Joint Spatial–Spectral Attention Network for Hyperspectral Image Classification
論文分析:高光譜圖像分類論文分析(四)
Lei Li, Jihao Yin , Senior Member , IEEE, Xiuping Jia, Senior Member , IEEE,
Sen Li, and Bingnan Han
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
資料集:Indian pines
環境：juputer notebook
tensorflow:2.3
訓練集:測試集:1:9
epoch:400
訓練batch_size=128
"""
作者:zyy
"""
import keras
from keras.layers import Conv2D,LeakyReLU, Conv3D, Flatten, Dense, MaxPooling2D,Reshape, BatchNormalization,Activation,DepthwiseConv2D,GlobalAveragePooling2D,GlobalAveragePooling1D
from keras.layers import Dropout, Input,Lambda
from keras.models import Model
from keras.optimizers import Adam
from keras.callbacks import ModelCheckpoint
from keras.utils import np_utils
from sklearn.decomposition import PCA
from sklearn.model_selection import train_test_split
from sklearn.metrics import confusion_matrix, accuracy_score, classification_report, cohen_kappa_score
import random
from operator import truediv
from plotly.offline import init_notebook_mode
import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt
%matplotlib inline
import scipy.io as sio
import os
import spectral
import glob
from sklearn.preprocessing import MinMaxScaler
def loadData():
data_path=os.path.join(r'D:\Program Files (x86)\Anaconda\jupyter_file_path','data')
data_path
data=sio.loadmat(os.path.join(data_path, 'Indian_pines_corrected.mat'))['indian_pines_corrected']
labels=sio.loadmat(os.path.join(data_path,'Indian_pines_gt.mat'))['indian_pines_gt']
return data,labels
def pca_change(X,num_components):
newX=np.reshape(X,(-1,X.shape[2]))
pca=PCA(n_components=num_components,whiten=True)
newX=pca.fit_transform(newX)
newX=np.reshape(newX,(X.shape[0],X.shape[1],num_components))
return newX
def padwithzeros(X,margin=2):
newX=np.zeros((X.shape[0]+2*margin,X.shape[1]+2*margin,X.shape[2]))
x_offset=margin
y_offset=margin
newX[x_offset:X.shape[0]+x_offset,y_offset:X.shape[1]+y_offset,:]=X
return newX
def creatCube(X,y,windowsize=25,removeZeroLabels=True):
margin=int((windowsize-1)/2) #margin=12
zeroPaddedX=padwithzeros(X,margin=margin)
patchesData=np.zeros((X.shape[0]*X.shape[1],windowsize,windowsize,X.shape[2])) #(145*145,25,25,30)
patchesLabels=np.zeros(X.shape[0]*X.shape[1])
patchIndex=0
for r in range(margin,zeroPaddedX.shape[0]-margin): #(12,145-12)=(12,132)
for c in range(margin,zeroPaddedX.shape[1]-margin): #(12,145-12)=(12,132)
patch=zeroPaddedX[r-margin:r+margin+1,c-margin:c+margin+1]
patchesData[patchIndex,:,:,:]=patch
patchesLabels[patchIndex]=y[r-margin,c-margin]
patchIndex=patchIndex+1
if removeZeroLabels:
patchesData=patchesData[patchesLabels>0,:,:,:]
patchesLabels=patchesLabels[patchesLabels>0]
patchesLabels-=1
return patchesData,patchesLabels
def splitTrainTest(X,y,Ratio,randoms=2019):
X_train,X_test,Y_train,Y_test=train_test_split(X,y,test_size=Ratio,random_state=randoms,
stratify=y)
return X_train,X_test,Y_train,Y_test
x,y=creatCube(x,y,windowsize=27) #打包成Cube
y=np_utils.to_categorical(y)
x.shape,y.shape,x_train.shape,y_train.shape
((10249, 27, 27, 20), (10249, 16), (1024, 27, 27, 20), (1024, 16))
網絡模型建構
def attention1_SA1(input_layer):
conv_11=Conv2D(filters=16,kernel_size=(1,1))(input_layer)
conv_12=Conv2D(filters=16,kernel_size=(1,1))(input_layer)
conv11_RES=Reshape((conv_11.shape[1]*conv_11.shape[2],conv_11.shape[3]))(conv_11)
conv12_RES=Reshape((conv_12.shape[1]*conv_12.shape[2],conv_12.shape[3]))(conv_12)
layer1=tf.matmul(conv11_RES,conv12_RES,transpose_b=True)
output_layer=Activation('softmax')(layer1)
return output_layer
def attention1_SA2(input_layer):
conv_2=Conv2D(filters=16,kernel_size=(3,3),padding='same')(input_layer)
conv_2_RES=Reshape((conv_2.shape[1]*conv_2.shape[2],conv_2.shape[3]))(conv_2)
return conv_2_RES
def attention1_SA3(input_layer):
conv_31=DepthwiseConv2D(kernel_size=(3,3))(input_layer)
conv_32=DepthwiseConv2D(kernel_size=(3,3))(input_layer)
conv32_RES=Reshape((conv_32.shape[1]*conv_32.shape[2],conv_32.shape[3]))(conv_32)
conv31_RES=Reshape((conv_31.shape[1]*conv_31.shape[2],conv_31.shape[3]))(conv_31)
layer1=tf.matmul(conv32_RES,conv31_RES,transpose_a=True)
output_layer=Activation('softmax')(layer1)
return output_layer
def attention_1_SA_FINFAL_1(input_layer):
output_1=attention1_SA1(input_layer)
output_2=attention1_SA2(input_layer)
output_3=attention1_SA3(input_layer)
output=tf.matmul(output_1,output_2)
output=tf.matmul(output,output_3)
output_layer=Reshape((27,27,16))(output)
return output_layer
def attention_moudle1(input_layer):
layer1=Conv2D(filters=16,kernel_size=(3,3),strides=1,padding='same')(input_layer)
layer1_1=attention_1_SA_FINFAL_1(layer1)
layer1_2=Conv2D(filters=16,kernel_size=(3,3),strides=1,padding='same')(layer1)
layer1_3=layer1
layer_output=tf.add(layer1_1,layer1_2)
layer_output=tf.add(layer_output,layer1)
layer_output=BatchNormalization()(layer_output)
layer_output=Activation('relu')(layer_output)
layer_output=MaxPooling2D(pool_size=(2,2),strides=2)(layer_output)
return layer_output
def attention2_SA1(input_layer):
conv_11=Conv2D(filters=32,kernel_size=(1,1))(input_layer)
conv_12=Conv2D(filters=32,kernel_size=(1,1))(input_layer)
conv11_RES=Reshape((conv_11.shape[1]*conv_11.shape[2],conv_11.shape[3]))(conv_11)
conv12_RES=Reshape((conv_12.shape[1]*conv_12.shape[2],conv_12.shape[3]))(conv_12)
layer1=tf.matmul(conv11_RES,conv12_RES,transpose_b=True)
output_layer=Activation('softmax')(layer1)
return output_layer
def attention2_SA2(input_layer):
conv_2=Conv2D(filters=32,kernel_size=(3,3),padding='same')(input_layer)
conv_2_RES=Reshape((conv_2.shape[1]*conv_2.shape[2],conv_2.shape[3]))(conv_2)
return conv_2_RES
def attention2_SA3(input_layer):
conv_31=DepthwiseConv2D(kernel_size=(3,3))(input_layer)
conv_32=DepthwiseConv2D(kernel_size=(3,3))(input_layer)
conv32_RES=Reshape((conv_32.shape[1]*conv_32.shape[2],conv_32.shape[3]))(conv_32)
conv31_RES=Reshape((conv_31.shape[1]*conv_31.shape[2],conv_31.shape[3]))(conv_31)
layer1=tf.matmul(conv32_RES,conv31_RES,transpose_a=True)
output_layer=Activation('softmax')(layer1)
return output_layer
def attention_2_SA_FINFAL_2(input_layer):
output_1=attention2_SA1(input_layer)
output_2=attention2_SA2(input_layer)
output_3=attention2_SA3(input_layer)
output=tf.matmul(output_1,output_2)
output=tf.matmul(output,output_3)
output_layer=Reshape((13,13,32))(output)
return output_layer
def attention_moudle2(input_layer):
layer1=Conv2D(filters=32,kernel_size=(3,3),strides=1,padding='same')(input_layer)
layer1_1=attention_2_SA_FINFAL_2(layer1)
layer1_2=Conv2D(filters=32,kernel_size=(3,3),strides=1,padding='same')(layer1)
layer1_3=layer1
layer_output=tf.add(layer1_1,layer1_2)
layer_output=tf.add(layer_output,layer1)
layer_output=BatchNormalization()(layer_output)
layer_output=Activation('relu')(layer_output)
layer_output=MaxPooling2D(pool_size=(2,2),strides=2)(layer_output)
return layer_output
def attention3_SA1(input_layer):
conv_11=Conv2D(filters=64,kernel_size=(1,1))(input_layer)
conv_12=Conv2D(filters=64,kernel_size=(1,1))(input_layer)
conv11_RES=Reshape((conv_11.shape[1]*conv_11.shape[2],conv_11.shape[3]))(conv_11)
conv12_RES=Reshape((conv_12.shape[1]*conv_12.shape[2],conv_12.shape[3]))(conv_12)
layer1=tf.matmul(conv11_RES,conv12_RES,transpose_b=True)
output_layer=Activation('softmax')(layer1)
return output_layer
def attention3_SA2(input_layer):
conv_2=Conv2D(filters=64,kernel_size=(3,3),padding='same')(input_layer)
conv_2_RES=Reshape((conv_2.shape[1]*conv_2.shape[2],conv_2.shape[3]))(conv_2)
return conv_2_RES
def attention3_SA3(input_layer):
conv_31=DepthwiseConv2D(kernel_size=(3,3))(input_layer)
conv_32=DepthwiseConv2D(kernel_size=(3,3))(input_layer)
conv32_RES=Reshape((conv_32.shape[1]*conv_32.shape[2],conv_32.shape[3]))(conv_32)
conv31_RES=Reshape((conv_31.shape[1]*conv_31.shape[2],conv_31.shape[3]))(conv_31)
layer1=tf.matmul(conv32_RES,conv31_RES,transpose_a=True)
output_layer=Activation('softmax')(layer1)
return output_layer
def attention_3_SA_FINFAL_3(input_layer):
output_1=attention3_SA1(input_layer)
output_2=attention3_SA2(input_layer)
output_3=attention3_SA3(input_layer)
output=tf.matmul(output_1,output_2)
output=tf.matmul(output,output_3)
output_layer=Reshape((6,6,64))(output)
return output_layer
def attention_moudle3(input_layer):
layer1=Conv2D(filters=64,kernel_size=(3,3),strides=1,padding='same')(input_layer)
layer1_1=attention_3_SA_FINFAL_3(layer1)
layer1_2=Conv2D(filters=64,kernel_size=(3,3),strides=1,padding='same')(layer1)
layer1_3=layer1
layer_output=tf.add(layer1_1,layer1_2)
layer_output=tf.add(layer_output,layer1)
layer_output=BatchNormalization()(layer_output)
layer_output=Activation('relu')(layer_output)
return layer_output
def classifer(input_layer):
out1=attention_moudle1(input_layer)
out2=attention_moudle2(out1)
out3=attention_moudle3(out2)
GAP=GlobalAveragePooling2D()(out3)
FC_layer=Dense(units=16,activation='softmax')(GAP)
return FC_layer
input_layer=Input((27,27,20))
output=classifer(input_layer)
Model: "model"
__________________________________________________________________________________________________
Layer (type) Output Shape Param # Connected to
==================================================================================================
input_1 (InputLayer) [(None, 27, 27, 20)] 0
__________________________________________________________________________________________________
conv2d (Conv2D) (None, 27, 27, 16) 2896 input_1[0][0]
__________________________________________________________________________________________________
conv2d_1 (Conv2D) (None, 27, 27, 16) 272 conv2d[0][0]
__________________________________________________________________________________________________
conv2d_2 (Conv2D) (None, 27, 27, 16) 272 conv2d[0][0]
__________________________________________________________________________________________________
reshape (Reshape) (None, 729, 16) 0 conv2d_1[0][0]
__________________________________________________________________________________________________
reshape_1 (Reshape) (None, 729, 16) 0 conv2d_2[0][0]
__________________________________________________________________________________________________
depthwise_conv2d_1 (DepthwiseCo (None, 25, 25, 16) 160 conv2d[0][0]
__________________________________________________________________________________________________
depthwise_conv2d (DepthwiseConv (None, 25, 25, 16) 160 conv2d[0][0]
__________________________________________________________________________________________________
tf_op_layer_BatchMatMulV2 (Tens [(None, 729, 729)] 0 reshape[0][0]
reshape_1[0][0]
__________________________________________________________________________________________________
conv2d_3 (Conv2D) (None, 27, 27, 16) 2320 conv2d[0][0]
__________________________________________________________________________________________________
reshape_3 (Reshape) (None, 625, 16) 0 depthwise_conv2d_1[0][0]
__________________________________________________________________________________________________
reshape_4 (Reshape) (None, 625, 16) 0 depthwise_conv2d[0][0]
__________________________________________________________________________________________________
activation (Activation) (None, 729, 729) 0 tf_op_layer_BatchMatMulV2[0][0]
__________________________________________________________________________________________________
reshape_2 (Reshape) (None, 729, 16) 0 conv2d_3[0][0]
__________________________________________________________________________________________________
tf_op_layer_BatchMatMulV2_1 (Te [(None, 16, 16)] 0 reshape_3[0][0]
reshape_4[0][0]
__________________________________________________________________________________________________
tf_op_layer_BatchMatMulV2_2 (Te [(None, 729, 16)] 0 activation[0][0]
reshape_2[0][0]
__________________________________________________________________________________________________
activation_1 (Activation) (None, 16, 16) 0 tf_op_layer_BatchMatMulV2_1[0][0]
__________________________________________________________________________________________________
tf_op_layer_BatchMatMulV2_3 (Te [(None, 729, 16)] 0 tf_op_layer_BatchMatMulV2_2[0][0]
activation_1[0][0]
__________________________________________________________________________________________________
reshape_5 (Reshape) (None, 27, 27, 16) 0 tf_op_layer_BatchMatMulV2_3[0][0]
__________________________________________________________________________________________________
conv2d_4 (Conv2D) (None, 27, 27, 16) 2320 conv2d[0][0]
__________________________________________________________________________________________________
tf_op_layer_Add (TensorFlowOpLa [(None, 27, 27, 16)] 0 reshape_5[0][0]
conv2d_4[0][0]
__________________________________________________________________________________________________
tf_op_layer_Add_1 (TensorFlowOp [(None, 27, 27, 16)] 0 tf_op_layer_Add[0][0]
conv2d[0][0]
__________________________________________________________________________________________________
batch_normalization (BatchNorma (None, 27, 27, 16) 64 tf_op_layer_Add_1[0][0]
__________________________________________________________________________________________________
activation_2 (Activation) (None, 27, 27, 16) 0 batch_normalization[0][0]
__________________________________________________________________________________________________
max_pooling2d (MaxPooling2D) (None, 13, 13, 16) 0 activation_2[0][0]
__________________________________________________________________________________________________
conv2d_5 (Conv2D) (None, 13, 13, 32) 4640 max_pooling2d[0][0]
__________________________________________________________________________________________________
conv2d_6 (Conv2D) (None, 13, 13, 32) 1056 conv2d_5[0][0]
__________________________________________________________________________________________________
conv2d_7 (Conv2D) (None, 13, 13, 32) 1056 conv2d_5[0][0]
__________________________________________________________________________________________________
reshape_6 (Reshape) (None, 169, 32) 0 conv2d_6[0][0]
__________________________________________________________________________________________________
reshape_7 (Reshape) (None, 169, 32) 0 conv2d_7[0][0]
__________________________________________________________________________________________________
depthwise_conv2d_3 (DepthwiseCo (None, 11, 11, 32) 320 conv2d_5[0][0]
__________________________________________________________________________________________________
depthwise_conv2d_2 (DepthwiseCo (None, 11, 11, 32) 320 conv2d_5[0][0]
__________________________________________________________________________________________________
tf_op_layer_BatchMatMulV2_4 (Te [(None, 169, 169)] 0 reshape_6[0][0]
reshape_7[0][0]
__________________________________________________________________________________________________
conv2d_8 (Conv2D) (None, 13, 13, 32) 9248 conv2d_5[0][0]
__________________________________________________________________________________________________
reshape_9 (Reshape) (None, 121, 32) 0 depthwise_conv2d_3[0][0]
__________________________________________________________________________________________________
reshape_10 (Reshape) (None, 121, 32) 0 depthwise_conv2d_2[0][0]
__________________________________________________________________________________________________
activation_3 (Activation) (None, 169, 169) 0 tf_op_layer_BatchMatMulV2_4[0][0]
__________________________________________________________________________________________________
reshape_8 (Reshape) (None, 169, 32) 0 conv2d_8[0][0]
__________________________________________________________________________________________________
tf_op_layer_BatchMatMulV2_5 (Te [(None, 32, 32)] 0 reshape_9[0][0]
reshape_10[0][0]
__________________________________________________________________________________________________
tf_op_layer_BatchMatMulV2_6 (Te [(None, 169, 32)] 0 activation_3[0][0]
reshape_8[0][0]
__________________________________________________________________________________________________
activation_4 (Activation) (None, 32, 32) 0 tf_op_layer_BatchMatMulV2_5[0][0]
__________________________________________________________________________________________________
tf_op_layer_BatchMatMulV2_7 (Te [(None, 169, 32)] 0 tf_op_layer_BatchMatMulV2_6[0][0]
activation_4[0][0]
__________________________________________________________________________________________________
reshape_11 (Reshape) (None, 13, 13, 32) 0 tf_op_layer_BatchMatMulV2_7[0][0]
__________________________________________________________________________________________________
conv2d_9 (Conv2D) (None, 13, 13, 32) 9248 conv2d_5[0][0]
__________________________________________________________________________________________________
tf_op_layer_Add_2 (TensorFlowOp [(None, 13, 13, 32)] 0 reshape_11[0][0]
conv2d_9[0][0]
__________________________________________________________________________________________________
tf_op_layer_Add_3 (TensorFlowOp [(None, 13, 13, 32)] 0 tf_op_layer_Add_2[0][0]
conv2d_5[0][0]
__________________________________________________________________________________________________
batch_normalization_1 (BatchNor (None, 13, 13, 32) 128 tf_op_layer_Add_3[0][0]
__________________________________________________________________________________________________
activation_5 (Activation) (None, 13, 13, 32) 0 batch_normalization_1[0][0]
__________________________________________________________________________________________________
max_pooling2d_1 (MaxPooling2D) (None, 6, 6, 32) 0 activation_5[0][0]
__________________________________________________________________________________________________
conv2d_10 (Conv2D) (None, 6, 6, 64) 18496 max_pooling2d_1[0][0]
__________________________________________________________________________________________________
conv2d_11 (Conv2D) (None, 6, 6, 64) 4160 conv2d_10[0][0]
__________________________________________________________________________________________________
conv2d_12 (Conv2D) (None, 6, 6, 64) 4160 conv2d_10[0][0]
__________________________________________________________________________________________________
reshape_12 (Reshape) (None, 36, 64) 0 conv2d_11[0][0]
__________________________________________________________________________________________________
reshape_13 (Reshape) (None, 36, 64) 0 conv2d_12[0][0]
__________________________________________________________________________________________________
depthwise_conv2d_5 (DepthwiseCo (None, 4, 4, 64) 640 conv2d_10[0][0]
__________________________________________________________________________________________________
depthwise_conv2d_4 (DepthwiseCo (None, 4, 4, 64) 640 conv2d_10[0][0]
__________________________________________________________________________________________________
tf_op_layer_BatchMatMulV2_8 (Te [(None, 36, 36)] 0 reshape_12[0][0]
reshape_13[0][0]
__________________________________________________________________________________________________
conv2d_13 (Conv2D) (None, 6, 6, 64) 36928 conv2d_10[0][0]
__________________________________________________________________________________________________
reshape_15 (Reshape) (None, 16, 64) 0 depthwise_conv2d_5[0][0]
__________________________________________________________________________________________________
reshape_16 (Reshape) (None, 16, 64) 0 depthwise_conv2d_4[0][0]
__________________________________________________________________________________________________
activation_6 (Activation) (None, 36, 36) 0 tf_op_layer_BatchMatMulV2_8[0][0]
__________________________________________________________________________________________________
reshape_14 (Reshape) (None, 36, 64) 0 conv2d_13[0][0]
__________________________________________________________________________________________________
tf_op_layer_BatchMatMulV2_9 (Te [(None, 64, 64)] 0 reshape_15[0][0]
reshape_16[0][0]
__________________________________________________________________________________________________
tf_op_layer_BatchMatMulV2_10 (T [(None, 36, 64)] 0 activation_6[0][0]
reshape_14[0][0]
__________________________________________________________________________________________________
activation_7 (Activation) (None, 64, 64) 0 tf_op_layer_BatchMatMulV2_9[0][0]
__________________________________________________________________________________________________
tf_op_layer_BatchMatMulV2_11 (T [(None, 36, 64)] 0 tf_op_layer_BatchMatMulV2_10[0][0
activation_7[0][0]
__________________________________________________________________________________________________
reshape_17 (Reshape) (None, 6, 6, 64) 0 tf_op_layer_BatchMatMulV2_11[0][0
__________________________________________________________________________________________________
conv2d_14 (Conv2D) (None, 6, 6, 64) 36928 conv2d_10[0][0]
__________________________________________________________________________________________________
tf_op_layer_Add_4 (TensorFlowOp [(None, 6, 6, 64)] 0 reshape_17[0][0]
conv2d_14[0][0]
__________________________________________________________________________________________________
tf_op_layer_Add_5 (TensorFlowOp [(None, 6, 6, 64)] 0 tf_op_layer_Add_4[0][0]
conv2d_10[0][0]
__________________________________________________________________________________________________
batch_normalization_2 (BatchNor (None, 6, 6, 64) 256 tf_op_layer_Add_5[0][0]
__________________________________________________________________________________________________
activation_8 (Activation) (None, 6, 6, 64) 0 batch_normalization_2[0][0]
__________________________________________________________________________________________________
global_average_pooling2d (Globa (None, 64) 0 activation_8[0][0]
__________________________________________________________________________________________________
dense (Dense) (None, 16) 1040 global_average_pooling2d[0][0]
==================================================================================================
Total params: 137,728
Trainable params: 137,504
Non-trainable params: 224
__________________________________________________________________________________________________
adam=Adam(lr=0.0001)
model.compile(loss='categorical_crossentropy',optimizer=adam,metrics=['accuracy'])
filepath='Joint Spatial-Spectral Attention Network.hdf5'
checkpoint=ModelCheckpoint(filepath,monitor='loss',verbose=1,save_best_only=True,
mode='min')
callback_list=[checkpoint]
Epoch 1/400
8/8 [==============================] - ETA: 0s - loss: 3.0383 - accuracy: 0.0234
Epoch 00001: loss improved from inf to 3.03825, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 3.0383 - accuracy: 0.0234
Epoch 2/400
8/8 [==============================] - ETA: 0s - loss: 2.7759 - accuracy: 0.0303
Epoch 00002: loss improved from 3.03825 to 2.77587, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 2.7759 - accuracy: 0.0303
Epoch 3/400
8/8 [==============================] - ETA: 0s - loss: 2.5516 - accuracy: 0.0938
Epoch 00003: loss improved from 2.77587 to 2.55160, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 2.5516 - accuracy: 0.0938
Epoch 4/400
8/8 [==============================] - ETA: 0s - loss: 2.3479 - accuracy: 0.1504
Epoch 00004: loss improved from 2.55160 to 2.34791, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 2.3479 - accuracy: 0.1504
Epoch 5/400
8/8 [==============================] - ETA: 0s - loss: 2.1612 - accuracy: 0.2197
Epoch 00005: loss improved from 2.34791 to 2.16123, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 2.1612 - accuracy: 0.2197
Epoch 6/400
8/8 [==============================] - ETA: 0s - loss: 1.9966 - accuracy: 0.3535
Epoch 00006: loss improved from 2.16123 to 1.99662, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 1.9966 - accuracy: 0.3535
Epoch 7/400
8/8 [==============================] - ETA: 0s - loss: 1.8277 - accuracy: 0.4658
Epoch 00007: loss improved from 1.99662 to 1.82772, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 1.8277 - accuracy: 0.4658
Epoch 8/400
8/8 [==============================] - ETA: 0s - loss: 1.6777 - accuracy: 0.5986
Epoch 00008: loss improved from 1.82772 to 1.67767, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 1.6777 - accuracy: 0.5986
Epoch 9/400
8/8 [==============================] - ETA: 0s - loss: 1.5471 - accuracy: 0.6572
Epoch 00009: loss improved from 1.67767 to 1.54714, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 1.5471 - accuracy: 0.6572
Epoch 10/400
8/8 [==============================] - ETA: 0s - loss: 1.4404 - accuracy: 0.6885
Epoch 00010: loss improved from 1.54714 to 1.44038, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 1.4404 - accuracy: 0.6885
Epoch 11/400
8/8 [==============================] - ETA: 0s - loss: 1.3371 - accuracy: 0.7139
Epoch 00011: loss improved from 1.44038 to 1.33707, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 1.3371 - accuracy: 0.7139
Epoch 12/400
8/8 [==============================] - ETA: 0s - loss: 1.2422 - accuracy: 0.7422
Epoch 00012: loss improved from 1.33707 to 1.24221, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 1.2422 - accuracy: 0.7422
Epoch 13/400
8/8 [==============================] - ETA: 0s - loss: 1.1711 - accuracy: 0.7637
Epoch 00013: loss improved from 1.24221 to 1.17107, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 1.1711 - accuracy: 0.7637
Epoch 14/400
8/8 [==============================] - ETA: 0s - loss: 1.0906 - accuracy: 0.7852
Epoch 00014: loss improved from 1.17107 to 1.09058, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 1.0906 - accuracy: 0.7852
Epoch 15/400
8/8 [==============================] - ETA: 0s - loss: 1.0282 - accuracy: 0.8066
Epoch 00015: loss improved from 1.09058 to 1.02823, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 1.0282 - accuracy: 0.8066
Epoch 16/400
8/8 [==============================] - ETA: 0s - loss: 0.9746 - accuracy: 0.8184
Epoch 00016: loss improved from 1.02823 to 0.97463, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.9746 - accuracy: 0.8184
Epoch 17/400
8/8 [==============================] - ETA: 0s - loss: 0.9149 - accuracy: 0.8271
Epoch 00017: loss improved from 0.97463 to 0.91490, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.9149 - accuracy: 0.8271
Epoch 18/400
8/8 [==============================] - ETA: 0s - loss: 0.8781 - accuracy: 0.8486
Epoch 00018: loss improved from 0.91490 to 0.87813, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.8781 - accuracy: 0.8486
Epoch 19/400
8/8 [==============================] - ETA: 0s - loss: 0.8366 - accuracy: 0.8604
Epoch 00019: loss improved from 0.87813 to 0.83664, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.8366 - accuracy: 0.8604
Epoch 20/400
8/8 [==============================] - ETA: 0s - loss: 0.7783 - accuracy: 0.8721
Epoch 00020: loss improved from 0.83664 to 0.77834, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.7783 - accuracy: 0.8721
Epoch 21/400
8/8 [==============================] - ETA: 0s - loss: 0.7355 - accuracy: 0.8857
Epoch 00021: loss improved from 0.77834 to 0.73554, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.7355 - accuracy: 0.8857
Epoch 22/400
8/8 [==============================] - ETA: 0s - loss: 0.7239 - accuracy: 0.8857
Epoch 00022: loss improved from 0.73554 to 0.72387, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.7239 - accuracy: 0.8857
Epoch 23/400
8/8 [==============================] - ETA: 0s - loss: 0.6845 - accuracy: 0.8877
Epoch 00023: loss improved from 0.72387 to 0.68453, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.6845 - accuracy: 0.8877
Epoch 24/400
8/8 [==============================] - ETA: 0s - loss: 0.6397 - accuracy: 0.8975
Epoch 00024: loss improved from 0.68453 to 0.63974, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.6397 - accuracy: 0.8975
Epoch 25/400
8/8 [==============================] - ETA: 0s - loss: 0.6169 - accuracy: 0.8984
Epoch 00025: loss improved from 0.63974 to 0.61687, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.6169 - accuracy: 0.8984
Epoch 26/400
8/8 [==============================] - ETA: 0s - loss: 0.5815 - accuracy: 0.9102
Epoch 00026: loss improved from 0.61687 to 0.58152, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.5815 - accuracy: 0.9102
Epoch 27/400
8/8 [==============================] - ETA: 0s - loss: 0.5599 - accuracy: 0.9043
Epoch 00027: loss improved from 0.58152 to 0.55985, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.5599 - accuracy: 0.9043
Epoch 28/400
8/8 [==============================] - ETA: 0s - loss: 0.5457 - accuracy: 0.9102
Epoch 00028: loss improved from 0.55985 to 0.54574, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.5457 - accuracy: 0.9102
Epoch 29/400
8/8 [==============================] - ETA: 0s - loss: 0.5144 - accuracy: 0.9170
Epoch 00029: loss improved from 0.54574 to 0.51441, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.5144 - accuracy: 0.9170
Epoch 30/400
8/8 [==============================] - ETA: 0s - loss: 0.5016 - accuracy: 0.9248
Epoch 00030: loss improved from 0.51441 to 0.50162, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.5016 - accuracy: 0.9248
Epoch 31/400
8/8 [==============================] - ETA: 0s - loss: 0.4719 - accuracy: 0.9277
Epoch 00031: loss improved from 0.50162 to 0.47186, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 0.4719 - accuracy: 0.9277
Epoch 32/400
8/8 [==============================] - ETA: 0s - loss: 0.4555 - accuracy: 0.9307
Epoch 00032: loss improved from 0.47186 to 0.45546, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.4555 - accuracy: 0.9307
Epoch 33/400
8/8 [==============================] - ETA: 0s - loss: 0.4489 - accuracy: 0.9238
Epoch 00033: loss improved from 0.45546 to 0.44891, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.4489 - accuracy: 0.9238
Epoch 34/400
8/8 [==============================] - ETA: 0s - loss: 0.4210 - accuracy: 0.9346
Epoch 00034: loss improved from 0.44891 to 0.42102, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 0.4210 - accuracy: 0.9346
Epoch 35/400
8/8 [==============================] - ETA: 0s - loss: 0.4161 - accuracy: 0.9346
Epoch 00035: loss improved from 0.42102 to 0.41609, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.4161 - accuracy: 0.9346
Epoch 36/400
8/8 [==============================] - ETA: 0s - loss: 0.3902 - accuracy: 0.9395
Epoch 00036: loss improved from 0.41609 to 0.39017, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.3902 - accuracy: 0.9395
Epoch 37/400
8/8 [==============================] - ETA: 0s - loss: 0.3944 - accuracy: 0.9424
Epoch 00037: loss did not improve from 0.39017
8/8 [==============================] - 1s 65ms/step - loss: 0.3944 - accuracy: 0.9424
Epoch 38/400
8/8 [==============================] - ETA: 0s - loss: 0.3725 - accuracy: 0.9424
Epoch 00038: loss improved from 0.39017 to 0.37245, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.3725 - accuracy: 0.9424
Epoch 39/400
8/8 [==============================] - ETA: 0s - loss: 0.3757 - accuracy: 0.9473
Epoch 00039: loss did not improve from 0.37245
8/8 [==============================] - 1s 66ms/step - loss: 0.3757 - accuracy: 0.9473
Epoch 40/400
8/8 [==============================] - ETA: 0s - loss: 0.3392 - accuracy: 0.9541
Epoch 00040: loss improved from 0.37245 to 0.33922, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.3392 - accuracy: 0.9541
Epoch 41/400
8/8 [==============================] - ETA: 0s - loss: 0.3367 - accuracy: 0.9561
Epoch 00041: loss improved from 0.33922 to 0.33670, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.3367 - accuracy: 0.9561
Epoch 42/400
8/8 [==============================] - ETA: 0s - loss: 0.3333 - accuracy: 0.9521
Epoch 00042: loss improved from 0.33670 to 0.33333, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.3333 - accuracy: 0.9521
Epoch 43/400
8/8 [==============================] - ETA: 0s - loss: 0.3354 - accuracy: 0.9619
Epoch 00043: loss did not improve from 0.33333
8/8 [==============================] - 1s 65ms/step - loss: 0.3354 - accuracy: 0.9619
Epoch 44/400
8/8 [==============================] - ETA: 0s - loss: 0.3122 - accuracy: 0.9609
Epoch 00044: loss improved from 0.33333 to 0.31216, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.3122 - accuracy: 0.9609
Epoch 45/400
8/8 [==============================] - ETA: 0s - loss: 0.2996 - accuracy: 0.9600
Epoch 00045: loss improved from 0.31216 to 0.29960, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.2996 - accuracy: 0.9600
Epoch 46/400
8/8 [==============================] - ETA: 0s - loss: 0.2936 - accuracy: 0.9609
Epoch 00046: loss improved from 0.29960 to 0.29356, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.2936 - accuracy: 0.9609
Epoch 47/400
8/8 [==============================] - ETA: 0s - loss: 0.2805 - accuracy: 0.9639
Epoch 00047: loss improved from 0.29356 to 0.28046, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.2805 - accuracy: 0.9639
Epoch 48/400
8/8 [==============================] - ETA: 0s - loss: 0.2644 - accuracy: 0.9707
Epoch 00048: loss improved from 0.28046 to 0.26440, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.2644 - accuracy: 0.9707
Epoch 49/400
8/8 [==============================] - ETA: 0s - loss: 0.2583 - accuracy: 0.9756
Epoch 00049: loss improved from 0.26440 to 0.25833, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.2583 - accuracy: 0.9756
Epoch 50/400
8/8 [==============================] - ETA: 0s - loss: 0.2509 - accuracy: 0.9756
Epoch 00050: loss improved from 0.25833 to 0.25088, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.2509 - accuracy: 0.9756
Epoch 51/400
8/8 [==============================] - ETA: 0s - loss: 0.2484 - accuracy: 0.9756
Epoch 00051: loss improved from 0.25088 to 0.24845, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.2484 - accuracy: 0.9756
Epoch 52/400
8/8 [==============================] - ETA: 0s - loss: 0.2354 - accuracy: 0.9766
Epoch 00052: loss improved from 0.24845 to 0.23536, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 0.2354 - accuracy: 0.9766
Epoch 53/400
8/8 [==============================] - ETA: 0s - loss: 0.2345 - accuracy: 0.9795
Epoch 00053: loss improved from 0.23536 to 0.23451, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.2345 - accuracy: 0.9795
Epoch 54/400
8/8 [==============================] - ETA: 0s - loss: 0.2222 - accuracy: 0.9834
Epoch 00054: loss improved from 0.23451 to 0.22221, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.2222 - accuracy: 0.9834
Epoch 55/400
8/8 [==============================] - ETA: 0s - loss: 0.2123 - accuracy: 0.9814
Epoch 00055: loss improved from 0.22221 to 0.21225, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 0.2123 - accuracy: 0.9814
Epoch 56/400
8/8 [==============================] - ETA: 0s - loss: 0.2040 - accuracy: 0.9844
Epoch 00056: loss improved from 0.21225 to 0.20397, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.2040 - accuracy: 0.9844
Epoch 57/400
8/8 [==============================] - ETA: 0s - loss: 0.2006 - accuracy: 0.9854
Epoch 00057: loss improved from 0.20397 to 0.20065, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.2006 - accuracy: 0.9854
Epoch 58/400
8/8 [==============================] - ETA: 0s - loss: 0.1911 - accuracy: 0.9854
Epoch 00058: loss improved from 0.20065 to 0.19109, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 0.1911 - accuracy: 0.9854
Epoch 59/400
8/8 [==============================] - ETA: 0s - loss: 0.1896 - accuracy: 0.9883
Epoch 00059: loss improved from 0.19109 to 0.18961, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.1896 - accuracy: 0.9883
Epoch 60/400
8/8 [==============================] - ETA: 0s - loss: 0.1809 - accuracy: 0.9873
Epoch 00060: loss improved from 0.18961 to 0.18088, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.1809 - accuracy: 0.9873
Epoch 61/400
8/8 [==============================] - ETA: 0s - loss: 0.1905 - accuracy: 0.9824
Epoch 00061: loss did not improve from 0.18088
8/8 [==============================] - 1s 64ms/step - loss: 0.1905 - accuracy: 0.9824
Epoch 62/400
8/8 [==============================] - ETA: 0s - loss: 0.1808 - accuracy: 0.9912
Epoch 00062: loss improved from 0.18088 to 0.18085, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.1808 - accuracy: 0.9912
Epoch 63/400
8/8 [==============================] - ETA: 0s - loss: 0.1665 - accuracy: 0.9902
Epoch 00063: loss improved from 0.18085 to 0.16648, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.1665 - accuracy: 0.9902
Epoch 64/400
8/8 [==============================] - ETA: 0s - loss: 0.1745 - accuracy: 0.9854
Epoch 00064: loss did not improve from 0.16648
8/8 [==============================] - 1s 67ms/step - loss: 0.1745 - accuracy: 0.9854
Epoch 65/400
8/8 [==============================] - ETA: 0s - loss: 0.1596 - accuracy: 0.9941
Epoch 00065: loss improved from 0.16648 to 0.15963, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.1596 - accuracy: 0.9941
Epoch 66/400
8/8 [==============================] - ETA: 0s - loss: 0.1530 - accuracy: 0.9951
Epoch 00066: loss improved from 0.15963 to 0.15298, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.1530 - accuracy: 0.9951
Epoch 67/400
8/8 [==============================] - ETA: 0s - loss: 0.1507 - accuracy: 0.9932
Epoch 00067: loss improved from 0.15298 to 0.15074, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.1507 - accuracy: 0.9932
Epoch 68/400
8/8 [==============================] - ETA: 0s - loss: 0.1499 - accuracy: 0.9951
Epoch 00068: loss improved from 0.15074 to 0.14993, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.1499 - accuracy: 0.9951
Epoch 69/400
8/8 [==============================] - ETA: 0s - loss: 0.1468 - accuracy: 0.9951
Epoch 00069: loss improved from 0.14993 to 0.14676, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.1468 - accuracy: 0.9951
Epoch 70/400
8/8 [==============================] - ETA: 0s - loss: 0.1410 - accuracy: 0.9961
Epoch 00070: loss improved from 0.14676 to 0.14104, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.1410 - accuracy: 0.9961
Epoch 71/400
8/8 [==============================] - ETA: 0s - loss: 0.1355 - accuracy: 0.9971
Epoch 00071: loss improved from 0.14104 to 0.13548, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 78ms/step - loss: 0.1355 - accuracy: 0.9971
Epoch 72/400
8/8 [==============================] - ETA: 0s - loss: 0.1344 - accuracy: 0.9932
Epoch 00072: loss improved from 0.13548 to 0.13439, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.1344 - accuracy: 0.9932
Epoch 73/400
8/8 [==============================] - ETA: 0s - loss: 0.1316 - accuracy: 0.9980
Epoch 00073: loss improved from 0.13439 to 0.13159, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.1316 - accuracy: 0.9980
Epoch 74/400
8/8 [==============================] - ETA: 0s - loss: 0.1272 - accuracy: 0.9990
Epoch 00074: loss improved from 0.13159 to 0.12725, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.1272 - accuracy: 0.9990
Epoch 75/400
8/8 [==============================] - ETA: 0s - loss: 0.1231 - accuracy: 0.9971
Epoch 00075: loss improved from 0.12725 to 0.12307, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.1231 - accuracy: 0.9971
Epoch 76/400
8/8 [==============================] - ETA: 0s - loss: 0.1239 - accuracy: 0.9971
Epoch 00076: loss did not improve from 0.12307
8/8 [==============================] - 1s 67ms/step - loss: 0.1239 - accuracy: 0.9971
Epoch 77/400
8/8 [==============================] - ETA: 0s - loss: 0.1248 - accuracy: 0.9961
Epoch 00077: loss did not improve from 0.12307
8/8 [==============================] - 1s 67ms/step - loss: 0.1248 - accuracy: 0.9961
Epoch 78/400
8/8 [==============================] - ETA: 0s - loss: 0.1161 - accuracy: 0.9980
Epoch 00078: loss improved from 0.12307 to 0.11610, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.1161 - accuracy: 0.9980
Epoch 79/400
8/8 [==============================] - ETA: 0s - loss: 0.1149 - accuracy: 0.9971
Epoch 00079: loss improved from 0.11610 to 0.11495, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.1149 - accuracy: 0.9971
Epoch 80/400
8/8 [==============================] - ETA: 0s - loss: 0.1087 - accuracy: 0.9971
Epoch 00080: loss improved from 0.11495 to 0.10866, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.1087 - accuracy: 0.9971
Epoch 81/400
8/8 [==============================] - ETA: 0s - loss: 0.1019 - accuracy: 1.0000
Epoch 00081: loss improved from 0.10866 to 0.10187, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.1019 - accuracy: 1.0000
Epoch 82/400
8/8 [==============================] - ETA: 0s - loss: 0.1020 - accuracy: 0.9980
Epoch 00082: loss did not improve from 0.10187
8/8 [==============================] - 1s 65ms/step - loss: 0.1020 - accuracy: 0.9980
Epoch 83/400
8/8 [==============================] - ETA: 0s - loss: 0.0968 - accuracy: 1.0000
Epoch 00083: loss improved from 0.10187 to 0.09682, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.0968 - accuracy: 1.0000
Epoch 84/400
8/8 [==============================] - ETA: 0s - loss: 0.0964 - accuracy: 0.9990
Epoch 00084: loss improved from 0.09682 to 0.09643, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0964 - accuracy: 0.9990
Epoch 85/400
8/8 [==============================] - ETA: 0s - loss: 0.0918 - accuracy: 1.0000
Epoch 00085: loss improved from 0.09643 to 0.09185, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.0918 - accuracy: 1.0000
Epoch 86/400
8/8 [==============================] - ETA: 0s - loss: 0.0908 - accuracy: 0.9990
Epoch 00086: loss improved from 0.09185 to 0.09079, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0908 - accuracy: 0.9990
Epoch 87/400
8/8 [==============================] - ETA: 0s - loss: 0.0939 - accuracy: 0.9990
Epoch 00087: loss did not improve from 0.09079
8/8 [==============================] - 1s 66ms/step - loss: 0.0939 - accuracy: 0.9990
Epoch 88/400
8/8 [==============================] - ETA: 0s - loss: 0.0877 - accuracy: 1.0000
Epoch 00088: loss improved from 0.09079 to 0.08774, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 0.0877 - accuracy: 1.0000
Epoch 89/400
8/8 [==============================] - ETA: 0s - loss: 0.0849 - accuracy: 1.0000
Epoch 00089: loss improved from 0.08774 to 0.08487, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0849 - accuracy: 1.0000
Epoch 90/400
8/8 [==============================] - ETA: 0s - loss: 0.0841 - accuracy: 1.0000
Epoch 00090: loss improved from 0.08487 to 0.08412, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0841 - accuracy: 1.0000
Epoch 91/400
8/8 [==============================] - ETA: 0s - loss: 0.0895 - accuracy: 1.0000
Epoch 00091: loss did not improve from 0.08412
8/8 [==============================] - 1s 65ms/step - loss: 0.0895 - accuracy: 1.0000
Epoch 92/400
8/8 [==============================] - ETA: 0s - loss: 0.0823 - accuracy: 1.0000
Epoch 00092: loss improved from 0.08412 to 0.08230, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.0823 - accuracy: 1.0000
Epoch 93/400
8/8 [==============================] - ETA: 0s - loss: 0.0793 - accuracy: 1.0000
Epoch 00093: loss improved from 0.08230 to 0.07927, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0793 - accuracy: 1.0000
Epoch 94/400
8/8 [==============================] - ETA: 0s - loss: 0.0797 - accuracy: 1.0000
Epoch 00094: loss did not improve from 0.07927
8/8 [==============================] - 1s 65ms/step - loss: 0.0797 - accuracy: 1.0000
Epoch 95/400
8/8 [==============================] - ETA: 0s - loss: 0.0774 - accuracy: 1.0000
Epoch 00095: loss improved from 0.07927 to 0.07742, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0774 - accuracy: 1.0000
Epoch 96/400
8/8 [==============================] - ETA: 0s - loss: 0.0742 - accuracy: 1.0000
Epoch 00096: loss improved from 0.07742 to 0.07417, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0742 - accuracy: 1.0000
Epoch 97/400
8/8 [==============================] - ETA: 0s - loss: 0.0758 - accuracy: 0.9990
Epoch 00097: loss did not improve from 0.07417
8/8 [==============================] - 1s 65ms/step - loss: 0.0758 - accuracy: 0.9990
Epoch 98/400
8/8 [==============================] - ETA: 0s - loss: 0.0728 - accuracy: 1.0000
Epoch 00098: loss improved from 0.07417 to 0.07279, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0728 - accuracy: 1.0000
Epoch 99/400
8/8 [==============================] - ETA: 0s - loss: 0.0719 - accuracy: 1.0000
Epoch 00099: loss improved from 0.07279 to 0.07192, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.0719 - accuracy: 1.0000
Epoch 100/400
8/8 [==============================] - ETA: 0s - loss: 0.0711 - accuracy: 1.0000
Epoch 00100: loss improved from 0.07192 to 0.07113, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 0.0711 - accuracy: 1.0000
Epoch 101/400
8/8 [==============================] - ETA: 0s - loss: 0.0682 - accuracy: 1.0000
Epoch 00101: loss improved from 0.07113 to 0.06819, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0682 - accuracy: 1.0000
Epoch 102/400
8/8 [==============================] - ETA: 0s - loss: 0.0713 - accuracy: 1.0000
Epoch 00102: loss did not improve from 0.06819
8/8 [==============================] - 1s 66ms/step - loss: 0.0713 - accuracy: 1.0000
Epoch 103/400
8/8 [==============================] - ETA: 0s - loss: 0.0708 - accuracy: 0.9990
Epoch 00103: loss did not improve from 0.06819
8/8 [==============================] - 1s 64ms/step - loss: 0.0708 - accuracy: 0.9990
Epoch 104/400
8/8 [==============================] - ETA: 0s - loss: 0.0677 - accuracy: 0.9990
Epoch 00104: loss improved from 0.06819 to 0.06775, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.0677 - accuracy: 0.9990
Epoch 105/400
8/8 [==============================] - ETA: 0s - loss: 0.0690 - accuracy: 1.0000
Epoch 00105: loss did not improve from 0.06775
8/8 [==============================] - 1s 67ms/step - loss: 0.0690 - accuracy: 1.0000
Epoch 106/400
8/8 [==============================] - ETA: 0s - loss: 0.0628 - accuracy: 1.0000
Epoch 00106: loss improved from 0.06775 to 0.06281, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 0.0628 - accuracy: 1.0000
Epoch 107/400
8/8 [==============================] - ETA: 0s - loss: 0.0642 - accuracy: 1.0000
Epoch 00107: loss did not improve from 0.06281
8/8 [==============================] - 1s 67ms/step - loss: 0.0642 - accuracy: 1.0000
Epoch 108/400
8/8 [==============================] - ETA: 0s - loss: 0.0591 - accuracy: 1.0000
Epoch 00108: loss improved from 0.06281 to 0.05909, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0591 - accuracy: 1.0000
Epoch 109/400
8/8 [==============================] - ETA: 0s - loss: 0.0597 - accuracy: 1.0000
Epoch 00109: loss did not improve from 0.05909
8/8 [==============================] - 1s 66ms/step - loss: 0.0597 - accuracy: 1.0000
Epoch 110/400
8/8 [==============================] - ETA: 0s - loss: 0.0580 - accuracy: 1.0000
Epoch 00110: loss improved from 0.05909 to 0.05802, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0580 - accuracy: 1.0000
Epoch 111/400
8/8 [==============================] - ETA: 0s - loss: 0.0599 - accuracy: 1.0000
Epoch 00111: loss did not improve from 0.05802
8/8 [==============================] - 1s 67ms/step - loss: 0.0599 - accuracy: 1.0000
Epoch 112/400
8/8 [==============================] - ETA: 0s - loss: 0.0579 - accuracy: 1.0000
Epoch 00112: loss improved from 0.05802 to 0.05794, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0579 - accuracy: 1.0000
Epoch 113/400
8/8 [==============================] - ETA: 0s - loss: 0.0588 - accuracy: 1.0000
Epoch 00113: loss did not improve from 0.05794
8/8 [==============================] - 1s 66ms/step - loss: 0.0588 - accuracy: 1.0000
Epoch 114/400
8/8 [==============================] - ETA: 0s - loss: 0.0578 - accuracy: 1.0000
Epoch 00114: loss improved from 0.05794 to 0.05776, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0578 - accuracy: 1.0000
Epoch 115/400
8/8 [==============================] - ETA: 0s - loss: 0.0511 - accuracy: 1.0000
Epoch 00115: loss improved from 0.05776 to 0.05110, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0511 - accuracy: 1.0000
Epoch 116/400
8/8 [==============================] - ETA: 0s - loss: 0.0531 - accuracy: 1.0000
Epoch 00116: loss did not improve from 0.05110
8/8 [==============================] - 1s 66ms/step - loss: 0.0531 - accuracy: 1.0000
Epoch 117/400
8/8 [==============================] - ETA: 0s - loss: 0.0535 - accuracy: 1.0000
Epoch 00117: loss did not improve from 0.05110
8/8 [==============================] - 1s 66ms/step - loss: 0.0535 - accuracy: 1.0000
Epoch 118/400
8/8 [==============================] - ETA: 0s - loss: 0.0553 - accuracy: 0.9990
Epoch 00118: loss did not improve from 0.05110
8/8 [==============================] - 1s 67ms/step - loss: 0.0553 - accuracy: 0.9990
Epoch 119/400
8/8 [==============================] - ETA: 0s - loss: 0.0500 - accuracy: 1.0000
Epoch 00119: loss improved from 0.05110 to 0.05003, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 0.0500 - accuracy: 1.0000
Epoch 120/400
8/8 [==============================] - ETA: 0s - loss: 0.0498 - accuracy: 1.0000
Epoch 00120: loss improved from 0.05003 to 0.04979, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0498 - accuracy: 1.0000
Epoch 121/400
8/8 [==============================] - ETA: 0s - loss: 0.0476 - accuracy: 1.0000
Epoch 00121: loss improved from 0.04979 to 0.04759, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0476 - accuracy: 1.0000
Epoch 122/400
8/8 [==============================] - ETA: 0s - loss: 0.0481 - accuracy: 1.0000
Epoch 00122: loss did not improve from 0.04759
8/8 [==============================] - 1s 65ms/step - loss: 0.0481 - accuracy: 1.0000
Epoch 123/400
8/8 [==============================] - ETA: 0s - loss: 0.0488 - accuracy: 1.0000
Epoch 00123: loss did not improve from 0.04759
8/8 [==============================] - 1s 67ms/step - loss: 0.0488 - accuracy: 1.0000
Epoch 124/400
8/8 [==============================] - ETA: 0s - loss: 0.0457 - accuracy: 1.0000
Epoch 00124: loss improved from 0.04759 to 0.04568, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.0457 - accuracy: 1.0000
Epoch 125/400
8/8 [==============================] - ETA: 0s - loss: 0.0508 - accuracy: 0.9990
Epoch 00125: loss did not improve from 0.04568
8/8 [==============================] - 1s 64ms/step - loss: 0.0508 - accuracy: 0.9990
Epoch 126/400
8/8 [==============================] - ETA: 0s - loss: 0.0490 - accuracy: 0.9990
Epoch 00126: loss did not improve from 0.04568
8/8 [==============================] - 1s 66ms/step - loss: 0.0490 - accuracy: 0.9990
Epoch 127/400
8/8 [==============================] - ETA: 0s - loss: 0.0497 - accuracy: 1.0000
Epoch 00127: loss did not improve from 0.04568
8/8 [==============================] - 1s 66ms/step - loss: 0.0497 - accuracy: 1.0000
Epoch 128/400
8/8 [==============================] - ETA: 0s - loss: 0.0514 - accuracy: 0.9971
Epoch 00128: loss did not improve from 0.04568
8/8 [==============================] - 1s 65ms/step - loss: 0.0514 - accuracy: 0.9971
Epoch 129/400
8/8 [==============================] - ETA: 0s - loss: 0.0467 - accuracy: 1.0000
Epoch 00129: loss did not improve from 0.04568
8/8 [==============================] - 1s 67ms/step - loss: 0.0467 - accuracy: 1.0000
Epoch 130/400
8/8 [==============================] - ETA: 0s - loss: 0.0464 - accuracy: 0.9990
Epoch 00130: loss did not improve from 0.04568
8/8 [==============================] - 1s 67ms/step - loss: 0.0464 - accuracy: 0.9990
Epoch 131/400
8/8 [==============================] - ETA: 0s - loss: 0.0510 - accuracy: 0.9990
Epoch 00131: loss did not improve from 0.04568
8/8 [==============================] - 1s 66ms/step - loss: 0.0510 - accuracy: 0.9990
Epoch 132/400
8/8 [==============================] - ETA: 0s - loss: 0.0449 - accuracy: 1.0000
Epoch 00132: loss improved from 0.04568 to 0.04487, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 0.0449 - accuracy: 1.0000
Epoch 133/400
8/8 [==============================] - ETA: 0s - loss: 0.0438 - accuracy: 0.9990
Epoch 00133: loss improved from 0.04487 to 0.04378, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0438 - accuracy: 0.9990
Epoch 134/400
8/8 [==============================] - ETA: 0s - loss: 0.0401 - accuracy: 1.0000
Epoch 00134: loss improved from 0.04378 to 0.04006, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0401 - accuracy: 1.0000
Epoch 135/400
8/8 [==============================] - ETA: 0s - loss: 0.0399 - accuracy: 1.0000
Epoch 00135: loss improved from 0.04006 to 0.03989, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.0399 - accuracy: 1.0000
Epoch 136/400
8/8 [==============================] - ETA: 0s - loss: 0.0407 - accuracy: 1.0000
Epoch 00136: loss did not improve from 0.03989
8/8 [==============================] - 1s 67ms/step - loss: 0.0407 - accuracy: 1.0000
Epoch 137/400
8/8 [==============================] - ETA: 0s - loss: 0.0409 - accuracy: 1.0000
Epoch 00137: loss did not improve from 0.03989
8/8 [==============================] - 1s 66ms/step - loss: 0.0409 - accuracy: 1.0000
Epoch 138/400
8/8 [==============================] - ETA: 0s - loss: 0.0381 - accuracy: 1.0000
Epoch 00138: loss improved from 0.03989 to 0.03809, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 0.0381 - accuracy: 1.0000
Epoch 139/400
8/8 [==============================] - ETA: 0s - loss: 0.0385 - accuracy: 1.0000
Epoch 00139: loss did not improve from 0.03809
8/8 [==============================] - 1s 66ms/step - loss: 0.0385 - accuracy: 1.0000
Epoch 140/400
8/8 [==============================] - ETA: 0s - loss: 0.0366 - accuracy: 1.0000
Epoch 00140: loss improved from 0.03809 to 0.03663, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0366 - accuracy: 1.0000
Epoch 141/400
8/8 [==============================] - ETA: 0s - loss: 0.0366 - accuracy: 1.0000
Epoch 00141: loss improved from 0.03663 to 0.03658, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.0366 - accuracy: 1.0000
Epoch 142/400
8/8 [==============================] - ETA: 0s - loss: 0.0346 - accuracy: 1.0000
Epoch 00142: loss improved from 0.03658 to 0.03464, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0346 - accuracy: 1.0000
Epoch 143/400
8/8 [==============================] - ETA: 0s - loss: 0.0353 - accuracy: 1.0000
Epoch 00143: loss did not improve from 0.03464
8/8 [==============================] - 1s 67ms/step - loss: 0.0353 - accuracy: 1.0000
Epoch 144/400
8/8 [==============================] - ETA: 0s - loss: 0.0352 - accuracy: 1.0000
Epoch 00144: loss did not improve from 0.03464
8/8 [==============================] - 1s 65ms/step - loss: 0.0352 - accuracy: 1.0000
Epoch 145/400
8/8 [==============================] - ETA: 0s - loss: 0.0336 - accuracy: 1.0000
Epoch 00145: loss improved from 0.03464 to 0.03365, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0336 - accuracy: 1.0000
Epoch 146/400
8/8 [==============================] - ETA: 0s - loss: 0.0349 - accuracy: 1.0000
Epoch 00146: loss did not improve from 0.03365
8/8 [==============================] - 1s 66ms/step - loss: 0.0349 - accuracy: 1.0000
Epoch 147/400
8/8 [==============================] - ETA: 0s - loss: 0.0332 - accuracy: 1.0000
Epoch 00147: loss improved from 0.03365 to 0.03321, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 0.0332 - accuracy: 1.0000
Epoch 148/400
8/8 [==============================] - ETA: 0s - loss: 0.0333 - accuracy: 1.0000
Epoch 00148: loss did not improve from 0.03321
8/8 [==============================] - 1s 67ms/step - loss: 0.0333 - accuracy: 1.0000
Epoch 149/400
8/8 [==============================] - ETA: 0s - loss: 0.0331 - accuracy: 1.0000
Epoch 00149: loss improved from 0.03321 to 0.03310, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0331 - accuracy: 1.0000
Epoch 150/400
8/8 [==============================] - ETA: 0s - loss: 0.0350 - accuracy: 1.0000
Epoch 00150: loss did not improve from 0.03310
8/8 [==============================] - 1s 66ms/step - loss: 0.0350 - accuracy: 1.0000
Epoch 151/400
8/8 [==============================] - ETA: 0s - loss: 0.0313 - accuracy: 1.0000
Epoch 00151: loss improved from 0.03310 to 0.03127, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.0313 - accuracy: 1.0000
Epoch 152/400
8/8 [==============================] - ETA: 0s - loss: 0.0340 - accuracy: 1.0000
Epoch 00152: loss did not improve from 0.03127
8/8 [==============================] - 1s 68ms/step - loss: 0.0340 - accuracy: 1.0000
Epoch 153/400
8/8 [==============================] - ETA: 0s - loss: 0.0308 - accuracy: 1.0000
Epoch 00153: loss improved from 0.03127 to 0.03077, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0308 - accuracy: 1.0000
Epoch 154/400
8/8 [==============================] - ETA: 0s - loss: 0.0332 - accuracy: 1.0000
Epoch 00154: loss did not improve from 0.03077
8/8 [==============================] - 1s 67ms/step - loss: 0.0332 - accuracy: 1.0000
Epoch 155/400
8/8 [==============================] - ETA: 0s - loss: 0.0310 - accuracy: 1.0000
Epoch 00155: loss did not improve from 0.03077
8/8 [==============================] - 1s 67ms/step - loss: 0.0310 - accuracy: 1.0000
Epoch 156/400
8/8 [==============================] - ETA: 0s - loss: 0.0309 - accuracy: 1.0000
Epoch 00156: loss did not improve from 0.03077
8/8 [==============================] - 1s 66ms/step - loss: 0.0309 - accuracy: 1.0000
Epoch 157/400
8/8 [==============================] - ETA: 0s - loss: 0.0291 - accuracy: 1.0000
Epoch 00157: loss improved from 0.03077 to 0.02910, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0291 - accuracy: 1.0000
Epoch 158/400
8/8 [==============================] - ETA: 0s - loss: 0.0287 - accuracy: 1.0000
Epoch 00158: loss improved from 0.02910 to 0.02869, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.0287 - accuracy: 1.0000
Epoch 159/400
8/8 [==============================] - ETA: 0s - loss: 0.0317 - accuracy: 1.0000
Epoch 00159: loss did not improve from 0.02869
8/8 [==============================] - 1s 66ms/step - loss: 0.0317 - accuracy: 1.0000
Epoch 160/400
8/8 [==============================] - ETA: 0s - loss: 0.0301 - accuracy: 1.0000
Epoch 00160: loss did not improve from 0.02869
8/8 [==============================] - 1s 67ms/step - loss: 0.0301 - accuracy: 1.0000
Epoch 161/400
8/8 [==============================] - ETA: 0s - loss: 0.0291 - accuracy: 1.0000
Epoch 00161: loss did not improve from 0.02869
8/8 [==============================] - 1s 67ms/step - loss: 0.0291 - accuracy: 1.0000
Epoch 162/400
8/8 [==============================] - ETA: 0s - loss: 0.0271 - accuracy: 1.0000
Epoch 00162: loss improved from 0.02869 to 0.02713, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.0271 - accuracy: 1.0000
Epoch 163/400
8/8 [==============================] - ETA: 0s - loss: 0.0291 - accuracy: 1.0000
Epoch 00163: loss did not improve from 0.02713
8/8 [==============================] - 1s 66ms/step - loss: 0.0291 - accuracy: 1.0000
Epoch 164/400
8/8 [==============================] - ETA: 0s - loss: 0.0293 - accuracy: 1.0000
Epoch 00164: loss did not improve from 0.02713
8/8 [==============================] - 1s 67ms/step - loss: 0.0293 - accuracy: 1.0000
Epoch 165/400
8/8 [==============================] - ETA: 0s - loss: 0.0275 - accuracy: 1.0000
Epoch 00165: loss did not improve from 0.02713
8/8 [==============================] - 1s 68ms/step - loss: 0.0275 - accuracy: 1.0000
Epoch 166/400
8/8 [==============================] - ETA: 0s - loss: 0.0278 - accuracy: 1.0000
Epoch 00166: loss did not improve from 0.02713
8/8 [==============================] - 1s 67ms/step - loss: 0.0278 - accuracy: 1.0000
Epoch 167/400
8/8 [==============================] - ETA: 0s - loss: 0.0265 - accuracy: 1.0000
Epoch 00167: loss improved from 0.02713 to 0.02650, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0265 - accuracy: 1.0000
Epoch 168/400
8/8 [==============================] - ETA: 0s - loss: 0.0261 - accuracy: 1.0000
Epoch 00168: loss improved from 0.02650 to 0.02608, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.0261 - accuracy: 1.0000
Epoch 169/400
8/8 [==============================] - ETA: 0s - loss: 0.0297 - accuracy: 1.0000
Epoch 00169: loss did not improve from 0.02608
8/8 [==============================] - 1s 66ms/step - loss: 0.0297 - accuracy: 1.0000
Epoch 170/400
8/8 [==============================] - ETA: 0s - loss: 0.0264 - accuracy: 1.0000
Epoch 00170: loss did not improve from 0.02608
8/8 [==============================] - 1s 67ms/step - loss: 0.0264 - accuracy: 1.0000
Epoch 171/400
8/8 [==============================] - ETA: 0s - loss: 0.0255 - accuracy: 1.0000
Epoch 00171: loss improved from 0.02608 to 0.02548, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0255 - accuracy: 1.0000
Epoch 172/400
8/8 [==============================] - ETA: 0s - loss: 0.0271 - accuracy: 1.0000
Epoch 00172: loss did not improve from 0.02548
8/8 [==============================] - 1s 66ms/step - loss: 0.0271 - accuracy: 1.0000
Epoch 173/400
8/8 [==============================] - ETA: 0s - loss: 0.0244 - accuracy: 1.0000
Epoch 00173: loss improved from 0.02548 to 0.02442, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 78ms/step - loss: 0.0244 - accuracy: 1.0000
Epoch 174/400
8/8 [==============================] - ETA: 0s - loss: 0.0238 - accuracy: 1.0000
Epoch 00174: loss improved from 0.02442 to 0.02379, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.0238 - accuracy: 1.0000
Epoch 175/400
8/8 [==============================] - ETA: 0s - loss: 0.0239 - accuracy: 1.0000
Epoch 00175: loss did not improve from 0.02379
8/8 [==============================] - 1s 68ms/step - loss: 0.0239 - accuracy: 1.0000
Epoch 176/400
8/8 [==============================] - ETA: 0s - loss: 0.0247 - accuracy: 1.0000
Epoch 00176: loss did not improve from 0.02379
8/8 [==============================] - 1s 68ms/step - loss: 0.0247 - accuracy: 1.0000
Epoch 177/400
8/8 [==============================] - ETA: 0s - loss: 0.0222 - accuracy: 1.0000
Epoch 00177: loss improved from 0.02379 to 0.02223, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.0222 - accuracy: 1.0000
Epoch 178/400
8/8 [==============================] - ETA: 0s - loss: 0.0232 - accuracy: 1.0000
Epoch 00178: loss did not improve from 0.02223
8/8 [==============================] - 1s 66ms/step - loss: 0.0232 - accuracy: 1.0000
Epoch 179/400
8/8 [==============================] - ETA: 0s - loss: 0.0241 - accuracy: 1.0000
Epoch 00179: loss did not improve from 0.02223
8/8 [==============================] - 1s 67ms/step - loss: 0.0241 - accuracy: 1.0000
Epoch 180/400
8/8 [==============================] - ETA: 0s - loss: 0.0247 - accuracy: 1.0000
Epoch 00180: loss did not improve from 0.02223
8/8 [==============================] - 1s 68ms/step - loss: 0.0247 - accuracy: 1.0000
Epoch 181/400
8/8 [==============================] - ETA: 0s - loss: 0.0215 - accuracy: 1.0000
Epoch 00181: loss improved from 0.02223 to 0.02146, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 78ms/step - loss: 0.0215 - accuracy: 1.0000
Epoch 182/400
8/8 [==============================] - ETA: 0s - loss: 0.0217 - accuracy: 1.0000
Epoch 00182: loss did not improve from 0.02146
8/8 [==============================] - 1s 68ms/step - loss: 0.0217 - accuracy: 1.0000
Epoch 183/400
8/8 [==============================] - ETA: 0s - loss: 0.0218 - accuracy: 1.0000
Epoch 00183: loss did not improve from 0.02146
8/8 [==============================] - 1s 68ms/step - loss: 0.0218 - accuracy: 1.0000
Epoch 184/400
8/8 [==============================] - ETA: 0s - loss: 0.0228 - accuracy: 1.0000
Epoch 00184: loss did not improve from 0.02146
8/8 [==============================] - 1s 66ms/step - loss: 0.0228 - accuracy: 1.0000
Epoch 185/400
8/8 [==============================] - ETA: 0s - loss: 0.0218 - accuracy: 1.0000
Epoch 00185: loss did not improve from 0.02146
8/8 [==============================] - 1s 66ms/step - loss: 0.0218 - accuracy: 1.0000
Epoch 186/400
8/8 [==============================] - ETA: 0s - loss: 0.0243 - accuracy: 1.0000
Epoch 00186: loss did not improve from 0.02146
8/8 [==============================] - 1s 66ms/step - loss: 0.0243 - accuracy: 1.0000
Epoch 187/400
8/8 [==============================] - ETA: 0s - loss: 0.0227 - accuracy: 1.0000
Epoch 00187: loss did not improve from 0.02146
8/8 [==============================] - 1s 67ms/step - loss: 0.0227 - accuracy: 1.0000
Epoch 188/400
8/8 [==============================] - ETA: 0s - loss: 0.0214 - accuracy: 1.0000
Epoch 00188: loss improved from 0.02146 to 0.02139, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 0.0214 - accuracy: 1.0000
Epoch 189/400
8/8 [==============================] - ETA: 0s - loss: 0.0216 - accuracy: 1.0000
Epoch 00189: loss did not improve from 0.02139
8/8 [==============================] - 1s 66ms/step - loss: 0.0216 - accuracy: 1.0000
Epoch 190/400
8/8 [==============================] - ETA: 0s - loss: 0.0234 - accuracy: 1.0000
Epoch 00190: loss did not improve from 0.02139
8/8 [==============================] - 1s 67ms/step - loss: 0.0234 - accuracy: 1.0000
Epoch 191/400
8/8 [==============================] - ETA: 0s - loss: 0.0227 - accuracy: 1.0000
Epoch 00191: loss did not improve from 0.02139
8/8 [==============================] - 1s 65ms/step - loss: 0.0227 - accuracy: 1.0000
Epoch 192/400
8/8 [==============================] - ETA: 0s - loss: 0.0217 - accuracy: 1.0000
Epoch 00192: loss did not improve from 0.02139
8/8 [==============================] - 1s 67ms/step - loss: 0.0217 - accuracy: 1.0000
Epoch 193/400
8/8 [==============================] - ETA: 0s - loss: 0.0227 - accuracy: 0.9990
Epoch 00193: loss did not improve from 0.02139
8/8 [==============================] - 1s 67ms/step - loss: 0.0227 - accuracy: 0.9990
Epoch 194/400
8/8 [==============================] - ETA: 0s - loss: 0.0195 - accuracy: 1.0000
Epoch 00194: loss improved from 0.02139 to 0.01952, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 0.0195 - accuracy: 1.0000
Epoch 195/400
8/8 [==============================] - ETA: 0s - loss: 0.0201 - accuracy: 1.0000
Epoch 00195: loss did not improve from 0.01952
8/8 [==============================] - 1s 66ms/step - loss: 0.0201 - accuracy: 1.0000
Epoch 196/400
8/8 [==============================] - ETA: 0s - loss: 0.0218 - accuracy: 1.0000
Epoch 00196: loss did not improve from 0.01952
8/8 [==============================] - 1s 66ms/step - loss: 0.0218 - accuracy: 1.0000
Epoch 197/400
8/8 [==============================] - ETA: 0s - loss: 0.0196 - accuracy: 1.0000
Epoch 00197: loss did not improve from 0.01952
8/8 [==============================] - 1s 66ms/step - loss: 0.0196 - accuracy: 1.0000
Epoch 198/400
8/8 [==============================] - ETA: 0s - loss: 0.0199 - accuracy: 1.0000
Epoch 00198: loss did not improve from 0.01952
8/8 [==============================] - 1s 65ms/step - loss: 0.0199 - accuracy: 1.0000
Epoch 199/400
8/8 [==============================] - ETA: 0s - loss: 0.0187 - accuracy: 1.0000
Epoch 00199: loss improved from 0.01952 to 0.01870, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.0187 - accuracy: 1.0000
Epoch 200/400
8/8 [==============================] - ETA: 0s - loss: 0.0187 - accuracy: 1.0000
Epoch 00200: loss improved from 0.01870 to 0.01868, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0187 - accuracy: 1.0000
Epoch 201/400
8/8 [==============================] - ETA: 0s - loss: 0.0192 - accuracy: 1.0000
Epoch 00201: loss did not improve from 0.01868
8/8 [==============================] - 1s 65ms/step - loss: 0.0192 - accuracy: 1.0000
Epoch 202/400
8/8 [==============================] - ETA: 0s - loss: 0.0175 - accuracy: 1.0000
Epoch 00202: loss improved from 0.01868 to 0.01748, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0175 - accuracy: 1.0000
Epoch 203/400
8/8 [==============================] - ETA: 0s - loss: 0.0181 - accuracy: 1.0000
Epoch 00203: loss did not improve from 0.01748
8/8 [==============================] - 1s 66ms/step - loss: 0.0181 - accuracy: 1.0000
Epoch 204/400
8/8 [==============================] - ETA: 0s - loss: 0.0246 - accuracy: 0.9990
Epoch 00204: loss did not improve from 0.01748
8/8 [==============================] - 1s 64ms/step - loss: 0.0246 - accuracy: 0.9990
Epoch 205/400
8/8 [==============================] - ETA: 0s - loss: 0.0345 - accuracy: 0.9961
Epoch 00205: loss did not improve from 0.01748
8/8 [==============================] - 1s 67ms/step - loss: 0.0345 - accuracy: 0.9961
Epoch 206/400
8/8 [==============================] - ETA: 0s - loss: 0.0234 - accuracy: 1.0000
Epoch 00206: loss did not improve from 0.01748
8/8 [==============================] - 1s 67ms/step - loss: 0.0234 - accuracy: 1.0000
Epoch 207/400
8/8 [==============================] - ETA: 0s - loss: 0.0477 - accuracy: 0.9932
Epoch 00207: loss did not improve from 0.01748
8/8 [==============================] - 1s 65ms/step - loss: 0.0477 - accuracy: 0.9932
Epoch 208/400
8/8 [==============================] - ETA: 0s - loss: 0.0501 - accuracy: 0.9951
Epoch 00208: loss did not improve from 0.01748
8/8 [==============================] - 1s 66ms/step - loss: 0.0501 - accuracy: 0.9951
Epoch 209/400
8/8 [==============================] - ETA: 0s - loss: 0.0415 - accuracy: 0.9922
Epoch 00209: loss did not improve from 0.01748
8/8 [==============================] - 1s 66ms/step - loss: 0.0415 - accuracy: 0.9922
Epoch 210/400
8/8 [==============================] - ETA: 0s - loss: 0.0302 - accuracy: 0.9980
Epoch 00210: loss did not improve from 0.01748
8/8 [==============================] - 1s 66ms/step - loss: 0.0302 - accuracy: 0.9980
Epoch 211/400
8/8 [==============================] - ETA: 0s - loss: 0.0281 - accuracy: 0.9990
Epoch 00211: loss did not improve from 0.01748
8/8 [==============================] - 1s 65ms/step - loss: 0.0281 - accuracy: 0.9990
Epoch 212/400
8/8 [==============================] - ETA: 0s - loss: 0.0251 - accuracy: 1.0000
Epoch 00212: loss did not improve from 0.01748
8/8 [==============================] - 1s 67ms/step - loss: 0.0251 - accuracy: 1.0000
Epoch 213/400
8/8 [==============================] - ETA: 0s - loss: 0.0201 - accuracy: 1.0000
Epoch 00213: loss did not improve from 0.01748
8/8 [==============================] - 1s 67ms/step - loss: 0.0201 - accuracy: 1.0000
Epoch 214/400
8/8 [==============================] - ETA: 0s - loss: 0.0188 - accuracy: 1.0000
Epoch 00214: loss did not improve from 0.01748
8/8 [==============================] - 1s 64ms/step - loss: 0.0188 - accuracy: 1.0000
Epoch 215/400
8/8 [==============================] - ETA: 0s - loss: 0.0192 - accuracy: 1.0000
Epoch 00215: loss did not improve from 0.01748
8/8 [==============================] - 1s 66ms/step - loss: 0.0192 - accuracy: 1.0000
Epoch 216/400
8/8 [==============================] - ETA: 0s - loss: 0.0179 - accuracy: 1.0000
Epoch 00216: loss did not improve from 0.01748
8/8 [==============================] - 1s 67ms/step - loss: 0.0179 - accuracy: 1.0000
Epoch 217/400
8/8 [==============================] - ETA: 0s - loss: 0.0194 - accuracy: 1.0000
Epoch 00217: loss did not improve from 0.01748
8/8 [==============================] - 1s 65ms/step - loss: 0.0194 - accuracy: 1.0000
Epoch 218/400
8/8 [==============================] - ETA: 0s - loss: 0.0183 - accuracy: 1.0000
Epoch 00218: loss did not improve from 0.01748
8/8 [==============================] - 1s 66ms/step - loss: 0.0183 - accuracy: 1.0000
Epoch 219/400
8/8 [==============================] - ETA: 0s - loss: 0.0174 - accuracy: 1.0000
Epoch 00219: loss improved from 0.01748 to 0.01740, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.0174 - accuracy: 1.0000
Epoch 220/400
8/8 [==============================] - ETA: 0s - loss: 0.0170 - accuracy: 1.0000
Epoch 00220: loss improved from 0.01740 to 0.01702, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 0.0170 - accuracy: 1.0000
Epoch 221/400
8/8 [==============================] - ETA: 0s - loss: 0.0169 - accuracy: 1.0000
Epoch 00221: loss improved from 0.01702 to 0.01694, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0169 - accuracy: 1.0000
Epoch 222/400
8/8 [==============================] - ETA: 0s - loss: 0.0158 - accuracy: 1.0000
Epoch 00222: loss improved from 0.01694 to 0.01580, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.0158 - accuracy: 1.0000
Epoch 223/400
8/8 [==============================] - ETA: 0s - loss: 0.0153 - accuracy: 1.0000
Epoch 00223: loss improved from 0.01580 to 0.01531, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.0153 - accuracy: 1.0000
Epoch 224/400
8/8 [==============================] - ETA: 0s - loss: 0.0163 - accuracy: 1.0000
Epoch 00224: loss did not improve from 0.01531
8/8 [==============================] - 1s 67ms/step - loss: 0.0163 - accuracy: 1.0000
Epoch 225/400
8/8 [==============================] - ETA: 0s - loss: 0.0155 - accuracy: 1.0000
Epoch 00225: loss did not improve from 0.01531
8/8 [==============================] - 1s 66ms/step - loss: 0.0155 - accuracy: 1.0000
Epoch 226/400
8/8 [==============================] - ETA: 0s - loss: 0.0149 - accuracy: 1.0000
Epoch 00226: loss improved from 0.01531 to 0.01494, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.0149 - accuracy: 1.0000
Epoch 227/400
8/8 [==============================] - ETA: 0s - loss: 0.0152 - accuracy: 1.0000
Epoch 00227: loss did not improve from 0.01494
8/8 [==============================] - 1s 66ms/step - loss: 0.0152 - accuracy: 1.0000
Epoch 228/400
8/8 [==============================] - ETA: 0s - loss: 0.0148 - accuracy: 1.0000
Epoch 00228: loss improved from 0.01494 to 0.01483, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0148 - accuracy: 1.0000
Epoch 229/400
8/8 [==============================] - ETA: 0s - loss: 0.0142 - accuracy: 1.0000
Epoch 00229: loss improved from 0.01483 to 0.01422, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0142 - accuracy: 1.0000
Epoch 230/400
8/8 [==============================] - ETA: 0s - loss: 0.0148 - accuracy: 1.0000
Epoch 00230: loss did not improve from 0.01422
8/8 [==============================] - 1s 66ms/step - loss: 0.0148 - accuracy: 1.0000
Epoch 231/400
8/8 [==============================] - ETA: 0s - loss: 0.0149 - accuracy: 1.0000
Epoch 00231: loss did not improve from 0.01422
8/8 [==============================] - 1s 67ms/step - loss: 0.0149 - accuracy: 1.0000
Epoch 232/400
8/8 [==============================] - ETA: 0s - loss: 0.0146 - accuracy: 1.0000
Epoch 00232: loss did not improve from 0.01422
8/8 [==============================] - 1s 67ms/step - loss: 0.0146 - accuracy: 1.0000
Epoch 233/400
8/8 [==============================] - ETA: 0s - loss: 0.0139 - accuracy: 1.0000
Epoch 00233: loss improved from 0.01422 to 0.01385, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 0.0139 - accuracy: 1.0000
Epoch 234/400
8/8 [==============================] - ETA: 0s - loss: 0.0134 - accuracy: 1.0000
Epoch 00234: loss improved from 0.01385 to 0.01345, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0134 - accuracy: 1.0000
Epoch 235/400
8/8 [==============================] - ETA: 0s - loss: 0.0145 - accuracy: 1.0000
Epoch 00235: loss did not improve from 0.01345
8/8 [==============================] - 1s 66ms/step - loss: 0.0145 - accuracy: 1.0000
Epoch 236/400
8/8 [==============================] - ETA: 0s - loss: 0.0135 - accuracy: 1.0000
Epoch 00236: loss did not improve from 0.01345
8/8 [==============================] - 1s 64ms/step - loss: 0.0135 - accuracy: 1.0000
Epoch 237/400
8/8 [==============================] - ETA: 0s - loss: 0.0137 - accuracy: 1.0000
Epoch 00237: loss did not improve from 0.01345
8/8 [==============================] - 1s 67ms/step - loss: 0.0137 - accuracy: 1.0000
Epoch 238/400
8/8 [==============================] - ETA: 0s - loss: 0.0138 - accuracy: 1.0000
Epoch 00238: loss did not improve from 0.01345
8/8 [==============================] - 1s 67ms/step - loss: 0.0138 - accuracy: 1.0000
Epoch 239/400
8/8 [==============================] - ETA: 0s - loss: 0.0129 - accuracy: 1.0000
Epoch 00239: loss improved from 0.01345 to 0.01289, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.0129 - accuracy: 1.0000
Epoch 240/400
8/8 [==============================] - ETA: 0s - loss: 0.0142 - accuracy: 1.0000
Epoch 00240: loss did not improve from 0.01289
8/8 [==============================] - 1s 67ms/step - loss: 0.0142 - accuracy: 1.0000
Epoch 241/400
8/8 [==============================] - ETA: 0s - loss: 0.0130 - accuracy: 1.0000
Epoch 00241: loss did not improve from 0.01289
8/8 [==============================] - 1s 67ms/step - loss: 0.0130 - accuracy: 1.0000
Epoch 242/400
8/8 [==============================] - ETA: 0s - loss: 0.0138 - accuracy: 1.0000
Epoch 00242: loss did not improve from 0.01289
8/8 [==============================] - 1s 65ms/step - loss: 0.0138 - accuracy: 1.0000
Epoch 243/400
8/8 [==============================] - ETA: 0s - loss: 0.0139 - accuracy: 1.0000
Epoch 00243: loss did not improve from 0.01289
8/8 [==============================] - 1s 67ms/step - loss: 0.0139 - accuracy: 1.0000
Epoch 244/400
8/8 [==============================] - ETA: 0s - loss: 0.0121 - accuracy: 1.0000
Epoch 00244: loss improved from 0.01289 to 0.01205, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.0121 - accuracy: 1.0000
Epoch 245/400
8/8 [==============================] - ETA: 0s - loss: 0.0132 - accuracy: 1.0000
Epoch 00245: loss did not improve from 0.01205
8/8 [==============================] - 1s 66ms/step - loss: 0.0132 - accuracy: 1.0000
Epoch 246/400
8/8 [==============================] - ETA: 0s - loss: 0.0134 - accuracy: 1.0000
Epoch 00246: loss did not improve from 0.01205
8/8 [==============================] - 1s 65ms/step - loss: 0.0134 - accuracy: 1.0000
Epoch 247/400
8/8 [==============================] - ETA: 0s - loss: 0.0131 - accuracy: 1.0000
Epoch 00247: loss did not improve from 0.01205
8/8 [==============================] - 1s 67ms/step - loss: 0.0131 - accuracy: 1.0000
Epoch 248/400
8/8 [==============================] - ETA: 0s - loss: 0.0125 - accuracy: 1.0000
Epoch 00248: loss did not improve from 0.01205
8/8 [==============================] - 1s 66ms/step - loss: 0.0125 - accuracy: 1.0000
Epoch 249/400
8/8 [==============================] - ETA: 0s - loss: 0.0118 - accuracy: 1.0000
Epoch 00249: loss improved from 0.01205 to 0.01176, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 0.0118 - accuracy: 1.0000
Epoch 250/400
8/8 [==============================] - ETA: 0s - loss: 0.0132 - accuracy: 1.0000
Epoch 00250: loss did not improve from 0.01176
8/8 [==============================] - 1s 67ms/step - loss: 0.0132 - accuracy: 1.0000
Epoch 251/400
8/8 [==============================] - ETA: 0s - loss: 0.0128 - accuracy: 1.0000
Epoch 00251: loss did not improve from 0.01176
8/8 [==============================] - 1s 68ms/step - loss: 0.0128 - accuracy: 1.0000
Epoch 252/400
8/8 [==============================] - ETA: 0s - loss: 0.0128 - accuracy: 1.0000
Epoch 00252: loss did not improve from 0.01176
8/8 [==============================] - 1s 66ms/step - loss: 0.0128 - accuracy: 1.0000
Epoch 253/400
8/8 [==============================] - ETA: 0s - loss: 0.0120 - accuracy: 1.0000
Epoch 00253: loss did not improve from 0.01176
8/8 [==============================] - 1s 67ms/step - loss: 0.0120 - accuracy: 1.0000
Epoch 254/400
8/8 [==============================] - ETA: 0s - loss: 0.0114 - accuracy: 1.0000
Epoch 00254: loss improved from 0.01176 to 0.01140, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.0114 - accuracy: 1.0000
Epoch 255/400
8/8 [==============================] - ETA: 0s - loss: 0.0114 - accuracy: 1.0000
Epoch 00255: loss did not improve from 0.01140
8/8 [==============================] - 1s 67ms/step - loss: 0.0114 - accuracy: 1.0000
Epoch 256/400
8/8 [==============================] - ETA: 0s - loss: 0.0112 - accuracy: 1.0000
Epoch 00256: loss improved from 0.01140 to 0.01125, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0112 - accuracy: 1.0000
Epoch 257/400
8/8 [==============================] - ETA: 0s - loss: 0.0116 - accuracy: 1.0000
Epoch 00257: loss did not improve from 0.01125
8/8 [==============================] - 1s 68ms/step - loss: 0.0116 - accuracy: 1.0000
Epoch 258/400
8/8 [==============================] - ETA: 0s - loss: 0.0122 - accuracy: 1.0000
Epoch 00258: loss did not improve from 0.01125
8/8 [==============================] - 1s 66ms/step - loss: 0.0122 - accuracy: 1.0000
Epoch 259/400
8/8 [==============================] - ETA: 0s - loss: 0.0110 - accuracy: 1.0000
Epoch 00259: loss improved from 0.01125 to 0.01098, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0110 - accuracy: 1.0000
Epoch 260/400
8/8 [==============================] - ETA: 0s - loss: 0.0109 - accuracy: 1.0000
Epoch 00260: loss improved from 0.01098 to 0.01091, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0109 - accuracy: 1.0000
Epoch 261/400
8/8 [==============================] - ETA: 0s - loss: 0.0108 - accuracy: 1.0000
Epoch 00261: loss improved from 0.01091 to 0.01080, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0108 - accuracy: 1.0000
Epoch 262/400
8/8 [==============================] - ETA: 0s - loss: 0.0128 - accuracy: 1.0000
Epoch 00262: loss did not improve from 0.01080
8/8 [==============================] - 1s 68ms/step - loss: 0.0128 - accuracy: 1.0000
Epoch 263/400
8/8 [==============================] - ETA: 0s - loss: 0.0111 - accuracy: 1.0000
Epoch 00263: loss did not improve from 0.01080
8/8 [==============================] - 1s 68ms/step - loss: 0.0111 - accuracy: 1.0000
Epoch 264/400
8/8 [==============================] - ETA: 0s - loss: 0.0103 - accuracy: 1.0000
Epoch 00264: loss improved from 0.01080 to 0.01034, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0103 - accuracy: 1.0000
Epoch 265/400
8/8 [==============================] - ETA: 0s - loss: 0.0118 - accuracy: 1.0000
Epoch 00265: loss did not improve from 0.01034
8/8 [==============================] - 1s 67ms/step - loss: 0.0118 - accuracy: 1.0000
Epoch 266/400
8/8 [==============================] - ETA: 0s - loss: 0.0116 - accuracy: 1.0000
Epoch 00266: loss did not improve from 0.01034
8/8 [==============================] - 1s 68ms/step - loss: 0.0116 - accuracy: 1.0000
Epoch 267/400
8/8 [==============================] - ETA: 0s - loss: 0.0111 - accuracy: 1.0000
Epoch 00267: loss did not improve from 0.01034
8/8 [==============================] - 1s 67ms/step - loss: 0.0111 - accuracy: 1.0000
Epoch 268/400
8/8 [==============================] - ETA: 0s - loss: 0.0098 - accuracy: 1.0000
Epoch 00268: loss improved from 0.01034 to 0.00983, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 78ms/step - loss: 0.0098 - accuracy: 1.0000
Epoch 269/400
8/8 [==============================] - ETA: 0s - loss: 0.0105 - accuracy: 1.0000
Epoch 00269: loss did not improve from 0.00983
8/8 [==============================] - 1s 67ms/step - loss: 0.0105 - accuracy: 1.0000
Epoch 270/400
8/8 [==============================] - ETA: 0s - loss: 0.0110 - accuracy: 1.0000
Epoch 00270: loss did not improve from 0.00983
8/8 [==============================] - 1s 67ms/step - loss: 0.0110 - accuracy: 1.0000
Epoch 271/400
8/8 [==============================] - ETA: 0s - loss: 0.0106 - accuracy: 1.0000
Epoch 00271: loss did not improve from 0.00983
8/8 [==============================] - 1s 67ms/step - loss: 0.0106 - accuracy: 1.0000
Epoch 272/400
8/8 [==============================] - ETA: 0s - loss: 0.0103 - accuracy: 1.0000
Epoch 00272: loss did not improve from 0.00983
8/8 [==============================] - 1s 68ms/step - loss: 0.0103 - accuracy: 1.0000
Epoch 273/400
8/8 [==============================] - ETA: 0s - loss: 0.0102 - accuracy: 1.0000
Epoch 00273: loss did not improve from 0.00983
8/8 [==============================] - 1s 67ms/step - loss: 0.0102 - accuracy: 1.0000
Epoch 274/400
8/8 [==============================] - ETA: 0s - loss: 0.0096 - accuracy: 1.0000
Epoch 00274: loss improved from 0.00983 to 0.00960, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0096 - accuracy: 1.0000
Epoch 275/400
8/8 [==============================] - ETA: 0s - loss: 0.0102 - accuracy: 1.0000
Epoch 00275: loss did not improve from 0.00960
8/8 [==============================] - 1s 68ms/step - loss: 0.0102 - accuracy: 1.0000
Epoch 276/400
8/8 [==============================] - ETA: 0s - loss: 0.0100 - accuracy: 1.0000
Epoch 00276: loss did not improve from 0.00960
8/8 [==============================] - 1s 67ms/step - loss: 0.0100 - accuracy: 1.0000
Epoch 277/400
8/8 [==============================] - ETA: 0s - loss: 0.0095 - accuracy: 1.0000
Epoch 00277: loss improved from 0.00960 to 0.00953, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0095 - accuracy: 1.0000
Epoch 278/400
8/8 [==============================] - ETA: 0s - loss: 0.0098 - accuracy: 1.0000
Epoch 00278: loss did not improve from 0.00953
8/8 [==============================] - 1s 67ms/step - loss: 0.0098 - accuracy: 1.0000
Epoch 279/400
8/8 [==============================] - ETA: 0s - loss: 0.0099 - accuracy: 1.0000
Epoch 00279: loss did not improve from 0.00953
8/8 [==============================] - 1s 67ms/step - loss: 0.0099 - accuracy: 1.0000
Epoch 280/400
8/8 [==============================] - ETA: 0s - loss: 0.0101 - accuracy: 1.0000
Epoch 00280: loss did not improve from 0.00953
8/8 [==============================] - 1s 68ms/step - loss: 0.0101 - accuracy: 1.0000
Epoch 281/400
8/8 [==============================] - ETA: 0s - loss: 0.0102 - accuracy: 1.0000
Epoch 00281: loss did not improve from 0.00953
8/8 [==============================] - 1s 67ms/step - loss: 0.0102 - accuracy: 1.0000
Epoch 282/400
8/8 [==============================] - ETA: 0s - loss: 0.0091 - accuracy: 1.0000
Epoch 00282: loss improved from 0.00953 to 0.00906, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0091 - accuracy: 1.0000
Epoch 283/400
8/8 [==============================] - ETA: 0s - loss: 0.0090 - accuracy: 1.0000
Epoch 00283: loss improved from 0.00906 to 0.00897, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.0090 - accuracy: 1.0000
Epoch 284/400
8/8 [==============================] - ETA: 0s - loss: 0.0089 - accuracy: 1.0000
Epoch 00284: loss improved from 0.00897 to 0.00894, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0089 - accuracy: 1.0000
Epoch 285/400
8/8 [==============================] - ETA: 0s - loss: 0.0093 - accuracy: 1.0000
Epoch 00285: loss did not improve from 0.00894
8/8 [==============================] - 1s 66ms/step - loss: 0.0093 - accuracy: 1.0000
Epoch 286/400
8/8 [==============================] - ETA: 0s - loss: 0.0089 - accuracy: 1.0000
Epoch 00286: loss did not improve from 0.00894
8/8 [==============================] - 1s 66ms/step - loss: 0.0089 - accuracy: 1.0000
Epoch 287/400
8/8 [==============================] - ETA: 0s - loss: 0.0090 - accuracy: 1.0000
Epoch 00287: loss did not improve from 0.00894
8/8 [==============================] - 1s 66ms/step - loss: 0.0090 - accuracy: 1.0000
Epoch 288/400
8/8 [==============================] - ETA: 0s - loss: 0.0089 - accuracy: 1.0000
Epoch 00288: loss improved from 0.00894 to 0.00888, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0089 - accuracy: 1.0000
Epoch 289/400
8/8 [==============================] - ETA: 0s - loss: 0.0092 - accuracy: 1.0000
Epoch 00289: loss did not improve from 0.00888
8/8 [==============================] - 1s 67ms/step - loss: 0.0092 - accuracy: 1.0000
Epoch 290/400
8/8 [==============================] - ETA: 0s - loss: 0.0101 - accuracy: 1.0000
Epoch 00290: loss did not improve from 0.00888
8/8 [==============================] - 1s 65ms/step - loss: 0.0101 - accuracy: 1.0000
Epoch 291/400
8/8 [==============================] - ETA: 0s - loss: 0.0087 - accuracy: 1.0000
Epoch 00291: loss improved from 0.00888 to 0.00868, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0087 - accuracy: 1.0000
Epoch 292/400
8/8 [==============================] - ETA: 0s - loss: 0.0093 - accuracy: 1.0000
Epoch 00292: loss did not improve from 0.00868
8/8 [==============================] - 1s 67ms/step - loss: 0.0093 - accuracy: 1.0000
Epoch 293/400
8/8 [==============================] - ETA: 0s - loss: 0.0087 - accuracy: 1.0000
Epoch 00293: loss improved from 0.00868 to 0.00866, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.0087 - accuracy: 1.0000
Epoch 294/400
8/8 [==============================] - ETA: 0s - loss: 0.0086 - accuracy: 1.0000
Epoch 00294: loss improved from 0.00866 to 0.00856, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.0086 - accuracy: 1.0000
Epoch 295/400
8/8 [==============================] - ETA: 0s - loss: 0.0084 - accuracy: 1.0000
Epoch 00295: loss improved from 0.00856 to 0.00838, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.0084 - accuracy: 1.0000
Epoch 296/400
8/8 [==============================] - ETA: 0s - loss: 0.0092 - accuracy: 1.0000
Epoch 00296: loss did not improve from 0.00838
8/8 [==============================] - 1s 65ms/step - loss: 0.0092 - accuracy: 1.0000
Epoch 297/400
8/8 [==============================] - ETA: 0s - loss: 0.0080 - accuracy: 1.0000
Epoch 00297: loss improved from 0.00838 to 0.00803, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.0080 - accuracy: 1.0000
Epoch 298/400
8/8 [==============================] - ETA: 0s - loss: 0.0086 - accuracy: 1.0000
Epoch 00298: loss did not improve from 0.00803
8/8 [==============================] - 1s 66ms/step - loss: 0.0086 - accuracy: 1.0000
Epoch 299/400
8/8 [==============================] - ETA: 0s - loss: 0.0077 - accuracy: 1.0000
Epoch 00299: loss improved from 0.00803 to 0.00768, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 0.0077 - accuracy: 1.0000
Epoch 300/400
8/8 [==============================] - ETA: 0s - loss: 0.0079 - accuracy: 1.0000
Epoch 00300: loss did not improve from 0.00768
8/8 [==============================] - 1s 66ms/step - loss: 0.0079 - accuracy: 1.0000
Epoch 301/400
8/8 [==============================] - ETA: 0s - loss: 0.0081 - accuracy: 1.0000
Epoch 00301: loss did not improve from 0.00768
8/8 [==============================] - 1s 66ms/step - loss: 0.0081 - accuracy: 1.0000
Epoch 302/400
8/8 [==============================] - ETA: 0s - loss: 0.0078 - accuracy: 1.0000
Epoch 00302: loss did not improve from 0.00768
8/8 [==============================] - 1s 65ms/step - loss: 0.0078 - accuracy: 1.0000
Epoch 303/400
8/8 [==============================] - ETA: 0s - loss: 0.0079 - accuracy: 1.0000
Epoch 00303: loss did not improve from 0.00768
8/8 [==============================] - 1s 67ms/step - loss: 0.0079 - accuracy: 1.0000
Epoch 304/400
8/8 [==============================] - ETA: 0s - loss: 0.0083 - accuracy: 1.0000
Epoch 00304: loss did not improve from 0.00768
8/8 [==============================] - 1s 66ms/step - loss: 0.0083 - accuracy: 1.0000
Epoch 305/400
8/8 [==============================] - ETA: 0s - loss: 0.0081 - accuracy: 1.0000
Epoch 00305: loss did not improve from 0.00768
8/8 [==============================] - 1s 65ms/step - loss: 0.0081 - accuracy: 1.0000
Epoch 306/400
8/8 [==============================] - ETA: 0s - loss: 0.0076 - accuracy: 1.0000
Epoch 00306: loss improved from 0.00768 to 0.00765, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.0076 - accuracy: 1.0000
Epoch 307/400
8/8 [==============================] - ETA: 0s - loss: 0.0078 - accuracy: 1.0000
Epoch 00307: loss did not improve from 0.00765
8/8 [==============================] - 1s 67ms/step - loss: 0.0078 - accuracy: 1.0000
Epoch 308/400
8/8 [==============================] - ETA: 0s - loss: 0.0078 - accuracy: 1.0000
Epoch 00308: loss did not improve from 0.00765
8/8 [==============================] - 1s 67ms/step - loss: 0.0078 - accuracy: 1.0000
Epoch 309/400
8/8 [==============================] - ETA: 0s - loss: 0.0079 - accuracy: 1.0000
Epoch 00309: loss did not improve from 0.00765
8/8 [==============================] - 1s 65ms/step - loss: 0.0079 - accuracy: 1.0000
Epoch 310/400
8/8 [==============================] - ETA: 0s - loss: 0.0078 - accuracy: 1.0000
Epoch 00310: loss did not improve from 0.00765
8/8 [==============================] - 1s 67ms/step - loss: 0.0078 - accuracy: 1.0000
Epoch 311/400
8/8 [==============================] - ETA: 0s - loss: 0.0077 - accuracy: 1.0000
Epoch 00311: loss did not improve from 0.00765
8/8 [==============================] - 1s 67ms/step - loss: 0.0077 - accuracy: 1.0000
Epoch 312/400
8/8 [==============================] - ETA: 0s - loss: 0.0079 - accuracy: 1.0000
Epoch 00312: loss did not improve from 0.00765
8/8 [==============================] - 1s 65ms/step - loss: 0.0079 - accuracy: 1.0000
Epoch 313/400
8/8 [==============================] - ETA: 0s - loss: 0.0074 - accuracy: 1.0000
Epoch 00313: loss improved from 0.00765 to 0.00736, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0074 - accuracy: 1.0000
Epoch 314/400
8/8 [==============================] - ETA: 0s - loss: 0.0073 - accuracy: 1.0000
Epoch 00314: loss improved from 0.00736 to 0.00731, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0073 - accuracy: 1.0000
Epoch 315/400
8/8 [==============================] - ETA: 0s - loss: 0.0071 - accuracy: 1.0000
Epoch 00315: loss improved from 0.00731 to 0.00713, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 0.0071 - accuracy: 1.0000
Epoch 316/400
8/8 [==============================] - ETA: 0s - loss: 0.0074 - accuracy: 1.0000
Epoch 00316: loss did not improve from 0.00713
8/8 [==============================] - 1s 66ms/step - loss: 0.0074 - accuracy: 1.0000
Epoch 317/400
8/8 [==============================] - ETA: 0s - loss: 0.0076 - accuracy: 1.0000
Epoch 00317: loss did not improve from 0.00713
8/8 [==============================] - 1s 66ms/step - loss: 0.0076 - accuracy: 1.0000
Epoch 318/400
8/8 [==============================] - ETA: 0s - loss: 0.0069 - accuracy: 1.0000
Epoch 00318: loss improved from 0.00713 to 0.00689, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 0.0069 - accuracy: 1.0000
Epoch 319/400
8/8 [==============================] - ETA: 0s - loss: 0.0074 - accuracy: 1.0000
Epoch 00319: loss did not improve from 0.00689
8/8 [==============================] - 1s 67ms/step - loss: 0.0074 - accuracy: 1.0000
Epoch 320/400
8/8 [==============================] - ETA: 0s - loss: 0.0069 - accuracy: 1.0000
Epoch 00320: loss improved from 0.00689 to 0.00688, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0069 - accuracy: 1.0000
Epoch 321/400
8/8 [==============================] - ETA: 0s - loss: 0.0078 - accuracy: 1.0000
Epoch 00321: loss did not improve from 0.00688
8/8 [==============================] - 1s 65ms/step - loss: 0.0078 - accuracy: 1.0000
Epoch 322/400
8/8 [==============================] - ETA: 0s - loss: 0.0081 - accuracy: 1.0000
Epoch 00322: loss did not improve from 0.00688
8/8 [==============================] - 1s 66ms/step - loss: 0.0081 - accuracy: 1.0000
Epoch 323/400
8/8 [==============================] - ETA: 0s - loss: 0.0071 - accuracy: 1.0000
Epoch 00323: loss did not improve from 0.00688
8/8 [==============================] - 1s 66ms/step - loss: 0.0071 - accuracy: 1.0000
Epoch 324/400
8/8 [==============================] - ETA: 0s - loss: 0.0069 - accuracy: 1.0000
Epoch 00324: loss improved from 0.00688 to 0.00686, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.0069 - accuracy: 1.0000
Epoch 325/400
8/8 [==============================] - ETA: 0s - loss: 0.0064 - accuracy: 1.0000
Epoch 00325: loss improved from 0.00686 to 0.00641, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 75ms/step - loss: 0.0064 - accuracy: 1.0000
Epoch 326/400
8/8 [==============================] - ETA: 0s - loss: 0.0064 - accuracy: 1.0000
Epoch 00326: loss improved from 0.00641 to 0.00637, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0064 - accuracy: 1.0000
Epoch 327/400
8/8 [==============================] - ETA: 0s - loss: 0.0065 - accuracy: 1.0000
Epoch 00327: loss did not improve from 0.00637
8/8 [==============================] - 1s 66ms/step - loss: 0.0065 - accuracy: 1.0000
Epoch 328/400
8/8 [==============================] - ETA: 0s - loss: 0.0074 - accuracy: 1.0000
Epoch 00328: loss did not improve from 0.00637
8/8 [==============================] - 1s 65ms/step - loss: 0.0074 - accuracy: 1.0000
Epoch 329/400
8/8 [==============================] - ETA: 0s - loss: 0.0070 - accuracy: 1.0000
Epoch 00329: loss did not improve from 0.00637
8/8 [==============================] - 1s 67ms/step - loss: 0.0070 - accuracy: 1.0000
Epoch 330/400
8/8 [==============================] - ETA: 0s - loss: 0.0075 - accuracy: 1.0000
Epoch 00330: loss did not improve from 0.00637
8/8 [==============================] - 1s 67ms/step - loss: 0.0075 - accuracy: 1.0000
Epoch 331/400
8/8 [==============================] - ETA: 0s - loss: 0.0071 - accuracy: 1.0000
Epoch 00331: loss did not improve from 0.00637
8/8 [==============================] - 1s 64ms/step - loss: 0.0071 - accuracy: 1.0000
Epoch 332/400
8/8 [==============================] - ETA: 0s - loss: 0.0071 - accuracy: 1.0000
Epoch 00332: loss did not improve from 0.00637
8/8 [==============================] - 1s 66ms/step - loss: 0.0071 - accuracy: 1.0000
Epoch 333/400
8/8 [==============================] - ETA: 0s - loss: 0.0066 - accuracy: 1.0000
Epoch 00333: loss did not improve from 0.00637
8/8 [==============================] - 1s 66ms/step - loss: 0.0066 - accuracy: 1.0000
Epoch 334/400
8/8 [==============================] - ETA: 0s - loss: 0.0120 - accuracy: 0.9980
Epoch 00334: loss did not improve from 0.00637
8/8 [==============================] - 1s 64ms/step - loss: 0.0120 - accuracy: 0.9980
Epoch 335/400
8/8 [==============================] - ETA: 0s - loss: 0.0066 - accuracy: 1.0000
Epoch 00335: loss did not improve from 0.00637
8/8 [==============================] - 1s 66ms/step - loss: 0.0066 - accuracy: 1.0000
Epoch 336/400
8/8 [==============================] - ETA: 0s - loss: 0.0065 - accuracy: 1.0000
Epoch 00336: loss did not improve from 0.00637
8/8 [==============================] - 1s 66ms/step - loss: 0.0065 - accuracy: 1.0000
Epoch 337/400
8/8 [==============================] - ETA: 0s - loss: 0.0066 - accuracy: 1.0000
Epoch 00337: loss did not improve from 0.00637
8/8 [==============================] - 1s 66ms/step - loss: 0.0066 - accuracy: 1.0000
Epoch 338/400
8/8 [==============================] - ETA: 0s - loss: 0.0076 - accuracy: 1.0000
Epoch 00338: loss did not improve from 0.00637
8/8 [==============================] - 1s 66ms/step - loss: 0.0076 - accuracy: 1.0000
Epoch 339/400
8/8 [==============================] - ETA: 0s - loss: 0.0066 - accuracy: 1.0000
Epoch 00339: loss did not improve from 0.00637
8/8 [==============================] - 1s 67ms/step - loss: 0.0066 - accuracy: 1.0000
Epoch 340/400
8/8 [==============================] - ETA: 0s - loss: 0.0072 - accuracy: 1.0000
Epoch 00340: loss did not improve from 0.00637
8/8 [==============================] - 1s 67ms/step - loss: 0.0072 - accuracy: 1.0000
Epoch 341/400
8/8 [==============================] - ETA: 0s - loss: 0.0150 - accuracy: 0.9980
Epoch 00341: loss did not improve from 0.00637
8/8 [==============================] - 1s 64ms/step - loss: 0.0150 - accuracy: 0.9980
Epoch 342/400
8/8 [==============================] - ETA: 0s - loss: 0.0068 - accuracy: 1.0000
Epoch 00342: loss did not improve from 0.00637
8/8 [==============================] - 1s 67ms/step - loss: 0.0068 - accuracy: 1.0000
Epoch 343/400
8/8 [==============================] - ETA: 0s - loss: 0.0072 - accuracy: 1.0000
Epoch 00343: loss did not improve from 0.00637
8/8 [==============================] - 1s 67ms/step - loss: 0.0072 - accuracy: 1.0000
Epoch 344/400
8/8 [==============================] - ETA: 0s - loss: 0.0095 - accuracy: 0.9990
Epoch 00344: loss did not improve from 0.00637
8/8 [==============================] - 1s 65ms/step - loss: 0.0095 - accuracy: 0.9990
Epoch 345/400
8/8 [==============================] - ETA: 0s - loss: 0.0092 - accuracy: 0.9990
Epoch 00345: loss did not improve from 0.00637
8/8 [==============================] - 1s 67ms/step - loss: 0.0092 - accuracy: 0.9990
Epoch 346/400
8/8 [==============================] - ETA: 0s - loss: 0.0098 - accuracy: 0.9990
Epoch 00346: loss did not improve from 0.00637
8/8 [==============================] - 1s 66ms/step - loss: 0.0098 - accuracy: 0.9990
Epoch 347/400
8/8 [==============================] - ETA: 0s - loss: 0.0117 - accuracy: 0.9990
Epoch 00347: loss did not improve from 0.00637
8/8 [==============================] - 1s 66ms/step - loss: 0.0117 - accuracy: 0.9990
Epoch 348/400
8/8 [==============================] - ETA: 0s - loss: 0.0202 - accuracy: 0.9971
Epoch 00348: loss did not improve from 0.00637
8/8 [==============================] - 1s 65ms/step - loss: 0.0202 - accuracy: 0.9971
Epoch 349/400
8/8 [==============================] - ETA: 0s - loss: 0.0118 - accuracy: 1.0000
Epoch 00349: loss did not improve from 0.00637
8/8 [==============================] - 1s 66ms/step - loss: 0.0118 - accuracy: 1.0000
Epoch 350/400
8/8 [==============================] - ETA: 0s - loss: 0.0142 - accuracy: 0.9980
Epoch 00350: loss did not improve from 0.00637
8/8 [==============================] - 1s 67ms/step - loss: 0.0142 - accuracy: 0.9980
Epoch 351/400
8/8 [==============================] - ETA: 0s - loss: 0.0159 - accuracy: 0.9961
Epoch 00351: loss did not improve from 0.00637
8/8 [==============================] - 1s 64ms/step - loss: 0.0159 - accuracy: 0.9961
Epoch 352/400
8/8 [==============================] - ETA: 0s - loss: 0.0174 - accuracy: 0.9980
Epoch 00352: loss did not improve from 0.00637
8/8 [==============================] - 1s 67ms/step - loss: 0.0174 - accuracy: 0.9980
Epoch 353/400
8/8 [==============================] - ETA: 0s - loss: 0.0205 - accuracy: 0.9971
Epoch 00353: loss did not improve from 0.00637
8/8 [==============================] - 1s 67ms/step - loss: 0.0205 - accuracy: 0.9971
Epoch 354/400
8/8 [==============================] - ETA: 0s - loss: 0.0236 - accuracy: 0.9961
Epoch 00354: loss did not improve from 0.00637
8/8 [==============================] - 1s 65ms/step - loss: 0.0236 - accuracy: 0.9961
Epoch 355/400
8/8 [==============================] - ETA: 0s - loss: 0.0105 - accuracy: 1.0000
Epoch 00355: loss did not improve from 0.00637
8/8 [==============================] - 1s 67ms/step - loss: 0.0105 - accuracy: 1.0000
Epoch 356/400
8/8 [==============================] - ETA: 0s - loss: 0.0092 - accuracy: 1.0000
Epoch 00356: loss did not improve from 0.00637
8/8 [==============================] - 1s 67ms/step - loss: 0.0092 - accuracy: 1.0000
Epoch 357/400
8/8 [==============================] - ETA: 0s - loss: 0.0080 - accuracy: 1.0000
Epoch 00357: loss did not improve from 0.00637
8/8 [==============================] - 1s 66ms/step - loss: 0.0080 - accuracy: 1.0000
Epoch 358/400
8/8 [==============================] - ETA: 0s - loss: 0.0079 - accuracy: 1.0000
Epoch 00358: loss did not improve from 0.00637
8/8 [==============================] - 1s 66ms/step - loss: 0.0079 - accuracy: 1.0000
Epoch 359/400
8/8 [==============================] - ETA: 0s - loss: 0.0066 - accuracy: 1.0000
Epoch 00359: loss did not improve from 0.00637
8/8 [==============================] - 1s 67ms/step - loss: 0.0066 - accuracy: 1.0000
Epoch 360/400
8/8 [==============================] - ETA: 0s - loss: 0.0091 - accuracy: 0.9990
Epoch 00360: loss did not improve from 0.00637
8/8 [==============================] - 1s 67ms/step - loss: 0.0091 - accuracy: 0.9990
Epoch 361/400
8/8 [==============================] - ETA: 0s - loss: 0.0064 - accuracy: 1.0000
Epoch 00361: loss did not improve from 0.00637
8/8 [==============================] - 1s 65ms/step - loss: 0.0064 - accuracy: 1.0000
Epoch 362/400
8/8 [==============================] - ETA: 0s - loss: 0.0063 - accuracy: 1.0000
Epoch 00362: loss improved from 0.00637 to 0.00627, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0063 - accuracy: 1.0000
Epoch 363/400
8/8 [==============================] - ETA: 0s - loss: 0.0064 - accuracy: 1.0000
Epoch 00363: loss did not improve from 0.00627
8/8 [==============================] - 1s 66ms/step - loss: 0.0064 - accuracy: 1.0000
Epoch 364/400
8/8 [==============================] - ETA: 0s - loss: 0.0057 - accuracy: 1.0000
Epoch 00364: loss improved from 0.00627 to 0.00568, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 74ms/step - loss: 0.0057 - accuracy: 1.0000
Epoch 365/400
8/8 [==============================] - ETA: 0s - loss: 0.0066 - accuracy: 1.0000
Epoch 00365: loss did not improve from 0.00568
8/8 [==============================] - 1s 67ms/step - loss: 0.0066 - accuracy: 1.0000
Epoch 366/400
8/8 [==============================] - ETA: 0s - loss: 0.0054 - accuracy: 1.0000
Epoch 00366: loss improved from 0.00568 to 0.00540, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0054 - accuracy: 1.0000
Epoch 367/400
8/8 [==============================] - ETA: 0s - loss: 0.0063 - accuracy: 1.0000
Epoch 00367: loss did not improve from 0.00540
8/8 [==============================] - 1s 64ms/step - loss: 0.0063 - accuracy: 1.0000
Epoch 368/400
8/8 [==============================] - ETA: 0s - loss: 0.0058 - accuracy: 1.0000
Epoch 00368: loss did not improve from 0.00540
8/8 [==============================] - 1s 67ms/step - loss: 0.0058 - accuracy: 1.0000
Epoch 369/400
8/8 [==============================] - ETA: 0s - loss: 0.0056 - accuracy: 1.0000
Epoch 00369: loss did not improve from 0.00540
8/8 [==============================] - 1s 67ms/step - loss: 0.0056 - accuracy: 1.0000
Epoch 370/400
8/8 [==============================] - ETA: 0s - loss: 0.0059 - accuracy: 1.0000
Epoch 00370: loss did not improve from 0.00540
8/8 [==============================] - 1s 65ms/step - loss: 0.0059 - accuracy: 1.0000
Epoch 371/400
8/8 [==============================] - ETA: 0s - loss: 0.0058 - accuracy: 1.0000
Epoch 00371: loss did not improve from 0.00540
8/8 [==============================] - 1s 66ms/step - loss: 0.0058 - accuracy: 1.0000
Epoch 372/400
8/8 [==============================] - ETA: 0s - loss: 0.0055 - accuracy: 1.0000
Epoch 00372: loss did not improve from 0.00540
8/8 [==============================] - 1s 66ms/step - loss: 0.0055 - accuracy: 1.0000
Epoch 373/400
8/8 [==============================] - ETA: 0s - loss: 0.0052 - accuracy: 1.0000
Epoch 00373: loss improved from 0.00540 to 0.00518, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0052 - accuracy: 1.0000
Epoch 374/400
8/8 [==============================] - ETA: 0s - loss: 0.0052 - accuracy: 1.0000
Epoch 00374: loss did not improve from 0.00518
8/8 [==============================] - 1s 66ms/step - loss: 0.0052 - accuracy: 1.0000
Epoch 375/400
8/8 [==============================] - ETA: 0s - loss: 0.0048 - accuracy: 1.0000
Epoch 00375: loss improved from 0.00518 to 0.00481, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.0048 - accuracy: 1.0000
Epoch 376/400
8/8 [==============================] - ETA: 0s - loss: 0.0052 - accuracy: 1.0000
Epoch 00376: loss did not improve from 0.00481
8/8 [==============================] - 1s 67ms/step - loss: 0.0052 - accuracy: 1.0000
Epoch 377/400
8/8 [==============================] - ETA: 0s - loss: 0.0051 - accuracy: 1.0000
Epoch 00377: loss did not improve from 0.00481
8/8 [==============================] - 1s 65ms/step - loss: 0.0051 - accuracy: 1.0000
Epoch 378/400
8/8 [==============================] - ETA: 0s - loss: 0.0050 - accuracy: 1.0000
Epoch 00378: loss did not improve from 0.00481
8/8 [==============================] - 1s 67ms/step - loss: 0.0050 - accuracy: 1.0000
Epoch 379/400
8/8 [==============================] - ETA: 0s - loss: 0.0052 - accuracy: 1.0000
Epoch 00379: loss did not improve from 0.00481
8/8 [==============================] - 1s 67ms/step - loss: 0.0052 - accuracy: 1.0000
Epoch 380/400
8/8 [==============================] - ETA: 0s - loss: 0.0057 - accuracy: 1.0000
Epoch 00380: loss did not improve from 0.00481
8/8 [==============================] - 1s 65ms/step - loss: 0.0057 - accuracy: 1.0000
Epoch 381/400
8/8 [==============================] - ETA: 0s - loss: 0.0046 - accuracy: 1.0000
Epoch 00381: loss improved from 0.00481 to 0.00465, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.0046 - accuracy: 1.0000
Epoch 382/400
8/8 [==============================] - ETA: 0s - loss: 0.0053 - accuracy: 1.0000
Epoch 00382: loss did not improve from 0.00465
8/8 [==============================] - 1s 67ms/step - loss: 0.0053 - accuracy: 1.0000
Epoch 383/400
8/8 [==============================] - ETA: 0s - loss: 0.0052 - accuracy: 1.0000
Epoch 00383: loss did not improve from 0.00465
8/8 [==============================] - 1s 65ms/step - loss: 0.0052 - accuracy: 1.0000
Epoch 384/400
8/8 [==============================] - ETA: 0s - loss: 0.0054 - accuracy: 1.0000
Epoch 00384: loss did not improve from 0.00465
8/8 [==============================] - 1s 67ms/step - loss: 0.0054 - accuracy: 1.0000
Epoch 385/400
8/8 [==============================] - ETA: 0s - loss: 0.0051 - accuracy: 1.0000
Epoch 00385: loss did not improve from 0.00465
8/8 [==============================] - 1s 67ms/step - loss: 0.0051 - accuracy: 1.0000
Epoch 386/400
8/8 [==============================] - ETA: 0s - loss: 0.0050 - accuracy: 1.0000
Epoch 00386: loss did not improve from 0.00465
8/8 [==============================] - 1s 66ms/step - loss: 0.0050 - accuracy: 1.0000
Epoch 387/400
8/8 [==============================] - ETA: 0s - loss: 0.0051 - accuracy: 1.0000
Epoch 00387: loss did not improve from 0.00465
8/8 [==============================] - 1s 66ms/step - loss: 0.0051 - accuracy: 1.0000
Epoch 388/400
8/8 [==============================] - ETA: 0s - loss: 0.0050 - accuracy: 1.0000
Epoch 00388: loss did not improve from 0.00465
8/8 [==============================] - 1s 67ms/step - loss: 0.0050 - accuracy: 1.0000
Epoch 389/400
8/8 [==============================] - ETA: 0s - loss: 0.0046 - accuracy: 1.0000
Epoch 00389: loss improved from 0.00465 to 0.00459, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0046 - accuracy: 1.0000
Epoch 390/400
8/8 [==============================] - ETA: 0s - loss: 0.0050 - accuracy: 1.0000
Epoch 00390: loss did not improve from 0.00459
8/8 [==============================] - 1s 65ms/step - loss: 0.0050 - accuracy: 1.0000
Epoch 391/400
8/8 [==============================] - ETA: 0s - loss: 0.0047 - accuracy: 1.0000
Epoch 00391: loss did not improve from 0.00459
8/8 [==============================] - 1s 66ms/step - loss: 0.0047 - accuracy: 1.0000
Epoch 392/400
8/8 [==============================] - ETA: 0s - loss: 0.0048 - accuracy: 1.0000
Epoch 00392: loss did not improve from 0.00459
8/8 [==============================] - 1s 67ms/step - loss: 0.0048 - accuracy: 1.0000
Epoch 393/400
8/8 [==============================] - ETA: 0s - loss: 0.0051 - accuracy: 1.0000
Epoch 00393: loss did not improve from 0.00459
8/8 [==============================] - 1s 65ms/step - loss: 0.0051 - accuracy: 1.0000
Epoch 394/400
8/8 [==============================] - ETA: 0s - loss: 0.0049 - accuracy: 1.0000
Epoch 00394: loss did not improve from 0.00459
8/8 [==============================] - 1s 67ms/step - loss: 0.0049 - accuracy: 1.0000
Epoch 395/400
8/8 [==============================] - ETA: 0s - loss: 0.0045 - accuracy: 1.0000
Epoch 00395: loss improved from 0.00459 to 0.00448, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 76ms/step - loss: 0.0045 - accuracy: 1.0000
Epoch 396/400
8/8 [==============================] - ETA: 0s - loss: 0.0049 - accuracy: 1.0000
Epoch 00396: loss did not improve from 0.00448
8/8 [==============================] - 1s 65ms/step - loss: 0.0049 - accuracy: 1.0000
Epoch 397/400
8/8 [==============================] - ETA: 0s - loss: 0.0047 - accuracy: 1.0000
Epoch 00397: loss did not improve from 0.00448
8/8 [==============================] - 1s 67ms/step - loss: 0.0047 - accuracy: 1.0000
Epoch 398/400
8/8 [==============================] - ETA: 0s - loss: 0.0044 - accuracy: 1.0000
Epoch 00398: loss improved from 0.00448 to 0.00441, saving model to Joint Spatial-Spectral Attention Network.hdf5
8/8 [==============================] - 1s 77ms/step - loss: 0.0044 - accuracy: 1.0000
Epoch 399/400
8/8 [==============================] - ETA: 0s - loss: 0.0047 - accuracy: 1.0000
Epoch 00399: loss did not improve from 0.00441
8/8 [==============================] - 1s 66ms/step - loss: 0.0047 - accuracy: 1.0000
Epoch 400/400
8/8 [==============================] - ETA: 0s - loss: 0.0050 - accuracy: 1.0000
Epoch 00400: loss did not improve from 0.00441
8/8 [==============================] - 1s 66ms/step - loss: 0.0050 - accuracy: 1.0000
classification = classification_report(np.argmax(y_test, axis=1), np.argmax(y_pred, axis=1),)
print(classification)
precision recall f1-score support
0 0.96 0.66 0.78 41
1 0.94 0.97 0.95 1285
2 0.96 0.87 0.92 747
3 0.87 0.97 0.92 213
4 0.95 0.97 0.96 435
5 0.97 0.94 0.96 657
6 0.83 0.80 0.82 25
7 1.00 1.00 1.00 430
8 1.00 0.89 0.94 18
9 0.93 0.94 0.93 875
10 0.97 0.98 0.97 2210
11 0.93 0.93 0.93 534
12 0.89 0.95 0.92 185
13 0.99 1.00 0.99 1139
14 0.97 0.92 0.94 347
15 0.86 0.88 0.87 84
accuracy 0.96 9225
macro avg 0.94 0.92 0.93 9225
weighted avg 0.96 0.96 0.96 9225
def AA_andEachClassAccuracy(confusion_matrix):
counter = confusion_matrix.shape[0]
list_diag = np.diag(confusion_matrix) #擷取confusion_matrix的主對角線所有數值
list_raw_sum = np.sum(confusion_matrix, axis=1) #将主對角線所有數求和
each_acc = np.nan_to_num(truediv(list_diag, list_raw_sum)) #list_diag/list_raw_sum 對角線各個數字/對角線所有數字的總和
average_acc = np.mean(each_acc)
return each_acc, average_acc
def reports (x_test,y_test):
#start = time.time()
Y_pred = model.predict(x_test)
y_pred = np.argmax(Y_pred, axis=1)
#end = time.time()
#print(end - start)
target_names = ['Alfalfa', 'Corn-notill', 'Corn-mintill', 'Corn',
'Grass-pasture', 'Grass-trees', 'Grass-pasture-mowed',
'Hay-windrowed', 'Oats', 'Soybean-notill', 'Soybean-mintill',
'Soybean-clean', 'Wheat', 'Woods', 'Buildings-Grass-Trees-Drives',
'Stone-Steel-Towers']
classification = classification_report(np.argmax(y_test, axis=1), y_pred, target_names=target_names)
oa = accuracy_score(np.argmax(y_test, axis=1), y_pred) #計算OA
confusion = confusion_matrix(np.argmax(y_test, axis=1), y_pred) #計算confusion
each_acc, aa = AA_andEachClassAccuracy(confusion) #計算each_acc和aa
kappa = cohen_kappa_score(np.argmax(y_test, axis=1), y_pred) #計算kappa
score = model.evaluate(x_test, y_test, batch_size=32)
Test_Loss = score[0]*100
Test_accuracy = score[1]*100
return classification, confusion, Test_Loss, Test_accuracy, oa*100, each_acc*100, aa*100, kappa*100
classification, confusion, Test_loss, Test_accuracy, oa, each_acc, aa, kappa = reports(x_test,y_test)
classification = str(classification)
confusion = str(confusion)
file_name = "Joint Spatial-Spectral Attention Network_report.txt"
with open(file_name, 'w') as x_file:
x_file.write('{} Test loss (%)'.format(Test_loss))
x_file.write('\n')
x_file.write('{} Test accuracy (%)'.format(Test_accuracy))
x_file.write('\n')
x_file.write('\n')
x_file.write('{} Kappa accuracy (%)'.format(kappa))
x_file.write('\n')
x_file.write('{} Overall accuracy (%)'.format(oa))
x_file.write('\n')
x_file.write('{} Average accuracy (%)'.format(aa))
x_file.write('\n')
x_file.write('\n')
x_file.write('{}'.format(classification))
x_file.write('\n')
x_file.write('{}'.format(confusion))
289/289 [==============================] - 2s 6ms/step - loss: 0.1687 - accuracy: 0.9558
16.86648577451706 Test loss (%)
95.5772340297699 Test accuracy (%)
94.95464400836495 Kappa accuracy (%)
95.57723577235772 Overall accuracy (%)
91.60848455559136 Average accuracy (%)
precision recall f1-score support
Alfalfa 0.96 0.66 0.78 41
Corn-notill 0.94 0.97 0.95 1285
Corn-mintill 0.96 0.87 0.92 747
Corn 0.87 0.97 0.92 213
Grass-pasture 0.95 0.97 0.96 435
Grass-trees 0.97 0.94 0.96 657
Grass-pasture-mowed 0.83 0.80 0.82 25
Hay-windrowed 1.00 1.00 1.00 430
Oats 1.00 0.89 0.94 18
Soybean-notill 0.93 0.94 0.93 875
Soybean-mintill 0.97 0.98 0.97 2210
Soybean-clean 0.93 0.93 0.93 534
Wheat 0.89 0.95 0.92 185
Woods 0.99 1.00 0.99 1139
Buildings-Grass-Trees-Drives 0.97 0.92 0.94 347
Stone-Steel-Towers 0.86 0.88 0.87 84
accuracy 0.96 9225
macro avg 0.94 0.92 0.93 9225
weighted avg 0.96 0.96 0.96 9225
[[ 27 3 0 1 0 0 2 0 0 8 0 0 0 0
0 0]
[ 1 1245 10 2 0 8 0 0 0 7 12 0 0 0
0 0]
[ 0 20 652 14 15 0 0 0 0 0 4 22 20 0
0 0]
[ 0 4 1 206 0 2 0 0 0 0 0 0 0 0
0 0]
[ 0 0 3 0 421 0 2 0 0 0 6 0 2 0
1 0]
[ 0 4 2 4 2 618 0 0 0 0 27 0 0 0
0 0]
[ 0 0 0 0 5 0 20 0 0 0 0 0 0 0
0 0]
[ 0 0 0 0 0 0 0 430 0 0 0 0 0 0
0 0]
[ 0 0 0 0 1 1 0 0 16 0 0 0 0 0
0 0]
[ 0 34 0 0 0 0 0 0 0 820 12 1 0 4
4 0]
[ 0 15 0 0 0 4 0 0 0 24 2159 0 0 2
6 0]
[ 0 0 2 8 1 1 0 0 0 21 0 495 0 0
0 6]
[ 0 0 6 0 0 0 0 0 0 0 3 0 176 0
0 0]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 1139
0 0]
[ 0 1 0 1 0 0 0 0 0 5 0 6 0 9
319 6]
[ 0 0 0 0 0 0 0 0 0 0 0 10 0 0
0 74]]
![TestLink導出用例轉換工具(XML2Excel)[圖]](data:image/gif;base64,R0lGODlhAQABAIAAAP///wAAACwAAAAAAQABAAACAkQBADs=)