machine-learning-digit-recognizer

MNIST Digit Recognizer with Machine Learning (CNN, MLP, KNN, SVM, Decision Tree, Random Forest)

https://www.kaggle.com/competitions/digit-recognizer

Keşifsel Veri Analizi (EDA) ve Derin Öğrenme Modeli

Proje kapsamında MNIST veri seti için keşifsel veri analizi aşağıdaki gibi yapılmıştır. Veri işleyiş yöntemleri burada görülebilir. Ayrıca scikit-learn kütüphanesindeki MLP (Multi-layer Perceptron) Classifier modeli ile veriler eğitilmiş ve %99.7 tahmin doğruluğuna ulaşılmıştır.

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns

test = pd.read_csv('test.csv')
train = pd.read_csv('train.csv')

train.head()

	label	pixel0	pixel1	pixel2	pixel3	pixel4	pixel5	pixel6	pixel7	pixel8	...	pixel774	pixel775	pixel776	pixel777	pixel778	pixel779	pixel780	pixel781	pixel782	pixel783
0	1	0	0	0	0	0	0	0	0	0	...	0	0	0	0	0	0	0	0	0	0
1	0	0	0	0	0	0	0	0	0	0	...	0	0	0	0	0	0	0	0	0	0
2	1	0	0	0	0	0	0	0	0	0	...	0	0	0	0	0	0	0	0	0	0
3	4	0	0	0	0	0	0	0	0	0	...	0	0	0	0	0	0	0	0	0	0
4	0	0	0	0	0	0	0	0	0	0	...	0	0	0	0	0	0	0	0	0	0

5 rows × 785 columns

train.describe()

	label	pixel0	pixel1	pixel2	pixel3	pixel4	pixel5	pixel6	pixel7	pixel8	...	pixel774	pixel775	pixel776	pixel777	pixel778	pixel779	pixel780	pixel781	pixel782	pixel783
count	42000.000000	42000.0	42000.0	42000.0	42000.0	42000.0	42000.0	42000.0	42000.0	42000.0	...	42000.000000	42000.000000	42000.000000	42000.00000	42000.000000	42000.000000	42000.0	42000.0	42000.0	42000.0
mean	4.456643	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	0.219286	0.117095	0.059024	0.02019	0.017238	0.002857	0.0	0.0	0.0	0.0
std	2.887730	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	6.312890	4.633819	3.274488	1.75987	1.894498	0.414264	0.0	0.0	0.0	0.0
min	0.000000	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	0.000000	0.000000	0.000000	0.00000	0.000000	0.000000	0.0	0.0	0.0	0.0
25%	2.000000	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	0.000000	0.000000	0.000000	0.00000	0.000000	0.000000	0.0	0.0	0.0	0.0
50%	4.000000	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	0.000000	0.000000	0.000000	0.00000	0.000000	0.000000	0.0	0.0	0.0	0.0
75%	7.000000	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	0.000000	0.000000	0.000000	0.00000	0.000000	0.000000	0.0	0.0	0.0	0.0
max	9.000000	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	...	254.000000	254.000000	253.000000	253.00000	254.000000	62.000000	0.0	0.0	0.0	0.0

8 rows × 785 columns

train.shape

(42000, 785)

# veri temizleme
train.isna().any().any()

# veri ve etiketlerini alma
train_data = train.loc[:, "pixel0":]
train_label = train.loc[:, "label"]

# verileri normalize etme
train_data = train_data/255.0
test = test/255.0

Veri setindeki örnek bir rakam veriler normalize edildikten sonra aşağıdaki gibi elde edilmiştir.

digit_array = train.loc[9, "pixel0":]
arr = np.array(digit_array) 

image_array = np.reshape(arr, (28,28))

digit_img = plt.imshow(image_array, cmap=plt.cm.binary)
plt.colorbar(digit_img)
print("Image Label: {}".format(train.loc[9, "label"]))

Image Label: 3

Veri standardizasyonu;

from sklearn.preprocessing import StandardScaler

standardized_scalar = StandardScaler()
standardized_data = standardized_scalar.fit_transform(train_data)
standardized_data.shape

(42000, 784)

Kovaryans matrisi;

cov_matrix = np.matmul(standardized_data.T, standardized_data)
cov_matrix.shape

(784, 784)

Özdeğer ve Özvektörler;

from scipy.linalg import eigh

lambdas, vectors = eigh(cov_matrix, eigvals=(782, 783))
vectors.shape

(784, 2)

vectors = vectors.T
vectors.shape

(2, 784)

Birim vektörler;

new_coordinates = np.matmul(vectors, standardized_data.T)
print(new_coordinates.shape)
new_coordinates = np.vstack((new_coordinates, train_label)).T

(2, 42000)

df_new = pd.DataFrame(new_coordinates, columns=["f1", "f2", "labels"])
df_new.head()

	f1	f2	labels
0	-5.226445	-5.140478	1.0
1	6.032996	19.292332	0.0
2	-1.705813	-7.644503	1.0
3	5.836139	-0.474207	4.0
4	6.024818	26.559574	0.0

Veri dağılımı;

sns.FacetGrid(df_new, hue="labels", height=6).map(plt.scatter, "f1", "f2").add_legend()
plt.show()

Veri histogramı;

Veri setinde diğerlerinden daha fazla sayıda 1 olduğu görülüyor.

sns.countplot(train.label)
print(list(train.label.value_counts().sort_index()))

[4132, 4684, 4177, 4351, 4072, 3795, 4137, 4401, 4063, 4188]


/opt/homebrew/Caskroom/miniforge/base/envs/tf_m1/lib/python3.8/site-packages/seaborn/_decorators.py:36: FutureWarning: Pass the following variable as a keyword arg: x. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.
  warnings.warn(

Multi-Layer Perceptron Classifier Derin Öğrenme Modeli ile Veri Eğitimi ve Sonuçları

Görüleceği üzere tahmin oranı %99.7 olarak çıkıyor.

x = train.drop(['label'],axis=1).values
y = train['label'].values.flatten()

from sklearn.model_selection import train_test_split
x_train,x_dev, y_train,  y_dev = train_test_split(x,y,random_state=42)

from sklearn.neural_network import MLPClassifier
from sklearn.metrics import classification_report
import scipy

clf = MLPClassifier(solver='lbfgs',hidden_layer_sizes=(1000,),  random_state=1 ) 
best_model = clf.fit(x/255.0, y)

train['label_predicted'] =  best_model.predict(x/255.0)
train['valid_predicted'] = train.apply(lambda row: row['label_predicted']==row['label'], axis=1)

print('Training Accuracy: {:3.2f} %'.format(best_model.score(x,y)*100))
y_test_pred = best_model.predict(x)

Training Accuracy: 99.70 %

from sklearn.metrics import confusion_matrix
mat = confusion_matrix(y, y_test_pred)
fig = plt.figure(figsize=(8,8))

sns.set(font_scale=2)
sns.heatmap(mat.T, annot=True, cbar = False, annot_kws={'size': 16}, fmt='.0f')
    
plt.title("Confusion Matrix")
plt.xlabel('Predicted Values')
plt.ylabel('True Values')
plt.show()

from sklearn.metrics import accuracy_score
print("Test Accuracy: " + accuracy_score(y, y_test_pred).__str__())

Test Accuracy: 0.9970238095238095

Yukarıda görülen confusion matriste değerler ne kadar orta çaprazda toplanırsa o kadar doğru tahmin yaptığımızı gösteriyor ve bakıldığında, köşelerin 0 ya da 0'a çok yakın değerler olduğu ve bu da uygulamış olduğumuz MLP Classifier algoritmasının verilerimiz için gayet iyi çalıştığını gösteriyor.

MNIST KNN

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.model_selection import train_test_split

test = pd.read_csv('test.csv')
train = pd.read_csv('train.csv')

x_Train=train.drop(labels="label",axis=1)
y_Train=train["label"]

x_Train=x_Train/255
x_Test=test/255
x_Train=x_Train.values.reshape(-1,28,28,1)
x_Test=x_Test.values.reshape(-1,28,28,1)
print("x train shape",x_Train.shape)
print("x test shape",x_Test.shape)

x train shape (42000, 28, 28, 1)
x test shape (28000, 28, 28, 1)

X_train, X_val, ml_train, ml_val = train_test_split(x_Train, y_Train, test_size = 0.1, random_state=2)

from tensorflow.keras.utils import to_categorical
Y_train=to_categorical(y_Train,num_classes=10)

X_train, X_val, Y_train, Y_val = train_test_split(x_Train, Y_train, test_size = 0.1, random_state=2)
print("x_train shape",X_train.shape)
print("x_val shape",X_val.shape)
print("y_train shape",Y_train.shape)
print("y_val shape",Y_val.shape)

x_train shape (37800, 28, 28, 1)
x_val shape (4200, 28, 28, 1)
y_train shape (37800, 10)
y_val shape (4200, 10)

from sklearn.neighbors import KNeighborsClassifier
knn = KNeighborsClassifier(n_neighbors=3)
knn.fit(train_x, ml_train)
train_accuracy = knn.score(train_x, ml_train)
test_accuracy = knn.score(test_x, ml_val) 
print(train_accuracy)
print(test_accuracy)
y_pred = knn.predict(test_x)

0.9902380952380953
0.9826190476190476

from sklearn.metrics import confusion_matrix
mat = confusion_matrix(ml_val,y_pred)
fig = plt.figure(figsize=(8,8))

sns.set(font_scale=2)
sns.heatmap(mat.T, annot=True, cbar = False, annot_kws={'size': 16}, fmt='.0f')
    
plt.title("Confusion Matrix")
plt.xlabel('Predicted Values')
plt.ylabel('True Values')
plt.show()

from sklearn.metrics import accuracy_score
print("Accuracy: " + accuracy_score(ml_val,y_pred).__str__())

Accuracy: 0.929047619047619

MNIST Decision Tree

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.model_selection import train_test_split

test = pd.read_csv('test.csv')
train = pd.read_csv('train.csv')

x_Train=train.drop(labels="label",axis=1)
y_Train=train["label"]

0    1
1    0
2    1
3    4
4    0
5    0
6    7
7    3
8    5
9    3
Name: label, dtype: int64

x_Train=x_Train/255
x_Test=test/255
x_Train=x_Train.values.reshape(-1,28,28,1)
x_Test=x_Test.values.reshape(-1,28,28,1)
print("x train shape",x_Train.shape)
print("x test shape",x_Test.shape)

x train shape (42000, 28, 28, 1)
x test shape (28000, 28, 28, 1)

X_train, X_val, ml_train, ml_val = train_test_split(x_Train, y_Train, test_size = 0.1, random_state=2)

from tensorflow.keras.utils import to_categorical
Y_train=to_categorical(y_Train,num_classes=10)

X_train, X_val, Y_train, Y_val = train_test_split(x_Train, Y_train, test_size = 0.1, random_state=2)
print("x_train shape",X_train.shape)
print("x_val shape",X_val.shape)
print("y_train shape",Y_train.shape)
print("y_val shape",Y_val.shape)

x_train shape (37800, 28, 28, 1)
x_val shape (4200, 28, 28, 1)
y_train shape (37800, 10)
y_val shape (4200, 10)

from sklearn import tree
clf = tree.DecisionTreeClassifier()
clf = clf.fit(train_x, ml_train)
y_pred = clf.predict(test_x)

from sklearn.metrics import confusion_matrix
mat = confusion_matrix(ml_val,y_pred)
fig = plt.figure(figsize=(8,8))

sns.set(font_scale=2)
sns.heatmap(mat.T, annot=True, cbar = False, annot_kws={'size': 16}, fmt='.0f')
    
plt.title("Confusion Matrix")
plt.xlabel('Predicted Values')
plt.ylabel('True Values')
plt.show()

from sklearn.metrics import accuracy_score
print("Accuracy: " + accuracy_score(ml_val,y_pred).__str__())

Accuracy: 0.9302380952380952

MNIST Random Forest

import numpy as np
import pandas as pd

import matplotlib.pyplot as plt
import seaborn as sns

from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier

from sklearn.metrics import accuracy_score

from sklearn.metrics import confusion_matrix

train = pd.read_csv('train.csv')
test = pd.read_csv('test.csv')

digits = train.drop(['label'], 1).values
digits = digits / 255.
label = train['label'].values

X = digits
y = label

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1, random_state=42)

/var/folders/kr/f0qmqqxs5p19_gxt95ch4jn00000gn/T/ipykernel_53665/1816080881.py:1: FutureWarning: In a future version of pandas all arguments of DataFrame.drop except for the argument 'labels' will be keyword-only
  digits = train.drop(['label'], 1).values

model = RandomForestClassifier(n_estimators=100, max_depth=10, random_state=0)
model.fit(X_train, y_train)

RandomForestClassifier(max_depth=10, random_state=0)

y_pred = model.predict(X_test)
rf_acc = accuracy_score(y_test, y_pred)

print("Model accuracy: {0:.3f}%".format(rf_acc * 100))

Model accuracy: 94.238%

np.unique(y_test, return_counts=True)

(array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9]),
 array([408, 471, 420, 506, 397, 339, 402, 438, 403, 416]))

mat = confusion_matrix(y_test, y_pred)
fig = plt.figure(figsize=(8,8))

sns.set(font_scale=2)
sns.heatmap(mat.T, annot=True, cbar = False, annot_kws={'size': 16}, fmt='.0f')
    
plt.title("Confusion Matrix")
plt.xlabel('Predicted Values')
plt.ylabel('True Values')
plt.show()

accuracy_score(y_test, y_pred)

0.8192857142857143

MNIST SVM

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.model_selection import train_test_split

test = pd.read_csv('test.csv')
train = pd.read_csv('train.csv')

x_Train=train.drop(labels="label",axis=1)
y_Train=train["label"]

x_Train=x_Train/255
x_Test=test/255
x_Train=x_Train.values.reshape(-1,28,28,1)
x_Test=x_Test.values.reshape(-1,28,28,1)
print("x train shape",x_Train.shape)
print("x test shape",x_Test.shape)

x train shape (42000, 28, 28, 1)
x test shape (28000, 28, 28, 1)

X_train, X_val, ml_train, ml_val = train_test_split(x_Train, y_Train, test_size = 0.1, random_state=2)

from tensorflow.keras.utils import to_categorical
Y_train=to_categorical(y_Train,num_classes=10)

X_train, X_val, Y_train, Y_val = train_test_split(x_Train, Y_train, test_size = 0.1, random_state=2)
print("x_train shape",X_train.shape)
print("x_val shape",X_val.shape)
print("y_train shape",Y_train.shape)
print("y_val shape",Y_val.shape)

x_train shape (37800, 28, 28, 1)
x_val shape (4200, 28, 28, 1)
y_train shape (37800, 10)
y_val shape (4200, 10)

from sklearn import svm
clf = svm.SVC(kernel='linear')
clf.fit(train_x, ml_train)

y_pred = clf.predict(test_x)

from sklearn.metrics import confusion_matrix
mat = confusion_matrix(ml_val,y_pred)
fig = plt.figure(figsize=(8,8))

sns.set(font_scale=2)
sns.heatmap(mat.T, annot=True, cbar = False, annot_kws={'size': 16}, fmt='.0f')
    
plt.title("Confusion Matrix")
plt.xlabel('Predicted Values')
plt.ylabel('True Values')
plt.show()

from sklearn.metrics import accuracy_score
print("Accuracy: " + accuracy_score(ml_val,y_pred).__str__())

Accuracy: 0.9785714285714285





0.9785714285714285

MNIST CNN

import pandas as pd
import numpy as np

import matplotlib.pyplot as plt
import seaborn as sns

from sklearn.model_selection import train_test_split
from sklearn.metrics import confusion_matrix

import tensorflow.keras
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Conv2D, Lambda, MaxPooling2D
from tensorflow.keras.layers import Dense, Dropout, Flatten
from tensorflow.keras.layers import BatchNormalization

from tensorflow.keras.preprocessing.image import ImageDataGenerator

from tensorflow.keras.utils import to_categorical

train = pd.read_csv('train.csv')
test = pd.read_csv('test.csv')

print(f"Training veri boyutu: {train.shape}\nTest veri boyutu: {test.shape}")

Training veri boyutu: (42000, 785)
Test veri boyutu: (28000, 784)

X = train.drop(['label'], 1).values
y = train['label'].values

/var/folders/kr/f0qmqqxs5p19_gxt95ch4jn00000gn/T/ipykernel_53609/2430759019.py:1: FutureWarning: In a future version of pandas all arguments of DataFrame.drop except for the argument 'labels' will be keyword-only
  X = train.drop(['label'], 1).values

X = X / 255.0
X = X.reshape(-1,28,28,1)

y = to_categorical(y)

print(f"Etiket boyutu: {y.shape}")

Etiket boyutu: (42000, 10)

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1, random_state=0)

X_train.shape, X_test.shape, y_train.shape, y_test.shape

((37800, 28, 28, 1), (4200, 28, 28, 1), (37800, 10), (4200, 10))

X_train__ = X_train.reshape(X_train.shape[0], 28, 28)

fig, axis = plt.subplots(1, 4, figsize=(20, 10))
for i, ax in enumerate(axis.flat):
    ax.imshow(X_train__[i], cmap='binary')
    digit = y_train[i].argmax()
    ax.set(title = f"Real Number: {digit}")

mean = np.mean(X_train)
std = np.std(X_train)

def standardize(x):
    return (x-mean)/std

model=Sequential()
  
model.add(Conv2D(filters=64, kernel_size = (3,3), activation="relu", input_shape=(28,28,1)))
model.add(Conv2D(filters=64, kernel_size = (3,3), activation="relu"))
model.add(MaxPooling2D(pool_size=(2,2)))
model.add(BatchNormalization())

model.add(Conv2D(filters=128, kernel_size = (3,3), activation="relu"))
model.add(Conv2D(filters=128, kernel_size = (3,3), activation="relu"))
model.add(MaxPooling2D(pool_size=(2,2)))
model.add(BatchNormalization())    

model.add(Conv2D(filters=256, kernel_size = (3,3), activation="relu"))
model.add(MaxPooling2D(pool_size=(2,2)))
model.add(BatchNormalization())
    
model.add(Flatten())
model.add(Dense(512,activation="relu"))
    
model.add(Dense(10,activation="softmax"))
    
model.compile(loss="categorical_crossentropy", optimizer="adam", metrics=["accuracy"])

model.summary()

Model: "sequential"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
conv2d (Conv2D)              (None, 26, 26, 64)        640       
_________________________________________________________________
conv2d_1 (Conv2D)            (None, 24, 24, 64)        36928     
_________________________________________________________________
max_pooling2d (MaxPooling2D) (None, 12, 12, 64)        0         
_________________________________________________________________
batch_normalization (BatchNo (None, 12, 12, 64)        256       
_________________________________________________________________
conv2d_2 (Conv2D)            (None, 10, 10, 128)       73856     
_________________________________________________________________
conv2d_3 (Conv2D)            (None, 8, 8, 128)         147584    
_________________________________________________________________
max_pooling2d_1 (MaxPooling2 (None, 4, 4, 128)         0         
_________________________________________________________________
batch_normalization_1 (Batch (None, 4, 4, 128)         512       
_________________________________________________________________
conv2d_4 (Conv2D)            (None, 2, 2, 256)         295168    
_________________________________________________________________
max_pooling2d_2 (MaxPooling2 (None, 1, 1, 256)         0         
_________________________________________________________________
batch_normalization_2 (Batch (None, 1, 1, 256)         1024      
_________________________________________________________________
flatten (Flatten)            (None, 256)               0         
_________________________________________________________________
dense (Dense)                (None, 512)               131584    
_________________________________________________________________
dense_1 (Dense)              (None, 10)                5130      
=================================================================
Total params: 692,682
Trainable params: 691,786
Non-trainable params: 896
_________________________________________________________________

datagen = ImageDataGenerator(
        featurewise_center=False,
        samplewise_center=False,
        featurewise_std_normalization=False,
        samplewise_std_normalization=False,
        zca_whitening=False,
        rotation_range=15,
        zoom_range = 0.01,
        width_shift_range=0.1,
        height_shift_range=0.1,
        horizontal_flip=False,
        vertical_flip=False)

train_gen = datagen.flow(X_train, y_train, batch_size=128)
test_gen = datagen.flow(X_test, y_test, batch_size=128)

epochs = 100
batch_size = 128
train_steps = X_train.shape[0] // batch_size
valid_steps = X_test.shape[0] // batch_size

es = tensorflow.keras.callbacks.EarlyStopping(
        monitor="val_acc",
        patience=10,
        verbose=1,
        mode="max",
        restore_best_weights=True, # 
     )

rp = tensorflow.keras.callbacks.ReduceLROnPlateau(
        monitor="val_acc",
        factor=0.2,
        patience=3,
        verbose=1,
        mode="max",
        min_lr=0.00001,
     )

history = model.fit_generator(train_gen, 
                              epochs = epochs, 
                              steps_per_epoch = train_steps,
                              validation_data = test_gen,
                              validation_steps = valid_steps, 
                              callbacks=[es, rp])

Epoch 1/100
WARNING:tensorflow:AutoGraph could not transform <function Model.make_train_function.<locals>.train_function at 0x2a1f99160> and will run it as-is.
Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.
Cause: unsupported operand type(s) for -: 'NoneType' and 'int'
To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert
WARNING: AutoGraph could not transform <function Model.make_train_function.<locals>.train_function at 0x2a1f99160> and will run it as-is.
Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.
Cause: unsupported operand type(s) for -: 'NoneType' and 'int'
To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert


/opt/homebrew/Caskroom/miniforge/base/envs/tf_m1/lib/python3.8/site-packages/tensorflow/python/keras/engine/training.py:1844: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.
  warnings.warn('`Model.fit_generator` is deprecated and '
2022-08-10 16:24:00.221521: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:116] None of the MLIR optimization passes are enabled (registered 2)
2022-08-10 16:24:00.221789: W tensorflow/core/platform/profile_utils/cpu_utils.cc:126] Failed to get CPU frequency: 0 Hz


295/295 [==============================] - ETA: 0s - loss: 0.3625 - accuracy: 0.8870WARNING:tensorflow:AutoGraph could not transform <function Model.make_test_function.<locals>.test_function at 0x2aa3e3040> and will run it as-is.
Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.
Cause: unsupported operand type(s) for -: 'NoneType' and 'int'
To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert
WARNING: AutoGraph could not transform <function Model.make_test_function.<locals>.test_function at 0x2aa3e3040> and will run it as-is.
Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.
Cause: unsupported operand type(s) for -: 'NoneType' and 'int'
To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert
295/295 [==============================] - 24s 82ms/step - loss: 0.3618 - accuracy: 0.8872 - val_loss: 1.4837 - val_accuracy: 0.4089
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 2/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0659 - accuracy: 0.9789 - val_loss: 0.0591 - val_accuracy: 0.9792
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 3/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0499 - accuracy: 0.9831 - val_loss: 0.0791 - val_accuracy: 0.9766
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 4/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0493 - accuracy: 0.9854 - val_loss: 0.0422 - val_accuracy: 0.9854
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 5/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0336 - accuracy: 0.9901 - val_loss: 0.0563 - val_accuracy: 0.9832
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 6/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0322 - accuracy: 0.9904 - val_loss: 0.0669 - val_accuracy: 0.9800
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 7/100
295/295 [==============================] - 25s 83ms/step - loss: 0.0332 - accuracy: 0.9891 - val_loss: 0.0759 - val_accuracy: 0.9807
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 8/100
295/295 [==============================] - 25s 84ms/step - loss: 0.0304 - accuracy: 0.9902 - val_loss: 0.0517 - val_accuracy: 0.9856
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 9/100
295/295 [==============================] - 25s 84ms/step - loss: 0.0266 - accuracy: 0.9915 - val_loss: 0.0383 - val_accuracy: 0.9883
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 10/100
295/295 [==============================] - 24s 82ms/step - loss: 0.0259 - accuracy: 0.9926 - val_loss: 0.0656 - val_accuracy: 0.9817
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 11/100
295/295 [==============================] - 25s 83ms/step - loss: 0.0275 - accuracy: 0.9908 - val_loss: 0.0429 - val_accuracy: 0.9880
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 12/100
295/295 [==============================] - 25s 83ms/step - loss: 0.0221 - accuracy: 0.9934 - val_loss: 0.0514 - val_accuracy: 0.9871
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 13/100
295/295 [==============================] - 25s 85ms/step - loss: 0.0220 - accuracy: 0.9935 - val_loss: 0.0483 - val_accuracy: 0.9846
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 14/100
295/295 [==============================] - 25s 84ms/step - loss: 0.0224 - accuracy: 0.9931 - val_loss: 0.0310 - val_accuracy: 0.9895
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 15/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0226 - accuracy: 0.9934 - val_loss: 0.0408 - val_accuracy: 0.9890
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 16/100
295/295 [==============================] - 24s 82ms/step - loss: 0.0159 - accuracy: 0.9949 - val_loss: 0.0367 - val_accuracy: 0.9893
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 17/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0194 - accuracy: 0.9940 - val_loss: 0.0277 - val_accuracy: 0.9929
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 18/100
295/295 [==============================] - 24s 82ms/step - loss: 0.0238 - accuracy: 0.9928 - val_loss: 0.0285 - val_accuracy: 0.9907
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 19/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0147 - accuracy: 0.9953 - val_loss: 0.0347 - val_accuracy: 0.9888
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 20/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0171 - accuracy: 0.9944 - val_loss: 0.0285 - val_accuracy: 0.9917
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 21/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0189 - accuracy: 0.9943 - val_loss: 0.0292 - val_accuracy: 0.9912
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 22/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0147 - accuracy: 0.9952 - val_loss: 0.0357 - val_accuracy: 0.9893
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 23/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0157 - accuracy: 0.9948 - val_loss: 0.0247 - val_accuracy: 0.9919
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 24/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0152 - accuracy: 0.9955 - val_loss: 0.0257 - val_accuracy: 0.9927
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 25/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0152 - accuracy: 0.9953 - val_loss: 0.0506 - val_accuracy: 0.9875
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 26/100
295/295 [==============================] - 24s 82ms/step - loss: 0.0145 - accuracy: 0.9953 - val_loss: 0.0446 - val_accuracy: 0.9878
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 27/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0182 - accuracy: 0.9945 - val_loss: 0.0275 - val_accuracy: 0.9912
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 28/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0119 - accuracy: 0.9962 - val_loss: 0.0293 - val_accuracy: 0.9919
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 29/100
295/295 [==============================] - 24s 82ms/step - loss: 0.0118 - accuracy: 0.9966 - val_loss: 0.0270 - val_accuracy: 0.9917
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 30/100
295/295 [==============================] - 24s 82ms/step - loss: 0.0124 - accuracy: 0.9962 - val_loss: 0.0286 - val_accuracy: 0.9919
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 31/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0127 - accuracy: 0.9956 - val_loss: 0.0342 - val_accuracy: 0.9900
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 32/100
295/295 [==============================] - 24s 82ms/step - loss: 0.0104 - accuracy: 0.9963 - val_loss: 0.0367 - val_accuracy: 0.9907
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 33/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0115 - accuracy: 0.9963 - val_loss: 0.0382 - val_accuracy: 0.9905
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 34/100
295/295 [==============================] - 24s 82ms/step - loss: 0.0102 - accuracy: 0.9967 - val_loss: 0.0534 - val_accuracy: 0.9861
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 35/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0115 - accuracy: 0.9961 - val_loss: 0.0346 - val_accuracy: 0.9917
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 36/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0122 - accuracy: 0.9962 - val_loss: 0.0490 - val_accuracy: 0.9868
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 37/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0124 - accuracy: 0.9963 - val_loss: 0.0200 - val_accuracy: 0.9937
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 38/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0088 - accuracy: 0.9973 - val_loss: 0.0296 - val_accuracy: 0.9912
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 39/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0100 - accuracy: 0.9971 - val_loss: 0.0400 - val_accuracy: 0.9912
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 40/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0068 - accuracy: 0.9974 - val_loss: 0.0351 - val_accuracy: 0.9917
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 41/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0097 - accuracy: 0.9967 - val_loss: 0.0341 - val_accuracy: 0.9932
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 42/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0114 - accuracy: 0.9965 - val_loss: 0.0252 - val_accuracy: 0.9917
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 43/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0092 - accuracy: 0.9976 - val_loss: 0.0329 - val_accuracy: 0.9902
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 44/100
295/295 [==============================] - 24s 81ms/step - loss: 0.0069 - accuracy: 0.9978 - val_loss: 0.0253 - val_accuracy: 0.9929
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 45/100
295/295 [==============================] - 25s 84ms/step - loss: 0.0072 - accuracy: 0.9975 - val_loss: 0.0330 - val_accuracy: 0.9910
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 46/100
295/295 [==============================] - 25s 83ms/step - loss: 0.0092 - accuracy: 0.9971 - val_loss: 0.0593 - val_accuracy: 0.9868
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 47/100
295/295 [==============================] - 24s 83ms/step - loss: 0.0071 - accuracy: 0.9973 - val_loss: 0.0309 - val_accuracy: 0.9915
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 48/100
295/295 [==============================] - 25s 84ms/step - loss: 0.0066 - accuracy: 0.9981 - val_loss: 0.0276 - val_accuracy: 0.9927
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 49/100
295/295 [==============================] - 25s 84ms/step - loss: 0.0085 - accuracy: 0.9972 - val_loss: 0.0219 - val_accuracy: 0.9924
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 50/100
295/295 [==============================] - 25s 86ms/step - loss: 0.0078 - accuracy: 0.9971 - val_loss: 0.0297 - val_accuracy: 0.9934
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 51/100
295/295 [==============================] - 25s 86ms/step - loss: 0.0098 - accuracy: 0.9964 - val_loss: 0.0366 - val_accuracy: 0.9922
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 52/100
295/295 [==============================] - 26s 88ms/step - loss: 0.0087 - accuracy: 0.9973 - val_loss: 0.0385 - val_accuracy: 0.9912
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 53/100
295/295 [==============================] - 26s 87ms/step - loss: 0.0053 - accuracy: 0.9980 - val_loss: 0.0238 - val_accuracy: 0.9922
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 54/100
295/295 [==============================] - 25s 86ms/step - loss: 0.0056 - accuracy: 0.9984 - val_loss: 0.0325 - val_accuracy: 0.9922
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 55/100
295/295 [==============================] - 26s 87ms/step - loss: 0.0092 - accuracy: 0.9969 - val_loss: 0.0241 - val_accuracy: 0.9944
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 56/100
295/295 [==============================] - 26s 87ms/step - loss: 0.0079 - accuracy: 0.9975 - val_loss: 0.0342 - val_accuracy: 0.9917
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 57/100
295/295 [==============================] - 26s 88ms/step - loss: 0.0091 - accuracy: 0.9971 - val_loss: 0.0233 - val_accuracy: 0.9937
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 58/100
295/295 [==============================] - 26s 89ms/step - loss: 0.0046 - accuracy: 0.9987 - val_loss: 0.0304 - val_accuracy: 0.9944
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 59/100
295/295 [==============================] - 26s 87ms/step - loss: 0.0064 - accuracy: 0.9977 - val_loss: 0.0277 - val_accuracy: 0.9939
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 60/100
295/295 [==============================] - 27s 92ms/step - loss: 0.0058 - accuracy: 0.9980 - val_loss: 0.0354 - val_accuracy: 0.9924
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 61/100
295/295 [==============================] - 27s 92ms/step - loss: 0.0080 - accuracy: 0.9969 - val_loss: 0.0198 - val_accuracy: 0.9958
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 62/100
295/295 [==============================] - 27s 91ms/step - loss: 0.0063 - accuracy: 0.9981 - val_loss: 0.0273 - val_accuracy: 0.9939
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 63/100
295/295 [==============================] - 26s 88ms/step - loss: 0.0097 - accuracy: 0.9972 - val_loss: 0.0274 - val_accuracy: 0.9939
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 64/100
295/295 [==============================] - 26s 90ms/step - loss: 0.0080 - accuracy: 0.9970 - val_loss: 0.0342 - val_accuracy: 0.9941
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 65/100
295/295 [==============================] - 28s 93ms/step - loss: 0.0054 - accuracy: 0.9982 - val_loss: 0.0379 - val_accuracy: 0.9897
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 66/100
295/295 [==============================] - 27s 91ms/step - loss: 0.0079 - accuracy: 0.9980 - val_loss: 0.0308 - val_accuracy: 0.9917
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 67/100
295/295 [==============================] - 27s 93ms/step - loss: 0.0049 - accuracy: 0.9983 - val_loss: 0.0285 - val_accuracy: 0.9927
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 68/100
295/295 [==============================] - 28s 94ms/step - loss: 0.0057 - accuracy: 0.9977 - val_loss: 0.0287 - val_accuracy: 0.9927
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 69/100
295/295 [==============================] - 28s 96ms/step - loss: 0.0083 - accuracy: 0.9976 - val_loss: 0.0321 - val_accuracy: 0.9939
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 70/100
295/295 [==============================] - 28s 94ms/step - loss: 0.0059 - accuracy: 0.9978 - val_loss: 0.0317 - val_accuracy: 0.9927
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 71/100
295/295 [==============================] - 28s 96ms/step - loss: 0.0083 - accuracy: 0.9973 - val_loss: 0.0514 - val_accuracy: 0.9922
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 72/100
295/295 [==============================] - 28s 94ms/step - loss: 0.0050 - accuracy: 0.9980 - val_loss: 0.0407 - val_accuracy: 0.9932
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 73/100
295/295 [==============================] - 28s 96ms/step - loss: 0.0069 - accuracy: 0.9981 - val_loss: 0.0244 - val_accuracy: 0.9941
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 74/100
295/295 [==============================] - 29s 98ms/step - loss: 0.0045 - accuracy: 0.9984 - val_loss: 0.0219 - val_accuracy: 0.9937
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 75/100
295/295 [==============================] - 28s 96ms/step - loss: 0.0033 - accuracy: 0.9987 - val_loss: 0.0249 - val_accuracy: 0.9927
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 76/100
295/295 [==============================] - 28s 96ms/step - loss: 0.0075 - accuracy: 0.9979 - val_loss: 0.0296 - val_accuracy: 0.9946
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 77/100
295/295 [==============================] - 29s 99ms/step - loss: 0.0054 - accuracy: 0.9981 - val_loss: 0.0268 - val_accuracy: 0.9937
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 78/100
295/295 [==============================] - 28s 94ms/step - loss: 0.0043 - accuracy: 0.9986 - val_loss: 0.0320 - val_accuracy: 0.9937
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 79/100
295/295 [==============================] - 29s 98ms/step - loss: 0.0064 - accuracy: 0.9981 - val_loss: 0.0223 - val_accuracy: 0.9946
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 80/100
295/295 [==============================] - 28s 95ms/step - loss: 0.0052 - accuracy: 0.9984 - val_loss: 0.0373 - val_accuracy: 0.9929
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 81/100
295/295 [==============================] - 29s 100ms/step - loss: 0.0041 - accuracy: 0.9987 - val_loss: 0.0250 - val_accuracy: 0.9939
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 82/100
295/295 [==============================] - 29s 99ms/step - loss: 0.0046 - accuracy: 0.9985 - val_loss: 0.0377 - val_accuracy: 0.9932
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 83/100
295/295 [==============================] - 29s 98ms/step - loss: 0.0057 - accuracy: 0.9983 - val_loss: 0.0357 - val_accuracy: 0.9929
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 84/100
295/295 [==============================] - 29s 97ms/step - loss: 0.0059 - accuracy: 0.9981 - val_loss: 0.0399 - val_accuracy: 0.9924
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 85/100
295/295 [==============================] - 28s 95ms/step - loss: 0.0036 - accuracy: 0.9990 - val_loss: 0.0383 - val_accuracy: 0.9927
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 86/100
295/295 [==============================] - 28s 96ms/step - loss: 0.0049 - accuracy: 0.9984 - val_loss: 0.0288 - val_accuracy: 0.9937
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 87/100
295/295 [==============================] - 28s 95ms/step - loss: 0.0074 - accuracy: 0.9981 - val_loss: 0.0378 - val_accuracy: 0.9941
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 88/100
295/295 [==============================] - 29s 97ms/step - loss: 0.0033 - accuracy: 0.9989 - val_loss: 0.0479 - val_accuracy: 0.9912
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 89/100
295/295 [==============================] - 28s 95ms/step - loss: 0.0056 - accuracy: 0.9985 - val_loss: 0.0264 - val_accuracy: 0.9922
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 90/100
295/295 [==============================] - 28s 96ms/step - loss: 0.0059 - accuracy: 0.9981 - val_loss: 0.0300 - val_accuracy: 0.9929
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 91/100
295/295 [==============================] - 28s 94ms/step - loss: 0.0042 - accuracy: 0.9985 - val_loss: 0.0342 - val_accuracy: 0.9929
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 92/100
295/295 [==============================] - 28s 94ms/step - loss: 0.0042 - accuracy: 0.9988 - val_loss: 0.0358 - val_accuracy: 0.9932
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 93/100
295/295 [==============================] - 29s 97ms/step - loss: 0.0034 - accuracy: 0.9989 - val_loss: 0.0253 - val_accuracy: 0.9946
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 94/100
295/295 [==============================] - 27s 93ms/step - loss: 0.0044 - accuracy: 0.9987 - val_loss: 0.0315 - val_accuracy: 0.9937
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 95/100
295/295 [==============================] - 27s 92ms/step - loss: 0.0051 - accuracy: 0.9979 - val_loss: 0.0362 - val_accuracy: 0.9946
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 96/100
295/295 [==============================] - 27s 90ms/step - loss: 0.0044 - accuracy: 0.9986 - val_loss: 0.0269 - val_accuracy: 0.9924
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 97/100
295/295 [==============================] - 26s 90ms/step - loss: 0.0047 - accuracy: 0.9984 - val_loss: 0.0269 - val_accuracy: 0.9934
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 98/100
295/295 [==============================] - 26s 89ms/step - loss: 0.0054 - accuracy: 0.9982 - val_loss: 0.0312 - val_accuracy: 0.9946
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 99/100
295/295 [==============================] - 27s 92ms/step - loss: 0.0033 - accuracy: 0.9987 - val_loss: 0.0403 - val_accuracy: 0.9927
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr
Epoch 100/100
295/295 [==============================] - 27s 93ms/step - loss: 0.0044 - accuracy: 0.9981 - val_loss: 0.0351 - val_accuracy: 0.9937
WARNING:tensorflow:Early stopping conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy
WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr

fig, ax = plt.subplots(2,1, figsize=(18, 10))
ax[0].plot(history.history['loss'], color='b', label="Training loss")
ax[0].plot(history.history['val_loss'], color='r', label="validation loss",axes =ax[0])
legend = ax[0].legend(loc='best', shadow=True)

ax[1].plot(history.history['accuracy'], color='b', label="Training accuracy")
ax[1].plot(history.history['val_accuracy'], color='r',label="Validation accuracy")
legend = ax[1].legend(loc='best', shadow=True)

from sklearn.metrics import confusion_matrix

fig = plt.figure(figsize=(8, 8))

y_pred = model.predict(X_test)

Y_pred = np.argmax(y_pred, 1)
Y_test = np.argmax(y_test, 1)

mat = confusion_matrix(Y_test, Y_pred)

sns.set(font_scale=2)
sns.heatmap(mat.T, annot=True, cbar = False, annot_kws={'size': 16}, fmt='.0f')

#sns.heatmap(mat.T, square=True, annot=True, cbar=False, cmap=plt.cm.PuBuGn,fmt="d")
#sns.heatmap(mat.T, square=True, annot=True, cbar=False, cmap=plt.cm.Blues, fmt='.0f')

plt.xlabel('Predicted Values')
plt.ylabel('True Values');
plt.show();

from sklearn.metrics import accuracy_score
print("Test Accuracy: " + accuracy_score(Y_test, Y_pred).__str__())

0.9954761904761905

y_pred = model.predict(X_test)
X_test__ = X_test.reshape(X_test.shape[0], 28, 28)

fig, axis = plt.subplots(4, 4, figsize=(12, 14))
for i, ax in enumerate(axis.flat):
    ax.imshow(X_test__[i], cmap='binary')
    ax.set(title = f"Real Number is {y_test[i].argmax()}\nPredict Number is {y_pred[i].argmax()}");