Tensorflow2（预课程）---3.1、Iris分类-层方式
Tensorflow2（预课程）---3.1、Iris分类-层方式
一、总结
一句话总结：
我们可以很清楚的发现，输入是4维，输出是1维，并且输出是分类问题，分3类（one_hot编码后），所以对应模型可以弄成4->n->3
# 构建容器 model = tf.keras.Sequential() # 输出层 model.add(tf.keras.Input(shape=(4,))) # 中间层 model.add(tf.keras.layers.Dense(10,activation='relu')) # 输出层 model.add(tf.keras.layers.Dense(3)) # 模型的结构 model.summary()
1、准确率不够优化方法？
1、增加训练次数：4-3的结构欠拟合，先训练500次，发现准确率不够，再训练5000次，发现欠拟合
2、增加网络层数：4-10-3，500次不够，换5000次，准确率为1
2、pandas打乱数据集？
(a)、data=data.sample(frac=1.0,random_state=116)#打乱所有数据
(b)、data=data.reset_index(drop=True) #打乱后的数据index也是乱的，用reset_index重新加一列index，drop=True表示丢弃原有index一列
3、对于分类问题，将分类转化为数字，然后one_hot编码？
(A)、转数字：train_y =data.iloc[:120,-1].replace({"Iris-setosa":0,"Iris-versicolor":1,"Iris-virginica":2})
(B)、one_hot编码：train_y = tf.one_hot(train_y, depth=3)
4、训练的时候加上准确率以及验证测试集？
(①)、metrics中加acc：model.compile(optimizer='adam',loss='mse',metrics=['acc'])
(②)、validation_data中放测试集数据：history = model.fit(train_x,train_y,epochs=5000,validation_data=(test_x,test_y))
二、Iris分类-层方式
博客对应课程的视频位置：
讲课讲法（准确率不够优化方法）：
1、增加训练次数：4-3的结构欠拟合，先训练500次，发现准确率不够，再训练5000次，发现欠拟合
2、增加网络层数：4-10-3，500次不够，换5000次，准确率为1步骤
1、读取数据集
2、拆分数据集（拆分成训练数据集和测试数据集）
3、构建模型
4、训练模型
5、检验模型需求
对Iris进行分类In [1]:
import pandas as pd
import numpy as np
import tensorflow as tf
import matplotlib.pyplot as plt
1、读取数据集
In [2]:
data = pd.read_csv('./iris.data',header=None)
# data.head() 前5行
data
Out[2]:
| 0 | 1 | 2 | 3 | 4 | |
|---|---|---|---|---|---|
| 0 | 5.1 | 3.5 | 1.4 | 0.2 | Iris-setosa |
| 1 | 4.9 | 3.0 | 1.4 | 0.2 | Iris-setosa |
| 2 | 4.7 | 3.2 | 1.3 | 0.2 | Iris-setosa |
| 3 | 4.6 | 3.1 | 1.5 | 0.2 | Iris-setosa |
| 4 | 5.0 | 3.6 | 1.4 | 0.2 | Iris-setosa |
| ... | ... | ... | ... | ... | ... |
| 145 | 6.7 | 3.0 | 5.2 | 2.3 | Iris-virginica |
| 146 | 6.3 | 2.5 | 5.0 | 1.9 | Iris-virginica |
| 147 | 6.5 | 3.0 | 5.2 | 2.0 | Iris-virginica |
| 148 | 6.2 | 3.4 | 5.4 | 2.3 | Iris-virginica |
| 149 | 5.9 | 3.0 | 5.1 | 1.8 | Iris-virginica |
150 rows × 5 columns
2、拆分数据集（拆分成训练数据集和测试数据集）
总共150行数据，前120训练，后30行预测
我需要打乱顺序
In [3]:
data=data.sample(frac=1.0,random_state=116)#打乱所有数据
data=data.reset_index(drop=True) #打乱后的数据index也是乱的，用reset_index重新加一列index，drop=True表示丢弃原有index一列
data
Out[3]:
| 0 | 1 | 2 | 3 | 4 | |
|---|---|---|---|---|---|
| 0 | 6.1 | 2.8 | 4.0 | 1.3 | Iris-versicolor |
| 1 | 6.3 | 3.3 | 4.7 | 1.6 | Iris-versicolor |
| 2 | 6.8 | 2.8 | 4.8 | 1.4 | Iris-versicolor |
| 3 | 5.3 | 3.7 | 1.5 | 0.2 | Iris-setosa |
| 4 | 5.4 | 3.4 | 1.7 | 0.2 | Iris-setosa |
| ... | ... | ... | ... | ... | ... |
| 145 | 7.4 | 2.8 | 6.1 | 1.9 | Iris-virginica |
| 146 | 5.1 | 3.8 | 1.6 | 0.2 | Iris-setosa |
| 147 | 5.0 | 3.3 | 1.4 | 0.2 | Iris-setosa |
| 148 | 6.7 | 3.3 | 5.7 | 2.1 | Iris-virginica |
| 149 | 5.5 | 2.3 | 4.0 | 1.3 | Iris-versicolor |
150 rows × 5 columns
In [4]:
train_x = data.iloc[:120,0:-1]
test_x = data.iloc[121:,0:-1]
train_y =data.iloc[:120,-1].replace({"Iris-setosa":0,"Iris-versicolor":1,"Iris-virginica":2})
test_y =data.iloc[121:,-1].replace({"Iris-setosa":0,"Iris-versicolor":1,"Iris-virginica":2})
In [5]:
train_y = tf.one_hot(train_y, depth=3)
test_y = tf.one_hot(test_y, depth=3)
3、构建模型
4->3 训练500次，测试集准确率只有0.72
4->10->3 训练500次，测试集准确率只有0.5172 这是因为训练次数太少了
4->3 训练5000次，测试集准确率也只有0.8左右
4->10->3 训练5000次，acc是1了
In [6]:
# 构建容器
model = tf.keras.Sequential()
# 输出层
model.add(tf.keras.Input(shape=(4,)))
# 中间层
model.add(tf.keras.layers.Dense(10,activation='relu'))
# 输出层
model.add(tf.keras.layers.Dense(3))
# 模型的结构
model.summary()
Model: "sequential" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= dense (Dense) (None, 10) 50 _________________________________________________________________ dense_1 (Dense) (None, 3) 33 ================================================================= Total params: 83 Trainable params: 83 Non-trainable params: 0 _________________________________________________________________
4、训练模型
In [7]:
# 配置优化函数和损失器
model.compile(optimizer='adam',loss='mse',metrics=['acc'])
# 开始训练
history = model.fit(train_x,train_y,epochs=5000,validation_data=(test_x,test_y))
Epoch 1/5000 4/4 [==============================] - 0s 34ms/step - loss: 7.6696 - acc: 0.3167 - val_loss: 5.7491 - val_acc: 0.4483 Epoch 2/5000 4/4 [==============================] - 0s 7ms/step - loss: 7.0254 - acc: 0.3167 - val_loss: 5.2289 - val_acc: 0.4483 Epoch 3/5000 4/4 [==============================] - 0s 10ms/step - loss: 6.3978 - acc: 0.3167 - val_loss: 4.7433 - val_acc: 0.4483 Epoch 4/5000 4/4 [==============================] - 0s 6ms/step - loss: 5.8265 - acc: 0.3167 - val_loss: 4.2884 - val_acc: 0.4483 Epoch 5/5000 4/4 [==============================] - 0s 6ms/step - loss: 5.3004 - acc: 0.3167 - val_loss: 3.8649 - val_acc: 0.4483 Epoch 6/5000 4/4 [==============================] - 0s 7ms/step - loss: 4.7814 - acc: 0.3333 - val_loss: 3.4773 - val_acc: 0.4483 Epoch 7/5000 4/4 [==============================] - 0s 6ms/step - loss: 4.3135 - acc: 0.3500 - val_loss: 3.1206 - val_acc: 0.4828 Epoch 8/5000 4/4 [==============================] - 0s 6ms/step - loss: 3.8878 - acc: 0.3667 - val_loss: 2.7925 - val_acc: 0.4828 Epoch 9/5000 4/4 [==============================] - 0s 6ms/step - loss: 3.4932 - acc: 0.4000 - val_loss: 2.4929 - val_acc: 0.5172 Epoch 10/5000 4/4 [==============================] - 0s 7ms/step - loss: 3.1345 - acc: 0.4417 - val_loss: 2.2200 - val_acc: 0.5862 Epoch 11/5000 4/4 [==============================] - 0s 7ms/step - loss: 2.8036 - acc: 0.4750 - val_loss: 1.9736 - val_acc: 0.6552 Epoch 12/5000 4/4 [==============================] - 0s 6ms/step - loss: 2.5092 - acc: 0.5167 - val_loss: 1.7506 - val_acc: 0.6897 Epoch 13/5000 4/4 [==============================] - 0s 7ms/step - loss: 2.2304 - acc: 0.5250 - val_loss: 1.5520 - val_acc: 0.6897 Epoch 14/5000 4/4 [==============================] - 0s 7ms/step - loss: 1.9880 - acc: 0.5500 - val_loss: 1.3734 - val_acc: 0.7586 Epoch 15/5000 4/4 [==============================] - 0s 7ms/step - loss: 1.7718 - acc: 0.5583 - val_loss: 1.2142 - val_acc: 0.7586 Epoch 16/5000 4/4 [==============================] - 0s 8ms/step - loss: 1.5773 - acc: 0.5667 - val_loss: 1.0727 - val_acc: 0.7586 Epoch 17/5000 4/4 [==============================] - 0s 8ms/step - loss: 1.4066 - acc: 0.5750 - val_loss: 0.9464 - val_acc: 0.7586 Epoch 18/5000 4/4 [==============================] - 0s 6ms/step - loss: 1.2499 - acc: 0.5750 - val_loss: 0.8361 - val_acc: 0.7586 Epoch 19/5000 4/4 [==============================] - 0s 6ms/step - loss: 1.1126 - acc: 0.5750 - val_loss: 0.7389 - val_acc: 0.7586 Epoch 20/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.9895 - acc: 0.5750 - val_loss: 0.6538 - val_acc: 0.7931 Epoch 21/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.8758 - acc: 0.5667 - val_loss: 0.5802 - val_acc: 0.7241 Epoch 22/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.7821 - acc: 0.5583 - val_loss: 0.5153 - val_acc: 0.6897 Epoch 23/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.6949 - acc: 0.5500 - val_loss: 0.4587 - val_acc: 0.6552 Epoch 24/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.6200 - acc: 0.5500 - val_loss: 0.4098 - val_acc: 0.6552 Epoch 25/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.5531 - acc: 0.5500 - val_loss: 0.3676 - val_acc: 0.6207 Epoch 26/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.4958 - acc: 0.5417 - val_loss: 0.3312 - val_acc: 0.5862 Epoch 27/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.4442 - acc: 0.5333 - val_loss: 0.2997 - val_acc: 0.6207 Epoch 28/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.3996 - acc: 0.5167 - val_loss: 0.2731 - val_acc: 0.6207 Epoch 29/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.3605 - acc: 0.5333 - val_loss: 0.2500 - val_acc: 0.6552 Epoch 30/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.3261 - acc: 0.5333 - val_loss: 0.2310 - val_acc: 0.6552 Epoch 31/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.2977 - acc: 0.5333 - val_loss: 0.2155 - val_acc: 0.6552 Epoch 32/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.2730 - acc: 0.5333 - val_loss: 0.2030 - val_acc: 0.6552 Epoch 33/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.2511 - acc: 0.5250 - val_loss: 0.1926 - val_acc: 0.6552 Epoch 34/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.2329 - acc: 0.5250 - val_loss: 0.1846 - val_acc: 0.6207 Epoch 35/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.2167 - acc: 0.5333 - val_loss: 0.1776 - val_acc: 0.6552 Epoch 36/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.2039 - acc: 0.5250 - val_loss: 0.1725 - val_acc: 0.5862 Epoch 37/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.1914 - acc: 0.5333 - val_loss: 0.1673 - val_acc: 0.6552 Epoch 38/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1817 - acc: 0.5333 - val_loss: 0.1638 - val_acc: 0.6207 Epoch 39/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.1734 - acc: 0.5583 - val_loss: 0.1607 - val_acc: 0.6207 Epoch 40/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1670 - acc: 0.5917 - val_loss: 0.1589 - val_acc: 0.6552 Epoch 41/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.1610 - acc: 0.6250 - val_loss: 0.1571 - val_acc: 0.6897 Epoch 42/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.1557 - acc: 0.6333 - val_loss: 0.1552 - val_acc: 0.6897 Epoch 43/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.1513 - acc: 0.6583 - val_loss: 0.1536 - val_acc: 0.6552 Epoch 44/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1477 - acc: 0.6750 - val_loss: 0.1524 - val_acc: 0.6552 Epoch 45/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.1445 - acc: 0.7000 - val_loss: 0.1515 - val_acc: 0.6552 Epoch 46/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1415 - acc: 0.7333 - val_loss: 0.1505 - val_acc: 0.6552 Epoch 47/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.1392 - acc: 0.7333 - val_loss: 0.1500 - val_acc: 0.6897 Epoch 48/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1372 - acc: 0.7417 - val_loss: 0.1495 - val_acc: 0.6897 Epoch 49/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1351 - acc: 0.7500 - val_loss: 0.1483 - val_acc: 0.6897 Epoch 50/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1333 - acc: 0.7667 - val_loss: 0.1474 - val_acc: 0.6897 Epoch 51/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1318 - acc: 0.7667 - val_loss: 0.1469 - val_acc: 0.6897 Epoch 52/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1305 - acc: 0.7667 - val_loss: 0.1461 - val_acc: 0.6897 Epoch 53/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1291 - acc: 0.7750 - val_loss: 0.1448 - val_acc: 0.6897 Epoch 54/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1280 - acc: 0.7750 - val_loss: 0.1445 - val_acc: 0.6897 Epoch 55/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1268 - acc: 0.7750 - val_loss: 0.1431 - val_acc: 0.6897 Epoch 56/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.1258 - acc: 0.7750 - val_loss: 0.1426 - val_acc: 0.6552 Epoch 57/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.1248 - acc: 0.7750 - val_loss: 0.1418 - val_acc: 0.6552 Epoch 58/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1240 - acc: 0.7750 - val_loss: 0.1413 - val_acc: 0.6552 Epoch 59/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.1230 - acc: 0.7667 - val_loss: 0.1400 - val_acc: 0.6552 Epoch 60/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.1223 - acc: 0.7750 - val_loss: 0.1386 - val_acc: 0.6552 Epoch 61/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1215 - acc: 0.7750 - val_loss: 0.1377 - val_acc: 0.6552 Epoch 62/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1208 - acc: 0.7750 - val_loss: 0.1368 - val_acc: 0.6552 Epoch 63/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1201 - acc: 0.7750 - val_loss: 0.1365 - val_acc: 0.6552 Epoch 64/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1195 - acc: 0.7750 - val_loss: 0.1355 - val_acc: 0.6552 Epoch 65/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1188 - acc: 0.7750 - val_loss: 0.1355 - val_acc: 0.6552 Epoch 66/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1183 - acc: 0.7750 - val_loss: 0.1344 - val_acc: 0.6552 Epoch 67/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.1175 - acc: 0.7750 - val_loss: 0.1342 - val_acc: 0.6552 Epoch 68/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1169 - acc: 0.7750 - val_loss: 0.1339 - val_acc: 0.6552 Epoch 69/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1165 - acc: 0.7667 - val_loss: 0.1334 - val_acc: 0.6552 Epoch 70/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1159 - acc: 0.7667 - val_loss: 0.1327 - val_acc: 0.6552 Epoch 71/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1154 - acc: 0.7667 - val_loss: 0.1317 - val_acc: 0.6552 Epoch 72/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1151 - acc: 0.7833 - val_loss: 0.1296 - val_acc: 0.7241 Epoch 73/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1144 - acc: 0.8000 - val_loss: 0.1292 - val_acc: 0.7241 Epoch 74/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.1140 - acc: 0.8000 - val_loss: 0.1294 - val_acc: 0.6897 Epoch 75/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.1135 - acc: 0.8000 - val_loss: 0.1290 - val_acc: 0.6897 Epoch 76/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1130 - acc: 0.8000 - val_loss: 0.1288 - val_acc: 0.6897 Epoch 77/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1126 - acc: 0.8000 - val_loss: 0.1278 - val_acc: 0.7241 Epoch 78/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1122 - acc: 0.8083 - val_loss: 0.1268 - val_acc: 0.7241 Epoch 79/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1118 - acc: 0.8083 - val_loss: 0.1269 - val_acc: 0.7241 Epoch 80/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.1113 - acc: 0.8083 - val_loss: 0.1264 - val_acc: 0.7241 Epoch 81/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.1109 - acc: 0.8083 - val_loss: 0.1258 - val_acc: 0.7241 Epoch 82/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.1105 - acc: 0.8167 - val_loss: 0.1249 - val_acc: 0.7241 Epoch 83/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.1101 - acc: 0.8167 - val_loss: 0.1247 - val_acc: 0.7241 Epoch 84/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1097 - acc: 0.8167 - val_loss: 0.1248 - val_acc: 0.7241 Epoch 85/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1093 - acc: 0.8167 - val_loss: 0.1240 - val_acc: 0.7241 Epoch 86/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1089 - acc: 0.8167 - val_loss: 0.1240 - val_acc: 0.7241 Epoch 87/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1086 - acc: 0.8167 - val_loss: 0.1231 - val_acc: 0.7241 Epoch 88/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1082 - acc: 0.8167 - val_loss: 0.1234 - val_acc: 0.7241 Epoch 89/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.1078 - acc: 0.8167 - val_loss: 0.1229 - val_acc: 0.7241 Epoch 90/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.1074 - acc: 0.8167 - val_loss: 0.1224 - val_acc: 0.7241 Epoch 91/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.1071 - acc: 0.8167 - val_loss: 0.1219 - val_acc: 0.7241 Epoch 92/5000 4/4 [==============================] - 0s 9ms/step - loss: 0.1068 - acc: 0.8167 - val_loss: 0.1214 - val_acc: 0.7241 Epoch 93/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1064 - acc: 0.8167 - val_loss: 0.1199 - val_acc: 0.7241 Epoch 94/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.1061 - acc: 0.8333 - val_loss: 0.1189 - val_acc: 0.7241 Epoch 95/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1058 - acc: 0.8333 - val_loss: 0.1187 - val_acc: 0.7241 Epoch 96/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.1054 - acc: 0.8333 - val_loss: 0.1189 - val_acc: 0.7241 Epoch 97/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1051 - acc: 0.8333 - val_loss: 0.1191 - val_acc: 0.7241 Epoch 98/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1049 - acc: 0.8333 - val_loss: 0.1204 - val_acc: 0.7241 Epoch 99/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1045 - acc: 0.8250 - val_loss: 0.1198 - val_acc: 0.7241 Epoch 100/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1041 - acc: 0.8250 - val_loss: 0.1196 - val_acc: 0.7241 Epoch 101/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.1038 - acc: 0.8250 - val_loss: 0.1184 - val_acc: 0.7241 Epoch 102/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1035 - acc: 0.8417 - val_loss: 0.1175 - val_acc: 0.7241 Epoch 103/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1033 - acc: 0.8500 - val_loss: 0.1177 - val_acc: 0.7241 Epoch 104/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1028 - acc: 0.8333 - val_loss: 0.1175 - val_acc: 0.7241 Epoch 105/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1026 - acc: 0.8500 - val_loss: 0.1165 - val_acc: 0.7241 Epoch 106/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1024 - acc: 0.8417 - val_loss: 0.1168 - val_acc: 0.7241 Epoch 107/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1021 - acc: 0.8583 - val_loss: 0.1151 - val_acc: 0.7241 Epoch 108/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1017 - acc: 0.8583 - val_loss: 0.1151 - val_acc: 0.7241 Epoch 109/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1014 - acc: 0.8667 - val_loss: 0.1151 - val_acc: 0.7241 Epoch 110/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.1011 - acc: 0.8667 - val_loss: 0.1150 - val_acc: 0.7241 Epoch 111/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1009 - acc: 0.8583 - val_loss: 0.1162 - val_acc: 0.7241 Epoch 112/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.1006 - acc: 0.8417 - val_loss: 0.1165 - val_acc: 0.7241 Epoch 113/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1005 - acc: 0.8417 - val_loss: 0.1165 - val_acc: 0.7241 Epoch 114/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.1001 - acc: 0.8417 - val_loss: 0.1146 - val_acc: 0.7241 Epoch 115/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0998 - acc: 0.8667 - val_loss: 0.1132 - val_acc: 0.7586 Epoch 116/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0996 - acc: 0.8667 - val_loss: 0.1126 - val_acc: 0.7586 Epoch 117/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0992 - acc: 0.8667 - val_loss: 0.1123 - val_acc: 0.7586 Epoch 118/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0991 - acc: 0.8667 - val_loss: 0.1120 - val_acc: 0.7586 Epoch 119/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0987 - acc: 0.8667 - val_loss: 0.1131 - val_acc: 0.7586 Epoch 120/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0986 - acc: 0.8667 - val_loss: 0.1125 - val_acc: 0.7586 Epoch 121/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0983 - acc: 0.8667 - val_loss: 0.1133 - val_acc: 0.7586 Epoch 122/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0980 - acc: 0.8667 - val_loss: 0.1133 - val_acc: 0.7586 Epoch 123/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0978 - acc: 0.8667 - val_loss: 0.1131 - val_acc: 0.7586 Epoch 124/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0977 - acc: 0.8583 - val_loss: 0.1138 - val_acc: 0.7241 Epoch 125/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0973 - acc: 0.8583 - val_loss: 0.1129 - val_acc: 0.7586 Epoch 126/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0971 - acc: 0.8583 - val_loss: 0.1110 - val_acc: 0.7586 Epoch 127/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0968 - acc: 0.8667 - val_loss: 0.1102 - val_acc: 0.7586 Epoch 128/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0966 - acc: 0.8750 - val_loss: 0.1095 - val_acc: 0.7586 Epoch 129/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0963 - acc: 0.8667 - val_loss: 0.1105 - val_acc: 0.7586 Epoch 130/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0961 - acc: 0.8667 - val_loss: 0.1105 - val_acc: 0.7586 Epoch 131/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0959 - acc: 0.8667 - val_loss: 0.1109 - val_acc: 0.7586 Epoch 132/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0957 - acc: 0.8667 - val_loss: 0.1107 - val_acc: 0.7586 Epoch 133/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0955 - acc: 0.8667 - val_loss: 0.1096 - val_acc: 0.7586 Epoch 134/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0952 - acc: 0.8667 - val_loss: 0.1095 - val_acc: 0.7586 Epoch 135/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0950 - acc: 0.8667 - val_loss: 0.1090 - val_acc: 0.7586 Epoch 136/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0948 - acc: 0.8667 - val_loss: 0.1086 - val_acc: 0.7586 Epoch 137/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0945 - acc: 0.8833 - val_loss: 0.1074 - val_acc: 0.7586 Epoch 138/5000 4/4 [==============================] - 0s 9ms/step - loss: 0.0943 - acc: 0.8833 - val_loss: 0.1066 - val_acc: 0.7586 Epoch 139/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0941 - acc: 0.8833 - val_loss: 0.1067 - val_acc: 0.7586 Epoch 140/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0940 - acc: 0.8833 - val_loss: 0.1061 - val_acc: 0.7586 Epoch 141/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0938 - acc: 0.8833 - val_loss: 0.1068 - val_acc: 0.7586 Epoch 142/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0935 - acc: 0.8833 - val_loss: 0.1061 - val_acc: 0.7586 Epoch 143/5000 4/4 [==============================] - 0s 9ms/step - loss: 0.0932 - acc: 0.8833 - val_loss: 0.1064 - val_acc: 0.7586 Epoch 144/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0930 - acc: 0.8833 - val_loss: 0.1064 - val_acc: 0.7586 Epoch 145/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0928 - acc: 0.8833 - val_loss: 0.1063 - val_acc: 0.7586 Epoch 146/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0926 - acc: 0.8833 - val_loss: 0.1068 - val_acc: 0.7586 Epoch 147/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0925 - acc: 0.8750 - val_loss: 0.1066 - val_acc: 0.7586 Epoch 148/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0922 - acc: 0.8833 - val_loss: 0.1061 - val_acc: 0.7586 Epoch 149/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0920 - acc: 0.8833 - val_loss: 0.1058 - val_acc: 0.7586 Epoch 150/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0918 - acc: 0.8833 - val_loss: 0.1056 - val_acc: 0.7586 Epoch 151/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0916 - acc: 0.8833 - val_loss: 0.1045 - val_acc: 0.7586 Epoch 152/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0914 - acc: 0.8833 - val_loss: 0.1038 - val_acc: 0.7586 Epoch 153/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0912 - acc: 0.8833 - val_loss: 0.1046 - val_acc: 0.7586 Epoch 154/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0910 - acc: 0.8833 - val_loss: 0.1037 - val_acc: 0.7586 Epoch 155/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0909 - acc: 0.8833 - val_loss: 0.1048 - val_acc: 0.7586 Epoch 156/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0905 - acc: 0.8833 - val_loss: 0.1049 - val_acc: 0.7586 Epoch 157/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0905 - acc: 0.8833 - val_loss: 0.1029 - val_acc: 0.7586 Epoch 158/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0901 - acc: 0.8833 - val_loss: 0.1023 - val_acc: 0.7586 Epoch 159/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0899 - acc: 0.8833 - val_loss: 0.1029 - val_acc: 0.7586 Epoch 160/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0896 - acc: 0.8833 - val_loss: 0.1025 - val_acc: 0.7586 Epoch 161/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0894 - acc: 0.8833 - val_loss: 0.1019 - val_acc: 0.7586 Epoch 162/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0893 - acc: 0.8833 - val_loss: 0.1022 - val_acc: 0.7586 Epoch 163/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0890 - acc: 0.8833 - val_loss: 0.1019 - val_acc: 0.7586 Epoch 164/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0888 - acc: 0.8833 - val_loss: 0.1008 - val_acc: 0.7586 Epoch 165/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0887 - acc: 0.8833 - val_loss: 0.1014 - val_acc: 0.7586 Epoch 166/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0884 - acc: 0.8833 - val_loss: 0.1008 - val_acc: 0.7586 Epoch 167/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0881 - acc: 0.8833 - val_loss: 0.1006 - val_acc: 0.7586 Epoch 168/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0879 - acc: 0.8833 - val_loss: 0.0996 - val_acc: 0.7586 Epoch 169/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0878 - acc: 0.8833 - val_loss: 0.0994 - val_acc: 0.7586 Epoch 170/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0875 - acc: 0.8833 - val_loss: 0.0987 - val_acc: 0.7586 Epoch 171/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0873 - acc: 0.8833 - val_loss: 0.0981 - val_acc: 0.7586 Epoch 172/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0871 - acc: 0.8833 - val_loss: 0.0977 - val_acc: 0.7586 Epoch 173/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0869 - acc: 0.8833 - val_loss: 0.0985 - val_acc: 0.7586 Epoch 174/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0866 - acc: 0.8833 - val_loss: 0.0988 - val_acc: 0.7586 Epoch 175/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0865 - acc: 0.8833 - val_loss: 0.0984 - val_acc: 0.7586 Epoch 176/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0863 - acc: 0.8833 - val_loss: 0.0981 - val_acc: 0.7586 Epoch 177/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0862 - acc: 0.8833 - val_loss: 0.0989 - val_acc: 0.7586 Epoch 178/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0858 - acc: 0.8833 - val_loss: 0.0984 - val_acc: 0.7586 Epoch 179/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0856 - acc: 0.8833 - val_loss: 0.0975 - val_acc: 0.7586 Epoch 180/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0855 - acc: 0.8833 - val_loss: 0.0959 - val_acc: 0.8276 Epoch 181/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0853 - acc: 0.9000 - val_loss: 0.0952 - val_acc: 0.8276 Epoch 182/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0852 - acc: 0.8917 - val_loss: 0.0958 - val_acc: 0.8276 Epoch 183/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0849 - acc: 0.8917 - val_loss: 0.0948 - val_acc: 0.8276 Epoch 184/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0846 - acc: 0.8917 - val_loss: 0.0957 - val_acc: 0.7931 Epoch 185/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0844 - acc: 0.8833 - val_loss: 0.0961 - val_acc: 0.7586 Epoch 186/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0841 - acc: 0.8833 - val_loss: 0.0960 - val_acc: 0.7586 Epoch 187/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0840 - acc: 0.8833 - val_loss: 0.0950 - val_acc: 0.8276 Epoch 188/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0838 - acc: 0.8917 - val_loss: 0.0950 - val_acc: 0.8276 Epoch 189/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0836 - acc: 0.8917 - val_loss: 0.0949 - val_acc: 0.8276 Epoch 190/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0833 - acc: 0.8917 - val_loss: 0.0944 - val_acc: 0.8276 Epoch 191/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0831 - acc: 0.8917 - val_loss: 0.0939 - val_acc: 0.8276 Epoch 192/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0830 - acc: 0.8917 - val_loss: 0.0934 - val_acc: 0.8276 Epoch 193/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0827 - acc: 0.9000 - val_loss: 0.0938 - val_acc: 0.8276 Epoch 194/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0828 - acc: 0.8917 - val_loss: 0.0948 - val_acc: 0.7586 Epoch 195/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0824 - acc: 0.8833 - val_loss: 0.0943 - val_acc: 0.7931 Epoch 196/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0821 - acc: 0.8917 - val_loss: 0.0928 - val_acc: 0.8621 Epoch 197/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0820 - acc: 0.9083 - val_loss: 0.0918 - val_acc: 0.8621 Epoch 198/5000 4/4 [==============================] - ETA: 0s - loss: 0.0830 - acc: 0.906 - 0s 9ms/step - loss: 0.0819 - acc: 0.9083 - val_loss: 0.0912 - val_acc: 0.8621 Epoch 199/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0816 - acc: 0.9083 - val_loss: 0.0910 - val_acc: 0.8621 Epoch 200/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0813 - acc: 0.9083 - val_loss: 0.0919 - val_acc: 0.8621 Epoch 201/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0811 - acc: 0.9000 - val_loss: 0.0929 - val_acc: 0.8276 Epoch 202/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0810 - acc: 0.8917 - val_loss: 0.0932 - val_acc: 0.8276 Epoch 203/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0808 - acc: 0.9000 - val_loss: 0.0922 - val_acc: 0.8621 Epoch 204/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0805 - acc: 0.9000 - val_loss: 0.0916 - val_acc: 0.8621 Epoch 205/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0803 - acc: 0.9000 - val_loss: 0.0909 - val_acc: 0.8621 Epoch 206/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0801 - acc: 0.9000 - val_loss: 0.0902 - val_acc: 0.8621 Epoch 207/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0799 - acc: 0.9083 - val_loss: 0.0895 - val_acc: 0.8621 Epoch 208/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0797 - acc: 0.9083 - val_loss: 0.0892 - val_acc: 0.8621 Epoch 209/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0795 - acc: 0.9083 - val_loss: 0.0887 - val_acc: 0.8621 Epoch 210/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0793 - acc: 0.9083 - val_loss: 0.0885 - val_acc: 0.8621 Epoch 211/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0791 - acc: 0.9083 - val_loss: 0.0891 - val_acc: 0.8621 Epoch 212/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0789 - acc: 0.9083 - val_loss: 0.0888 - val_acc: 0.8621 Epoch 213/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0789 - acc: 0.9000 - val_loss: 0.0898 - val_acc: 0.8621 Epoch 214/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0786 - acc: 0.9000 - val_loss: 0.0902 - val_acc: 0.8621 Epoch 215/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0783 - acc: 0.9000 - val_loss: 0.0899 - val_acc: 0.8621 Epoch 216/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0785 - acc: 0.9000 - val_loss: 0.0877 - val_acc: 0.8621 Epoch 217/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0779 - acc: 0.9083 - val_loss: 0.0872 - val_acc: 0.8621 Epoch 218/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0779 - acc: 0.9000 - val_loss: 0.0881 - val_acc: 0.8621 Epoch 219/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0775 - acc: 0.9000 - val_loss: 0.0879 - val_acc: 0.8621 Epoch 220/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0773 - acc: 0.9083 - val_loss: 0.0871 - val_acc: 0.8621 Epoch 221/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0771 - acc: 0.9083 - val_loss: 0.0874 - val_acc: 0.8621 Epoch 222/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0770 - acc: 0.9083 - val_loss: 0.0867 - val_acc: 0.8621 Epoch 223/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0767 - acc: 0.9083 - val_loss: 0.0869 - val_acc: 0.8621 Epoch 224/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0765 - acc: 0.9083 - val_loss: 0.0872 - val_acc: 0.8621 Epoch 225/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0763 - acc: 0.9000 - val_loss: 0.0869 - val_acc: 0.8621 Epoch 226/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0761 - acc: 0.9083 - val_loss: 0.0863 - val_acc: 0.8621 Epoch 227/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0759 - acc: 0.9167 - val_loss: 0.0859 - val_acc: 0.8621 Epoch 228/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0757 - acc: 0.9167 - val_loss: 0.0855 - val_acc: 0.8621 Epoch 229/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0755 - acc: 0.9167 - val_loss: 0.0852 - val_acc: 0.8966 Epoch 230/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0753 - acc: 0.9167 - val_loss: 0.0852 - val_acc: 0.8621 Epoch 231/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0751 - acc: 0.9167 - val_loss: 0.0847 - val_acc: 0.8966 Epoch 232/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0750 - acc: 0.9167 - val_loss: 0.0840 - val_acc: 0.8966 Epoch 233/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0747 - acc: 0.9167 - val_loss: 0.0842 - val_acc: 0.8966 Epoch 234/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0746 - acc: 0.9167 - val_loss: 0.0844 - val_acc: 0.8966 Epoch 235/5000 4/4 [==============================] - 0s 10ms/step - loss: 0.0743 - acc: 0.9167 - val_loss: 0.0840 - val_acc: 0.8966 Epoch 236/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0743 - acc: 0.9167 - val_loss: 0.0827 - val_acc: 0.9310 Epoch 237/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0740 - acc: 0.9250 - val_loss: 0.0828 - val_acc: 0.9310 Epoch 238/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0738 - acc: 0.9250 - val_loss: 0.0828 - val_acc: 0.9310 Epoch 239/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0736 - acc: 0.9167 - val_loss: 0.0827 - val_acc: 0.9310 Epoch 240/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0734 - acc: 0.9250 - val_loss: 0.0824 - val_acc: 0.9310 Epoch 241/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0733 - acc: 0.9250 - val_loss: 0.0823 - val_acc: 0.9310 Epoch 242/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0730 - acc: 0.9250 - val_loss: 0.0820 - val_acc: 0.9310 Epoch 243/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0729 - acc: 0.9167 - val_loss: 0.0823 - val_acc: 0.9310 Epoch 244/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0726 - acc: 0.9167 - val_loss: 0.0821 - val_acc: 0.9310 Epoch 245/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0726 - acc: 0.9333 - val_loss: 0.0805 - val_acc: 0.9310 Epoch 246/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0724 - acc: 0.9250 - val_loss: 0.0809 - val_acc: 0.9310 Epoch 247/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0720 - acc: 0.9250 - val_loss: 0.0806 - val_acc: 0.9310 Epoch 248/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0718 - acc: 0.9250 - val_loss: 0.0805 - val_acc: 0.9310 Epoch 249/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0717 - acc: 0.9250 - val_loss: 0.0804 - val_acc: 0.9310 Epoch 250/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0715 - acc: 0.9250 - val_loss: 0.0802 - val_acc: 0.9310 Epoch 251/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0713 - acc: 0.9333 - val_loss: 0.0798 - val_acc: 0.9310 Epoch 252/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0711 - acc: 0.9250 - val_loss: 0.0801 - val_acc: 0.9310 Epoch 253/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0710 - acc: 0.9250 - val_loss: 0.0801 - val_acc: 0.9310 Epoch 254/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0708 - acc: 0.9333 - val_loss: 0.0790 - val_acc: 0.9310 Epoch 255/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0706 - acc: 0.9417 - val_loss: 0.0790 - val_acc: 0.9310 Epoch 256/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0706 - acc: 0.9417 - val_loss: 0.0777 - val_acc: 0.9310 Epoch 257/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0702 - acc: 0.9417 - val_loss: 0.0781 - val_acc: 0.9310 Epoch 258/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0700 - acc: 0.9417 - val_loss: 0.0782 - val_acc: 0.9310 Epoch 259/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0702 - acc: 0.9417 - val_loss: 0.0774 - val_acc: 0.9310 Epoch 260/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0698 - acc: 0.9333 - val_loss: 0.0787 - val_acc: 0.9310 Epoch 261/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0695 - acc: 0.9333 - val_loss: 0.0779 - val_acc: 0.9310 Epoch 262/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0692 - acc: 0.9417 - val_loss: 0.0775 - val_acc: 0.9310 Epoch 263/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0691 - acc: 0.9417 - val_loss: 0.0773 - val_acc: 0.9310 Epoch 264/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0689 - acc: 0.9417 - val_loss: 0.0766 - val_acc: 0.9310 Epoch 265/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0687 - acc: 0.9417 - val_loss: 0.0764 - val_acc: 0.9310 Epoch 266/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0685 - acc: 0.9417 - val_loss: 0.0763 - val_acc: 0.9310 Epoch 267/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0683 - acc: 0.9417 - val_loss: 0.0761 - val_acc: 0.9310 Epoch 268/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0681 - acc: 0.9417 - val_loss: 0.0762 - val_acc: 0.9310 Epoch 269/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0680 - acc: 0.9417 - val_loss: 0.0764 - val_acc: 0.9310 Epoch 270/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0678 - acc: 0.9417 - val_loss: 0.0756 - val_acc: 0.9310 Epoch 271/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0676 - acc: 0.9417 - val_loss: 0.0756 - val_acc: 0.9310 Epoch 272/5000 4/4 [==============================] - 0s 9ms/step - loss: 0.0674 - acc: 0.9417 - val_loss: 0.0754 - val_acc: 0.9310 Epoch 273/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0673 - acc: 0.9417 - val_loss: 0.0748 - val_acc: 0.9310 Epoch 274/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0671 - acc: 0.9417 - val_loss: 0.0744 - val_acc: 0.9310 Epoch 275/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0669 - acc: 0.9417 - val_loss: 0.0743 - val_acc: 0.9310 Epoch 276/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0668 - acc: 0.9500 - val_loss: 0.0736 - val_acc: 0.9310 Epoch 277/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0666 - acc: 0.9500 - val_loss: 0.0742 - val_acc: 0.9310 Epoch 278/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0665 - acc: 0.9417 - val_loss: 0.0750 - val_acc: 0.9310 Epoch 279/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0662 - acc: 0.9417 - val_loss: 0.0743 - val_acc: 0.9310 Epoch 280/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0661 - acc: 0.9417 - val_loss: 0.0734 - val_acc: 0.9310 Epoch 281/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0659 - acc: 0.9417 - val_loss: 0.0730 - val_acc: 0.9310 Epoch 282/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0658 - acc: 0.9417 - val_loss: 0.0732 - val_acc: 0.9310 Epoch 283/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0656 - acc: 0.9500 - val_loss: 0.0724 - val_acc: 0.9310 Epoch 284/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0655 - acc: 0.9500 - val_loss: 0.0722 - val_acc: 0.9310 Epoch 285/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0653 - acc: 0.9500 - val_loss: 0.0720 - val_acc: 0.9310 Epoch 286/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0651 - acc: 0.9500 - val_loss: 0.0721 - val_acc: 0.9310 Epoch 287/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0649 - acc: 0.9500 - val_loss: 0.0724 - val_acc: 0.9310 Epoch 288/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0648 - acc: 0.9417 - val_loss: 0.0724 - val_acc: 0.9310 Epoch 289/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0647 - acc: 0.9417 - val_loss: 0.0725 - val_acc: 0.9310 Epoch 290/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0645 - acc: 0.9417 - val_loss: 0.0714 - val_acc: 0.9310 Epoch 291/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0643 - acc: 0.9500 - val_loss: 0.0709 - val_acc: 0.9310 Epoch 292/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0643 - acc: 0.9500 - val_loss: 0.0702 - val_acc: 0.9310 Epoch 293/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0640 - acc: 0.9500 - val_loss: 0.0702 - val_acc: 0.9310 Epoch 294/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0640 - acc: 0.9500 - val_loss: 0.0708 - val_acc: 0.9310 Epoch 295/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0638 - acc: 0.9500 - val_loss: 0.0702 - val_acc: 0.9310 Epoch 296/5000 4/4 [==============================] - 0s 9ms/step - loss: 0.0636 - acc: 0.9500 - val_loss: 0.0697 - val_acc: 0.9310 Epoch 297/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0635 - acc: 0.9500 - val_loss: 0.0701 - val_acc: 0.9310 Epoch 298/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0633 - acc: 0.9500 - val_loss: 0.0699 - val_acc: 0.9310 Epoch 299/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0634 - acc: 0.9500 - val_loss: 0.0700 - val_acc: 0.9310 Epoch 300/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0630 - acc: 0.9500 - val_loss: 0.0690 - val_acc: 0.9310 Epoch 301/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0632 - acc: 0.9583 - val_loss: 0.0678 - val_acc: 0.9310 Epoch 302/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0627 - acc: 0.9583 - val_loss: 0.0685 - val_acc: 0.9310 Epoch 303/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0626 - acc: 0.9583 - val_loss: 0.0695 - val_acc: 0.9310 Epoch 304/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0625 - acc: 0.9500 - val_loss: 0.0699 - val_acc: 0.9310 Epoch 305/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0625 - acc: 0.9500 - val_loss: 0.0683 - val_acc: 0.9310 Epoch 306/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0622 - acc: 0.9500 - val_loss: 0.0685 - val_acc: 0.9310 Epoch 307/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0620 - acc: 0.9500 - val_loss: 0.0683 - val_acc: 0.9310 Epoch 308/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0618 - acc: 0.9583 - val_loss: 0.0674 - val_acc: 0.9310 Epoch 309/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0617 - acc: 0.9583 - val_loss: 0.0670 - val_acc: 0.9310 Epoch 310/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0617 - acc: 0.9583 - val_loss: 0.0674 - val_acc: 0.9310 Epoch 311/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0614 - acc: 0.9583 - val_loss: 0.0670 - val_acc: 0.9310 Epoch 312/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0614 - acc: 0.9583 - val_loss: 0.0665 - val_acc: 0.9310 Epoch 313/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0613 - acc: 0.9583 - val_loss: 0.0676 - val_acc: 0.9310 Epoch 314/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0611 - acc: 0.9583 - val_loss: 0.0671 - val_acc: 0.9310 Epoch 315/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0610 - acc: 0.9583 - val_loss: 0.0672 - val_acc: 0.9310 Epoch 316/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0608 - acc: 0.9583 - val_loss: 0.0667 - val_acc: 0.9310 Epoch 317/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0607 - acc: 0.9583 - val_loss: 0.0658 - val_acc: 0.9310 Epoch 318/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0606 - acc: 0.9583 - val_loss: 0.0652 - val_acc: 0.9310 Epoch 319/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0605 - acc: 0.9583 - val_loss: 0.0658 - val_acc: 0.9310 Epoch 320/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0603 - acc: 0.9583 - val_loss: 0.0658 - val_acc: 0.9310 Epoch 321/5000 4/4 [==============================] - 0s 10ms/step - loss: 0.0602 - acc: 0.9583 - val_loss: 0.0651 - val_acc: 0.9310 Epoch 322/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0600 - acc: 0.9583 - val_loss: 0.0647 - val_acc: 0.9310 Epoch 323/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0600 - acc: 0.9583 - val_loss: 0.0644 - val_acc: 0.9310 Epoch 324/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0598 - acc: 0.9583 - val_loss: 0.0646 - val_acc: 0.9310 Epoch 325/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0597 - acc: 0.9583 - val_loss: 0.0650 - val_acc: 0.9310 Epoch 326/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0596 - acc: 0.9583 - val_loss: 0.0642 - val_acc: 0.9310 Epoch 327/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0594 - acc: 0.9583 - val_loss: 0.0643 - val_acc: 0.9310 Epoch 328/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0593 - acc: 0.9583 - val_loss: 0.0637 - val_acc: 0.9310 Epoch 329/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0592 - acc: 0.9583 - val_loss: 0.0636 - val_acc: 0.9310 Epoch 330/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0591 - acc: 0.9583 - val_loss: 0.0636 - val_acc: 0.9310 Epoch 331/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0589 - acc: 0.9583 - val_loss: 0.0637 - val_acc: 0.9310 Epoch 332/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0590 - acc: 0.9583 - val_loss: 0.0641 - val_acc: 0.9310 Epoch 333/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0587 - acc: 0.9583 - val_loss: 0.0631 - val_acc: 0.9310 Epoch 334/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0586 - acc: 0.9583 - val_loss: 0.0624 - val_acc: 0.9310 Epoch 335/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0585 - acc: 0.9583 - val_loss: 0.0621 - val_acc: 0.9310 Epoch 336/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0586 - acc: 0.9583 - val_loss: 0.0613 - val_acc: 0.9655 Epoch 337/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0583 - acc: 0.9583 - val_loss: 0.0620 - val_acc: 0.9310 Epoch 338/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0583 - acc: 0.9583 - val_loss: 0.0630 - val_acc: 0.9310 Epoch 339/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0581 - acc: 0.9583 - val_loss: 0.0625 - val_acc: 0.9310 Epoch 340/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0579 - acc: 0.9583 - val_loss: 0.0622 - val_acc: 0.9310 Epoch 341/5000 4/4 [==============================] - 0s 9ms/step - loss: 0.0578 - acc: 0.9583 - val_loss: 0.0617 - val_acc: 0.9310 Epoch 342/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0577 - acc: 0.9583 - val_loss: 0.0613 - val_acc: 0.9655 Epoch 343/5000 4/4 [==============================] - 0s 10ms/step - loss: 0.0576 - acc: 0.9583 - val_loss: 0.0612 - val_acc: 0.9655 Epoch 344/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0574 - acc: 0.9583 - val_loss: 0.0618 - val_acc: 0.9310 Epoch 345/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0574 - acc: 0.9583 - val_loss: 0.0622 - val_acc: 0.9310 Epoch 346/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0574 - acc: 0.9583 - val_loss: 0.0614 - val_acc: 0.9310 Epoch 347/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0571 - acc: 0.9583 - val_loss: 0.0610 - val_acc: 0.9310 Epoch 348/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0573 - acc: 0.9583 - val_loss: 0.0601 - val_acc: 0.9655 Epoch 349/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0569 - acc: 0.9583 - val_loss: 0.0606 - val_acc: 0.9655 Epoch 350/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0569 - acc: 0.9583 - val_loss: 0.0603 - val_acc: 0.9655 Epoch 351/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0567 - acc: 0.9583 - val_loss: 0.0606 - val_acc: 0.9310 Epoch 352/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0567 - acc: 0.9583 - val_loss: 0.0605 - val_acc: 0.9310 Epoch 353/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0566 - acc: 0.9583 - val_loss: 0.0598 - val_acc: 0.9655 Epoch 354/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0565 - acc: 0.9583 - val_loss: 0.0602 - val_acc: 0.9655 Epoch 355/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0563 - acc: 0.9583 - val_loss: 0.0601 - val_acc: 0.9655 Epoch 356/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0563 - acc: 0.9583 - val_loss: 0.0588 - val_acc: 0.9655 Epoch 357/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0561 - acc: 0.9583 - val_loss: 0.0585 - val_acc: 0.9655 Epoch 358/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0561 - acc: 0.9583 - val_loss: 0.0591 - val_acc: 0.9655 Epoch 359/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0559 - acc: 0.9583 - val_loss: 0.0588 - val_acc: 0.9655 Epoch 360/5000 4/4 [==============================] - 0s 6ms/step - loss: 0.0558 - acc: 0.9583 - val_loss: 0.0585 - val_acc: 0.9655 Epoch 361/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0558 - acc: 0.9583 - val_loss: 0.0586 - val_acc: 0.9655 Epoch 362/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0556 - acc: 0.9583 - val_loss: 0.0577 - val_acc: 0.9655 Epoch 363/5000 4/4 [==============================] - 0s 10ms/step - loss: 0.0555 - acc: 0.9583 - val_loss: 0.0574 - val_acc: 1.0000 Epoch 364/5000 4/4 [==============================] - 0s 9ms/step - loss: 0.0554 - acc: 0.9583 - val_loss: 0.0573 - val_acc: 1.0000 Epoch 365/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0553 - acc: 0.9583 - val_loss: 0.0573 - val_acc: 1.0000 Epoch 366/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0552 - acc: 0.9583 - val_loss: 0.0577 - val_acc: 0.9655 Epoch 367/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0551 - acc: 0.9583 - val_loss: 0.0578 - val_acc: 0.9655 Epoch 368/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0551 - acc: 0.9583 - val_loss: 0.0574 - val_acc: 0.9655 Epoch 369/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0549 - acc: 0.9583 - val_loss: 0.0572 - val_acc: 0.9655 Epoch 370/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0548 - acc: 0.9583 - val_loss: 0.0574 - val_acc: 0.9655 Epoch 371/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0548 - acc: 0.9583 - val_loss: 0.0568 - val_acc: 1.0000 Epoch 372/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0547 - acc: 0.9583 - val_loss: 0.0565 - val_acc: 1.0000 Epoch 373/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0546 - acc: 0.9583 - val_loss: 0.0571 - val_acc: 0.9655 Epoch 374/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0545 - acc: 0.9583 - val_loss: 0.0572 - val_acc: 0.9655 Epoch 375/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0544 - acc: 0.9583 - val_loss: 0.0566 - val_acc: 0.9655 Epoch 376/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0543 - acc: 0.9583 - val_loss: 0.0558 - val_acc: 1.0000 Epoch 377/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0542 - acc: 0.9583 - val_loss: 0.0558 - val_acc: 1.0000 Epoch 378/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0541 - acc: 0.9583 - val_loss: 0.0558 - val_acc: 1.0000 Epoch 379/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0541 - acc: 0.9583 - val_loss: 0.0562 - val_acc: 1.0000 Epoch 380/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0540 - acc: 0.9583 - val_loss: 0.0554 - val_acc: 1.0000 Epoch 381/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0540 - acc: 0.9583 - val_loss: 0.0559 - val_acc: 1.0000 Epoch 382/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0538 - acc: 0.9583 - val_loss: 0.0557 - val_acc: 1.0000 Epoch 383/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0538 - acc: 0.9583 - val_loss: 0.0546 - val_acc: 1.0000 Epoch 384/5000 4/4 [==============================] - 0s 9ms/step - loss: 0.0537 - acc: 0.9583 - val_loss: 0.0547 - val_acc: 1.0000 Epoch 385/5000 4/4 [==============================] - 0s 9ms/step - loss: 0.0537 - acc: 0.9583 - val_loss: 0.0545 - val_acc: 1.0000 Epoch 386/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0536 - acc: 0.9583 - val_loss: 0.0556 - val_acc: 1.0000 Epoch 387/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0534 - acc: 0.9583 - val_loss: 0.0556 - val_acc: 1.0000 Epoch 388/5000 4/4 [==============================] - 0s 9ms/step - loss: 0.0535 - acc: 0.9583 - val_loss: 0.0550 - val_acc: 1.0000 Epoch 389/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0533 - acc: 0.9583 - val_loss: 0.0551 - val_acc: 1.0000 Epoch 390/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0533 - acc: 0.9583 - val_loss: 0.0555 - val_acc: 1.0000 Epoch 391/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0532 - acc: 0.9583 - val_loss: 0.0552 - val_acc: 1.0000 Epoch 392/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0531 - acc: 0.9583 - val_loss: 0.0549 - val_acc: 1.0000 Epoch 393/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0531 - acc: 0.9583 - val_loss: 0.0539 - val_acc: 1.0000 Epoch 394/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0530 - acc: 0.9583 - val_loss: 0.0538 - val_acc: 1.0000 Epoch 395/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0533 - acc: 0.9583 - val_loss: 0.0530 - val_acc: 1.0000 Epoch 396/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0529 - acc: 0.9583 - val_loss: 0.0543 - val_acc: 1.0000 Epoch 397/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0530 - acc: 0.9583 - val_loss: 0.0550 - val_acc: 1.0000 Epoch 398/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0529 - acc: 0.9583 - val_loss: 0.0536 - val_acc: 1.0000 Epoch 399/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0526 - acc: 0.9583 - val_loss: 0.0534 - val_acc: 1.0000 Epoch 400/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0527 - acc: 0.9583 - val_loss: 0.0528 - val_acc: 1.0000 Epoch 401/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0526 - acc: 0.9583 - val_loss: 0.0538 - val_acc: 1.0000 Epoch 402/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0524 - acc: 0.9583 - val_loss: 0.0537 - val_acc: 1.0000 Epoch 403/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0523 - acc: 0.9583 - val_loss: 0.0530 - val_acc: 1.0000 Epoch 404/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0522 - acc: 0.9583 - val_loss: 0.0525 - val_acc: 1.0000 ........................................ Epoch 4990/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0207 - acc: 0.9833 - val_loss: 0.0066 - val_acc: 1.0000 Epoch 4991/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0201 - acc: 0.9833 - val_loss: 0.0062 - val_acc: 1.0000 Epoch 4992/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0202 - acc: 0.9750 - val_loss: 0.0063 - val_acc: 1.0000 Epoch 4993/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0200 - acc: 0.9833 - val_loss: 0.0065 - val_acc: 1.0000 Epoch 4994/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0202 - acc: 0.9833 - val_loss: 0.0065 - val_acc: 1.0000 Epoch 4995/5000 4/4 [==============================] - 0s 9ms/step - loss: 0.0204 - acc: 0.9833 - val_loss: 0.0063 - val_acc: 1.0000 Epoch 4996/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0200 - acc: 0.9833 - val_loss: 0.0063 - val_acc: 1.0000 Epoch 4997/5000 4/4 [==============================] - 0s 8ms/step - loss: 0.0204 - acc: 0.9750 - val_loss: 0.0063 - val_acc: 1.0000 Epoch 4998/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0204 - acc: 0.9833 - val_loss: 0.0064 - val_acc: 1.0000 Epoch 4999/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0201 - acc: 0.9833 - val_loss: 0.0066 - val_acc: 1.0000 Epoch 5000/5000 4/4 [==============================] - 0s 7ms/step - loss: 0.0201 - acc: 0.9833 - val_loss: 0.0064 - val_acc: 1.0000
In [8]:
# 把epoch当横左边，把loss当纵坐标
plt.plot(history.epoch,history.history.get('loss'))
plt.show()
In [9]:
plt.plot(history.epoch,history.history.get('val_acc'))
plt.title("test data acc")
plt.show()
5、检验模型
In [10]:
pridict_y=model.predict(test_x)
print(pridict_y)
print(test_y)
[[ 1.0565193e+00 -2.6889920e-02 -1.4720589e-02] [-6.2308311e-03 1.0293320e+00 -1.9791573e-02] [-6.1598510e-02 1.1274689e+00 -6.0042202e-02] [-5.4454565e-02 9.0536302e-01 1.5964857e-01] [ 3.8922548e-02 7.4594215e-02 9.0920663e-01] [-4.3319345e-02 1.0453464e+00 4.2278469e-03] [ 6.5247595e-02 1.0222425e+00 -8.0829531e-02] [-4.1642666e-02 9.7465175e-01 7.5081468e-02] [-3.0709654e-02 1.0207720e+00 9.9766850e-03] [ 9.2975128e-01 6.5912709e-02 4.1824669e-02] [ 9.1035056e-01 9.0161189e-02 3.5780758e-02] [ 9.9654549e-01 -2.0576075e-02 4.8277467e-02] [ 9.5139623e-01 8.6834028e-02 -2.6680857e-02] [ 9.1800302e-01 1.1167152e-01 -4.8157275e-03] [ 6.9141388e-05 8.7840325e-01 1.3010237e-01] [-2.4678916e-02 9.6513385e-01 7.0031434e-02] [ 9.7465920e-01 7.3628679e-02 -7.7703744e-02] [-1.3774335e-02 8.9502794e-01 1.2773290e-01] [ 8.9697993e-01 1.2705387e-01 2.8292835e-03] [ 5.7709873e-02 9.7175795e-01 -2.4247855e-02] [-3.4542620e-02 8.8021189e-01 1.6434848e-01] [ 2.1483630e-02 2.5068894e-02 9.7806406e-01] [ 3.8601160e-03 3.1894189e-01 7.0536506e-01] [ 9.4212979e-01 8.0834672e-02 1.5474647e-02] [ 2.0926297e-03 6.7882553e-02 9.5009637e-01] [ 1.0249877e+00 1.8325776e-02 -5.5281550e-02] [ 9.9468207e-01 2.7860314e-02 2.5103986e-03] [ 1.6328961e-02 2.0112351e-02 1.0003220e+00] [-1.8210649e-02 9.1097170e-01 1.1046210e-01]] tf.Tensor( [[1. 0. 0.] [0. 1. 0.] [0. 1. 0.] [0. 1. 0.] [0. 0. 1.] [0. 1. 0.] [0. 1. 0.] [0. 1. 0.] [0. 1. 0.] [1. 0. 0.] [1. 0. 0.] [1. 0. 0.] [1. 0. 0.] [1. 0. 0.] [0. 1. 0.] [0. 1. 0.] [1. 0. 0.] [0. 1. 0.] [1. 0. 0.] [0. 1. 0.] [0. 1. 0.] [0. 0. 1.] [0. 0. 1.] [1. 0. 0.] [0. 0. 1.] [1. 0. 0.] [1. 0. 0.] [0. 0. 1.] [0. 1. 0.]], shape=(29, 3), dtype=float32)
In [11]:
# 在pridict_y中找最大值的索引，横向
pridict_y = tf.argmax(pridict_y, axis=1)
print(pridict_y)
#
test_y = tf.argmax(test_y, axis=1)
print(test_y)
tf.Tensor([0 1 1 1 2 1 1 1 1 0 0 0 0 0 1 1 0 1 0 1 1 2 2 0 2 0 0 2 1], shape=(29,), dtype=int64) tf.Tensor([0 1 1 1 2 1 1 1 1 0 0 0 0 0 1 1 0 1 0 1 1 2 2 0 2 0 0 2 1], shape=(29,), dtype=int64)
In [ ]:
我的旨在学过的东西不再忘记（主要使用艾宾浩斯遗忘曲线算法及其它智能学习复习算法）的偏公益性质的完全免费的编程视频学习网站：
【读书编程笔记】fanrenyi.com；有各种前端、后端、算法、大数据、人工智能等课程。
版权申明：欢迎转载，但请注明出处
一些博文中有一些参考内容因时间久远找不到来源了没有注明，如果侵权请联系我删除。
2023-12-12，找女朋友，非诚勿扰，vx：fan404006308
AI交流资料群：753014672
