771
Epoch 950/1000
2/2 [==============================] - 1s 724ms/step - loss: 25.3754 - val_loss: 46.6540
Epoch 951/1000
2/2 [==============================] - 1s 717ms/step - loss: 23.5237 - val_loss: 42.4122
Epoch 952/1000
2/2 [==============================] - 1s 674ms/step - loss: 23.9127 - val_loss: 41.6710
Epoch 953/1000
2/2 [==============================] - 1s 700ms/step - loss: 28.6852 - val_loss: 41.4525
Epoch 954/1000
2/2 [==============================] - 1s 713ms/step - loss: 25.2797 - val_loss: 41.2844
Epoch 955/1000
2/2 [==============================] - 1s 646ms/step - loss: 24.9585 - val_loss: 46.2278
Epoch 956/1000
2/2 [==============================] - 1s 686ms/step - loss: 28.2044 - val_loss: 37.7526
Epoch 957/1000
2/2 [==============================] - 1s 689ms/step - loss: 28.4031 - val_loss: 38.8367
Epoch 958/1000
2/2 [==============================] - 1s 670ms/step - loss: 27.4076 - val_loss: 36.2606
Epoch 959/1000
2/2 [==============================] - 1s 655ms/step - loss: 30.1536 - val_loss: 36.6165
Epoch 960/1000
2/2 [==============================] - 1s 680ms/step - loss: 26.9103 - val_loss: 43.2022
Epoch 961/1000
2/2 [==============================] - 1s 700ms/step - loss: 25.6185 - val_loss: 39.7292
Epoch 962/1000
2/2 [==============================] - 1s 673ms/step - loss: 26.6749 - val_loss: 41.8903
Epoch 963/1000
2/2 [==============================] - 1s 699ms/step - loss: 24.4171 - val_loss: 41.4557
Epoch 964/1000
2/2 [==============================] - 1s 690ms/step - loss: 25.4216 - val_loss: 42.8401
Epoch 965/1000
2/2 [==============================] - 1s 694ms/step - loss: 26.3831 - val_loss: 41.9555
Epoch 966/1000
2/2 [==============================] - 1s 685ms/step - loss: 29.2405 - val_loss: 44.8700
Epoch 967/1000
2/2 [==============================] - 1s 691ms/step - loss: 25.8807 - val_loss: 39.7436
Epoch 968/1000
2/2 [==============================] - 1s 703ms/step - loss: 28.6898 - val_loss: 39.3143
Epoch 969/1000
2/2 [==============================] - 1s 667ms/step - loss: 27.7574 - val_loss: 35.7768
Epoch 970/1000
2/2 [==============================] - 1s 673ms/step - loss: 26.9536 - val_loss: 39.7704
Epoch 971/1000
2/2 [==============================] - 1s 698ms/step - loss: 26.4141 - val_loss: 34.2213
Epoch 972/1000
2/2 [==============================] - 1s 709ms/step - loss: 25.8178 - val_loss: 37.2995
Epoch 973/1000
2/2 [==============================] - 1s 671ms/step - loss: 27.7513 - val_loss: 34.8151
Epoch 974/1000
2/2 [==============================] - 1s 672ms/step - loss: 23.8862 - val_loss: 34.7832
Epoch 975/1000
2/2 [==============================] - 1s 677ms/step - loss: 27.3787 - val_loss: 42.3015
Epoch 976/1000
2/2 [==============================] - 1s 694ms/step - loss: 26.3037 - val_loss: 41.3247
Epoch 977/1000
2/2 [==============================] - 1s 664ms/step - loss: 31.3219 - val_loss: 37.3191
Epoch 978/1000
2/2 [==============================] - 1s 686ms/step - loss: 26.3275 - val_loss: 38.5429
Epoch 979/1000
2/2 [==============================] - 1s 670ms/step - loss: 26.3383 - val_loss: 37.6338
Epoch 980/1000
2/2 [==============================] - 1s 690ms/step - loss: 25.3666 - val_loss: 41.0435
Epoch 981/1000
2/2 [==============================] - 1s 686ms/step - loss: 24.9208 - val_loss: 37.5422
Epoch 982/1000
2/2 [==============================] - 1s 670ms/step - loss: 27.9078 - val_loss: 44.2080
Epoch 983/1000
2/2 [==============================] - 1s 694ms/step - loss: 26.2418 - val_loss: 46.7854
Epoch 984/1000
2/2 [==============================] - 1s 679ms/step - loss: 25.0556 - val_loss: 38.5311
Epoch 985/1000
2/2 [==============================] - 1s 684ms/step - loss: 25.7830 - val_loss: 40.2685
Epoch 986/1000
2/2 [==============================] - 1s 680ms/step - loss: 26.6815 - val_loss: 45.0740
Epoch 987/1000
2/2 [==============================] - 1s 706ms/step - loss: 24.9262 - val_loss: 48.2825
Epoch 988/1000
2/2 [==============================] - 1s 691ms/step - loss: 25.2088 - val_loss: 41.4746
Epoch 989/1000
2/2 [==============================] - 1s 716ms/step - loss: 23.9640 - val_loss: 45.4634
Epoch 990/1000
2/2 [==============================] - 1s 686ms/step - loss: 25.5766 - val_loss: 41.1606
Epoch 991/1000
2/2 [==============================] - 1s 650ms/step - loss: 29.0632 - val_loss: 47.6571
Epoch 992/1000
2/2 [==============================] - 1s 708ms/step - loss: 24.7369 - val_loss: 35.0405
Epoch 993/1000
2/2 [==============================] - 1s 689ms/step - loss: 23.8154 - val_loss: 37.5268
Epoch 994/1000
2/2 [==============================] - 1s 710ms/step - loss: 25.5490 - val_loss: 34.8837
Epoch 995/1000
2/2 [==============================] - 1s 685ms/step - loss: 26.3559 - val_loss: 39.3258
Epoch 996/1000
2/2 [==============================] - 1s 689ms/step - loss: 25.3620 - val_loss: 36.5128
Epoch 997/1000
2/2 [==============================] - 1s 655ms/step - loss: 26.9342 - val_loss: 36.6954
Epoch 998/1000
2/2 [==============================] - 1s 692ms/step - loss: 26.0467 - val_loss: 42.7822
Epoch 999/1000
2/2 [==============================] - 1s 679ms/step - loss: 25.7131 - val_loss: 42.0424
Epoch 1000/1000
2/2 [==============================] - 1s 693ms/step - loss: 25.4310 - val_loss: 41.9392

keras-yolo3-实时眼睛鼻子嘴巴监测系统

keras-yolo3-实时眼睛鼻子嘴巴监测系统

keras-yolo3-实时口罩监测系统监测系统

.

2020.01.02 20:54

你可能感兴趣的:(人工智能)