這篇介紹如何用資料對vgg16進行訓練

Finetuning最重要的一個步驟就是模型的重新訓練與儲存。  首先對於模型的值的輸出，在類中已經做了定義，因此只需要將定義的模型類初始化後輸出賦予一個特定的變數即可。

vgg = model.vgg16(x_imgs)
fc3_cat_and_dog = vgg.probs
loss = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=fc3_cat_and_dog, labels=y_imgs))
optimizer = tf.train.GradientDescentOptimizer(learning_rate=0.001).minimize(loss)
 

這裡同時定義了損失函式已經最小化方法，完整程式碼如下：

完整程式碼 train.py檔案

import os
os.environ['TF_CPP_MIN_LOG_LEVEL']='2'

import tensorflow as tf
import VGG16_model as model
import create_and_read_TFRecord2 as reader2

if __name__ == '__main__':

    X_train, y_train = reader2.get_file('./train')
    image_batch, label_batch = reader2.get_batch(X_train, y_train, 224, 224, 25, 256)

    x_imgs = tf.placeholder(tf.float32, [None, 224, 224, 3])
    y_imgs = tf.placeholder(tf.int32, [None, 2])

    vgg = model.vgg16(x_imgs)
    fc3_cat_and_dog = vgg.probs
    loss = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=fc3_cat_and_dog, labels=y_imgs))
    optimizer = tf.train.GradientDescentOptimizer(learning_rate=0.001).minimize(loss)

    correct_prediction = tf.equal(tf.arg_max(y_imgs, 1), tf.arg_max(fc3_cat_and_dog, 1))
    accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))

    sess = tf.Session()
    sess.run(tf.global_variables_initializer())
    vgg.load_weights('./vgg16_weights.npz', sess)
    saver = vgg.saver()

    coord = tf.train.Coordinator()
    threads = tf.train.start_queue_runners(coord=coord, sess=sess)

    import time
    start_time = time.time()

    for i in range(200):
        image, label = sess.run([image_batch, label_batch])
        labels = reader2.onehot(label)
        sess.run(optimizer, feed_dict={x_imgs: image, y_imgs: labels})
        loss_record = sess.run(loss, feed_dict={x_imgs: image, y_imgs: labels})
        accuracy_record = sess.run(accuracy, feed_dict={x_imgs: image, y_imgs: labels})
        print("the loss is %f " % loss_record, "the accuracy is %f" % accuracy_record)
        end_time = time.time()
        print('time: ', (end_time - start_time))
        start_time = end_time
        print("----------epoch %d is finished---------------" % i)

    saver.save(sess, "./model/")
    print("Optimization Finished!")


 

在訓練函式中使用了Tensorflow的佇列方式進行資料輸入，而對於權重的重新載入也使用的是前面文章類似的方式，最終資料進行200次迭代，儲存模型在model資料夾中。

訓練的結果如下：

C:\Users\303\Anaconda3\python.exe C:/Users/303/Desktop/vgg/train.py -----------all done--------------- the loss is 0.758111  the accuracy is 0.520000 time:  93.22740054130554 ----------epoch 0 is finished--------------- the loss is 0.700010  the accuracy is 0.400000 time:  37.96240019798279 ----------epoch 1 is finished--------------- the loss is 0.661964  the accuracy is 0.480000 time:  27.612000226974487 ----------epoch 2 is finished--------------- the loss is 0.696854  the accuracy is 0.440000 time:  27.752400159835815 ----------epoch 3 is finished--------------- the loss is 0.692208  the accuracy is 0.560000 time:  27.409200191497803 ----------epoch 4 is finished--------------- the loss is 0.695500  the accuracy is 0.600000 time:  27.69000005722046 ----------epoch 5 is finished--------------- the loss is 0.690480  the accuracy is 0.440000 time:  27.346800565719604 ----------epoch 6 is finished--------------- the loss is 0.673299  the accuracy is 0.400000 time:  27.299999952316284 ----------epoch 7 is finished--------------- the loss is 0.685414  the accuracy is 0.560000 time:  27.159600257873535 ----------epoch 8 is finished--------------- the loss is 0.678527  the accuracy is 0.560000 time:  27.36240005493164 ----------epoch 9 is finished--------------- the loss is 0.703208  the accuracy is 0.560000 time:  27.300000190734863 ----------epoch 10 is finished--------------- the loss is 0.678527  the accuracy is 0.440000 time:  28.25160026550293 ----------epoch 11 is finished--------------- the loss is 0.693578  the accuracy is 0.680000 time:  28.14240026473999 ----------epoch 12 is finished--------------- the loss is 0.688089  the accuracy is 0.560000 time:  28.25160002708435 ----------epoch 13 is finished--------------- the loss is 0.691666  the accuracy is 0.520000 time:  27.93560028076172 ----------epoch 14 is finished--------------- the loss is 0.695733  the accuracy is 0.360000 time:  27.705600261688232 ----------epoch 15 is finished--------------- the loss is 0.676656  the accuracy is 0.560000 time:  27.76800012588501 ----------epoch 16 is finished--------------- the loss is 0.680016  the accuracy is 0.760000 time:  27.783600091934204 ----------epoch 17 is finished--------------- the loss is 0.675143  the accuracy is 0.680000 time:  27.82360029220581 ----------epoch 18 is finished--------------- the loss is 0.674448  the accuracy is 0.400000 time:  27.721199989318848 ----------epoch 19 is finished--------------- the loss is 0.683596  the accuracy is 0.640000 time:  27.556400060653687 ----------epoch 20 is finished--------------- the loss is 0.685636  the accuracy is 0.760000 time:  27.816800355911255 ----------epoch 21 is finished--------------- the loss is 0.667558  the accuracy is 0.560000 time:  27.721200227737427 ----------epoch 22 is finished--------------- the loss is 0.647418  the accuracy is 0.720000 time:  27.87720012664795 ----------epoch 23 is finished--------------- the loss is 0.688498  the accuracy is 0.680000 time:  27.861600160598755 ----------epoch 24 is finished--------------- the loss is 0.693147  the accuracy is 0.680000 time:  27.80120015144348 ----------epoch 25 is finished--------------- the loss is 0.672804  the accuracy is 0.600000 time:  27.6326003074646 ----------epoch 26 is finished--------------- the loss is 0.670962  the accuracy is 0.640000 time:  27.752400159835815 ----------epoch 27 is finished--------------- the loss is 0.689062  the accuracy is 0.600000 time:  27.666800022125244 ----------epoch 28 is finished--------------- the loss is 0.673509  the accuracy is 0.640000 time:  27.675600290298462 ----------epoch 29 is finished--------------- the loss is 0.654030  the accuracy is 0.760000 time:  27.58080005645752 ----------epoch 30 is finished--------------- the loss is 0.602225  the accuracy is 0.640000 time:  27.783600330352783 ----------epoch 31 is finished--------------- the loss is 0.611574  the accuracy is 0.640000 time:  27.600600242614746 ----------epoch 32 is finished--------------- the loss is 0.637754  the accuracy is 0.720000 time:  27.720799922943115 ----------epoch 33 is finished--------------- the loss is 0.639839  the accuracy is 0.720000 time:  27.215200185775757 ----------epoch 34 is finished--------------- the loss is 0.548652  the accuracy is 0.640000 time:  27.306600332260132 ----------epoch 35 is finished--------------- the loss is 0.648707  the accuracy is 0.640000 time:  27.30780005455017 ----------epoch 36 is finished--------------- the loss is 0.673585  the accuracy is 0.760000 time:  27.269400358200073 ----------epoch 37 is finished--------------- the loss is 0.567087  the accuracy is 0.800000 time:  27.429999828338623 ----------epoch 38 is finished--------------- the loss is 0.652346  the accuracy is 0.920000 time:  27.326200246810913 ----------epoch 39 is finished--------------- the loss is 0.580749  the accuracy is 0.680000 time:  28.331400156021118 ----------epoch 40 is finished--------------- the loss is 0.619812  the accuracy is 0.760000 time:  29.592000007629395 ----------epoch 41 is finished--------------- the loss is 0.676218  the accuracy is 0.680000 time:  29.66480040550232 ----------epoch 42 is finished--------------- the loss is 0.523080  the accuracy is 0.760000 time:  30.444000244140625 ----------epoch 43 is finished--------------- the loss is 0.636938  the accuracy is 0.680000 time:  29.430800199508667 ----------epoch 44 is finished--------------- the loss is 0.565380  the accuracy is 0.720000 time:  29.14680004119873 ----------epoch 45 is finished--------------- the loss is 0.607241  the accuracy is 0.760000 time:  29.552000284194946 ----------epoch 46 is finished--------------- the loss is 0.668076  the accuracy is 0.720000 time:  30.087400197982788 ----------epoch 47 is finished--------------- the loss is 0.615282  the accuracy is 0.760000 time:  29.361200094223022 ----------epoch 48 is finished--------------- the loss is 0.572243  the accuracy is 0.680000 time:  28.427799940109253 ----------epoch 49 is finished--------------- the loss is 0.626822  the accuracy is 0.720000 time:  29.176600217819214 ----------epoch 50 is finished--------------- the loss is 0.600773  the accuracy is 0.720000 time:  28.78540015220642 ----------epoch 51 is finished--------------- the loss is 0.573486  the accuracy is 0.720000 time:  28.928600311279297 ----------epoch 52 is finished--------------- the loss is 0.598439  the accuracy is 0.680000 time:  28.34980010986328 ----------epoch 53 is finished--------------- the loss is 0.575304  the accuracy is 0.800000 time:  28.56060028076172 ----------epoch 54 is finished--------------- the loss is 0.449136  the accuracy is 0.760000 time:  27.85759997367859 ----------epoch 55 is finished--------------- the loss is 0.584953  the accuracy is 0.720000 time:  27.222000122070312 ----------epoch 56 is finished--------------- the loss is 0.534731  the accuracy is 0.800000 time:  27.420000314712524 ----------epoch 57 is finished--------------- the loss is 0.523127  the accuracy is 0.720000 time:  27.40619993209839 ----------epoch 58 is finished--------------- the loss is 0.579338  the accuracy is 0.680000 time:  27.410800218582153 ----------epoch 59 is finished--------------- the loss is 0.533000  the accuracy is 0.760000 time:  27.298800230026245 ----------epoch 60 is finished--------------- the loss is 0.610720  the accuracy is 0.800000 time:  27.620800256729126 ----------epoch 61 is finished--------------- the loss is 0.626162  the accuracy is 0.760000 time:  27.49239993095398 ----------epoch 62 is finished--------------- the loss is 0.589299  the accuracy is 0.840000 time:  27.429200172424316 ----------epoch 63 is finished--------------- the loss is 0.496789  the accuracy is 0.760000 time:  27.400400161743164 ----------epoch 64 is finished--------------- the loss is 0.569393  the accuracy is 0.680000 time:  27.337600231170654 ----------epoch 65 is finished--------------- the loss is 0.448170  the accuracy is 0.840000 time:  27.45860004425049 ----------epoch 66 is finished--------------- the loss is 0.587917  the accuracy is 0.680000 time:  27.365400314331055 ----------epoch 67 is finished--------------- the loss is 0.554151  the accuracy is 0.880000 time:  27.627599954605103 ----------epoch 68 is finished--------------- the loss is 0.635504  the accuracy is 0.920000 time:  27.450200080871582 ----------epoch 69 is finished--------------- the loss is 0.526137  the accuracy is 0.800000 time:  27.54040026664734 ----------epoch 70 is finished--------------- the loss is 0.487640  the accuracy is 0.800000 time:  27.451200246810913 ----------epoch 71 is finished--------------- the loss is 0.494419  the accuracy is 0.760000 time:  27.231199979782104 ----------epoch 72 is finished--------------- the loss is 0.601800  the accuracy is 0.840000 time:  27.487000226974487 ----------epoch 73 is finished--------------- the loss is 0.414525  the accuracy is 0.720000 time:  27.597599983215332 ----------epoch 74 is finished--------------- the loss is 0.530127  the accuracy is 0.880000 time:  27.871800422668457 ----------epoch 75 is finished--------------- the loss is 0.472716  the accuracy is 0.760000 time:  27.755200147628784 ----------epoch 76 is finished--------------- the loss is 0.581156  the accuracy is 0.760000 time:  27.348999977111816 ----------epoch 77 is finished--------------- the loss is 0.588855  the accuracy is 0.840000 time:  27.505800247192383 ----------epoch 78 is finished--------------- the loss is 0.442660  the accuracy is 0.800000 time:  27.877000093460083 ----------epoch 79 is finished--------------- the loss is 0.513920  the accuracy is 0.760000 time:  27.692400217056274 ----------epoch 80 is finished--------------- the loss is 0.570069  the accuracy is 0.800000 time:  27.851400136947632 ----------epoch 81 is finished--------------- the loss is 0.574013  the accuracy is 0.720000 time:  27.560199975967407 ----------epoch 82 is finished--------------- the loss is 0.621346  the accuracy is 0.960000 time:  28.02780032157898 ----------epoch 83 is finished--------------- the loss is 0.491623  the accuracy is 0.800000 time:  27.84940004348755 ----------epoch 84 is finished--------------- the loss is 0.549356  the accuracy is 0.680000 time:  27.50480031967163 ----------epoch 85 is finished--------------- the loss is 0.605569  the accuracy is 0.720000 time:  27.891799926757812 ----------epoch 86 is finished--------------- the loss is 0.489809  the accuracy is 0.720000 time:  27.671600341796875 ----------epoch 87 is finished--------------- the loss is 0.548903  the accuracy is 0.760000 time:  27.834800004959106 ----------epoch 88 is finished--------------- the loss is 0.557381  the accuracy is 0.800000 time:  27.87980031967163 ----------epoch 89 is finished--------------- the loss is 0.614677  the accuracy is 0.720000 time:  27.79800009727478 ----------epoch 90 is finished--------------- the loss is 0.487653  the accuracy is 0.920000 time:  27.67300033569336 ----------epoch 91 is finished--------------- the loss is 0.439404  the accuracy is 0.720000 time:  27.524999856948853 ----------epoch 92 is finished--------------- the loss is 0.514191  the accuracy is 0.760000 time:  27.502800464630127 ----------epoch 93 is finished--------------- the loss is 0.608245  the accuracy is 0.880000 time:  27.73099994659424 ----------epoch 94 is finished--------------- the loss is 0.351427  the accuracy is 0.840000 time:  27.83899998664856 ----------epoch 95 is finished--------------- the loss is 0.643810  the accuracy is 0.800000 time:  27.814600229263306 ----------epoch 96 is finished--------------- the loss is 0.382391  the accuracy is 0.880000 time:  27.59920024871826 ----------epoch 97 is finished--------------- the loss is 0.514393  the accuracy is 0.800000 time:  27.685400247573853 ----------epoch 98 is finished--------------- the loss is 0.389810  the accuracy is 0.800000 time:  28.022000074386597 ----------epoch 99 is finished--------------- the loss is 0.455801  the accuracy is 0.760000 time:  27.57540011405945 ----------epoch 100 is finished--------------- the loss is 0.419581  the accuracy is 0.800000 time:  27.84120011329651 ----------epoch 101 is finished--------------- the loss is 0.471398  the accuracy is 0.840000 time:  27.59599995613098 ----------epoch 102 is finished--------------- the loss is 0.456360  the accuracy is 0.840000 time:  27.46400022506714 ----------epoch 103 is finished--------------- the loss is 0.528342  the accuracy is 0.800000 time:  27.807400226593018 ----------epoch 104 is finished--------------- the loss is 0.523615  the accuracy is 0.880000 time:  27.818000316619873 ----------epoch 105 is finished--------------- the loss is 0.467946  the accuracy is 0.880000 time:  27.59220004081726 ----------epoch 106 is finished--------------- the loss is 0.557764  the accuracy is 0.840000 time:  27.674400329589844 ----------epoch 107 is finished--------------- the loss is 0.572827  the accuracy is 0.920000 time:  27.546200037002563 ----------epoch 108 is finished--------------- the loss is 0.506509  the accuracy is 0.720000 time:  27.511000156402588 ----------epoch 109 is finished--------------- the loss is 0.483258  the accuracy is 0.760000 time:  27.842000246047974 ----------epoch 110 is finished--------------- the loss is 0.588461  the accuracy is 0.840000 time:  27.560999870300293 ----------epoch 111 is finished--------------- the loss is 0.395232  the accuracy is 0.680000 time:  28.07260036468506 ----------epoch 112 is finished--------------- the loss is 0.573396  the accuracy is 0.880000 time:  27.79580020904541 ----------epoch 113 is finished--------------- the loss is 0.378605  the accuracy is 0.920000 time:  27.718600034713745 ----------epoch 114 is finished--------------- the loss is 0.516754  the accuracy is 0.920000 time:  27.76640009880066 ----------epoch 115 is finished--------------- the loss is 0.486750  the accuracy is 0.920000 time:  27.51260018348694 ----------epoch 116 is finished--------------- the loss is 0.449040  the accuracy is 0.880000 time:  27.76200032234192 ----------epoch 117 is finished--------------- the loss is 0.564241  the accuracy is 0.920000 time:  27.56940007209778 ----------epoch 118 is finished--------------- the loss is 0.476609  the accuracy is 0.880000 time:  27.84000015258789 ----------epoch 119 is finished--------------- the loss is 0.442881  the accuracy is 0.840000 time:  27.701200246810913 ----------epoch 120 is finished--------------- the loss is 0.410049  the accuracy is 0.840000 time:  27.618799924850464 ----------epoch 121 is finished--------------- the loss is 0.486768  the accuracy is 0.840000 time:  27.57740020751953 ----------epoch 122 is finished--------------- the loss is 0.438721  the accuracy is 0.800000 time:  27.670600175857544 ----------epoch 123 is finished--------------- the loss is 0.555945  the accuracy is 0.960000 time:  27.57260012626648 ----------epoch 124 is finished--------------- the loss is 0.376361  the accuracy is 0.840000 time:  27.803800344467163 ----------epoch 125 is finished--------------- the loss is 0.384597  the accuracy is 0.880000 time:  27.729599952697754 ----------epoch 126 is finished--------------- the loss is 0.393353  the accuracy is 0.880000 time:  27.59500026702881 ----------epoch 127 is finished--------------- the loss is 0.582737  the accuracy is 0.920000 time:  27.850000143051147 ----------epoch 128 is finished--------------- the loss is 0.530217  the accuracy is 0.880000 time:  27.79579997062683 ----------epoch 129 is finished--------------- the loss is 0.505465  the accuracy is 0.920000 time:  27.616400241851807 ----------epoch 130 is finished--------------- the loss is 0.454727  the accuracy is 0.840000 time:  27.714800357818604 ----------epoch 131 is finished--------------- the loss is 0.478746  the accuracy is 0.760000 time:  27.530200004577637 ----------epoch 132 is finished--------------- the loss is 0.460549  the accuracy is 0.880000 time:  27.57480025291443 ----------epoch 133 is finished--------------- the loss is 0.549542  the accuracy is 0.880000 time:  27.84220004081726 ----------epoch 134 is finished--------------- the loss is 0.519628  the accuracy is 0.880000 time:  27.43340015411377 ----------epoch 135 is finished--------------- the loss is 0.546381  the accuracy is 0.880000 time:  27.611600160598755 ----------epoch 136 is finished--------------- the loss is 0.564635  the accuracy is 0.840000 time:  27.526400089263916 ----------epoch 137 is finished--------------- the loss is 0.365546  the accuracy is 0.920000 time:  27.882800340652466 ----------epoch 138 is finished--------------- the loss is 0.487230  the accuracy is 0.800000 time:  27.65059995651245 ----------epoch 139 is finished--------------- the loss is 0.522421  the accuracy is 0.880000 time:  27.713800191879272 ----------epoch 140 is finished--------------- the loss is 0.417020  the accuracy is 0.800000 time:  27.84060001373291 ----------epoch 141 is finished--------------- the loss is 0.319831  the accuracy is 0.880000 time:  27.86180019378662 ----------epoch 142 is finished--------------- the loss is 0.477082  the accuracy is 0.760000 time:  27.462600231170654 ----------epoch 143 is finished--------------- the loss is 0.510756  the accuracy is 0.840000 time:  27.606200218200684 ----------epoch 144 is finished--------------- the loss is 0.458786  the accuracy is 0.880000 time:  27.80500030517578 ----------epoch 145 is finished--------------- the loss is 0.472080  the accuracy is 0.840000 time:  27.635599851608276 ----------epoch 146 is finished--------------- the loss is 0.584948  the accuracy is 0.840000 time:  27.65820026397705 ----------epoch 147 is finished--------------- the loss is 0.492688  the accuracy is 0.920000 time:  27.647400379180908 ----------epoch 148 is finished--------------- the loss is 0.465814  the accuracy is 0.960000 time:  27.7221999168396 ----------epoch 149 is finished--------------- the loss is 0.629802  the accuracy is 0.840000 time:  27.763800144195557 ----------epoch 150 is finished--------------- the loss is 0.566602  the accuracy is 0.800000 time:  27.61579990386963 ----------epoch 151 is finished--------------- the loss is 0.480911  the accuracy is 0.800000 time:  27.80180048942566 ----------epoch 152 is finished--------------- the loss is 0.550858  the accuracy is 0.800000 time:  27.554399967193604 ----------epoch 153 is finished--------------- the loss is 0.472405  the accuracy is 0.920000 time:  27.86620020866394 ----------epoch 154 is finished--------------- the loss is 0.555857  the accuracy is 0.840000 time:  27.531800270080566 ----------epoch 155 is finished--------------- the loss is 0.544163  the accuracy is 0.920000 time:  27.48580002784729 ----------epoch 156 is finished--------------- the loss is 0.447131  the accuracy is 0.800000 time:  27.74880027770996 ----------epoch 157 is finished--------------- the loss is 0.512945  the accuracy is 0.800000 time:  27.713599920272827 ----------epoch 158 is finished--------------- the loss is 0.422867  the accuracy is 0.880000 time:  27.759000301361084 ----------epoch 159 is finished--------------- the loss is 0.451860  the accuracy is 0.880000 time:  27.888200283050537 ----------epoch 160 is finished--------------- the loss is 0.561305  the accuracy is 0.800000 time:  27.659800052642822 ----------epoch 161 is finished--------------- the loss is 0.420359  the accuracy is 0.840000 time:  27.467600107192993 ----------epoch 162 is finished--------------- the loss is 0.523465  the accuracy is 0.880000 time:  27.79419994354248 ----------epoch 163 is finished--------------- the loss is 0.568682  the accuracy is 0.880000 time:  27.65660047531128 ----------epoch 164 is finished--------------- the loss is 0.482351  the accuracy is 0.920000 time:  27.829400062561035 ----------epoch 165 is finished--------------- the loss is 0.548296  the accuracy is 0.880000 time:  27.62720012664795 ----------epoch 166 is finished--------------- the loss is 0.456741  the accuracy is 0.880000 time:  27.60700011253357 ----------epoch 167 is finished--------------- the loss is 0.457089  the accuracy is 0.800000 time:  27.6742000579834 ----------epoch 168 is finished--------------- the loss is 0.454088  the accuracy is 0.880000 time:  27.770400285720825 ----------epoch 169 is finished--------------- the loss is 0.636062  the accuracy is 0.880000 time:  28.01960015296936 ----------epoch 170 is finished--------------- the loss is 0.427030  the accuracy is 0.960000 time:  27.59720015525818 ----------epoch 171 is finished--------------- the loss is 0.533497  the accuracy is 0.760000 time:  27.63319993019104 ----------epoch 172 is finished--------------- the loss is 0.683608  the accuracy is 0.840000 time:  27.600000143051147 ----------epoch 173 is finished--------------- the loss is 0.414515  the accuracy is 0.920000 time:  27.706400156021118 ----------epoch 174 is finished--------------- the loss is 0.478144  the accuracy is 0.840000 time:  27.615600109100342 ----------epoch 175 is finished--------------- the loss is 0.460551  the accuracy is 0.960000 time:  27.55440044403076 ----------epoch 176 is finished--------------- the loss is 0.417252  the accuracy is 0.920000 time:  27.61140012741089 ----------epoch 177 is finished--------------- the loss is 0.563965  the accuracy is 0.880000 time:  27.632399797439575 ----------epoch 178 is finished--------------- the loss is 0.364149  the accuracy is 0.960000 time:  27.843200206756592 ----------epoch 179 is finished--------------- the loss is 0.473918  the accuracy is 0.880000 time:  27.80520009994507 ----------epoch 180 is finished--------------- the loss is 0.520914  the accuracy is 0.800000 time:  27.72420024871826 ----------epoch 181 is finished--------------- the loss is 0.408743  the accuracy is 0.800000 time:  27.618600130081177 ----------epoch 182 is finished--------------- the loss is 0.518027  the accuracy is 0.840000 time:  27.498400449752808 ----------epoch 183 is finished--------------- the loss is 0.349911  the accuracy is 0.880000 time:  27.69539976119995 ----------epoch 184 is finished--------------- the loss is 0.481384  the accuracy is 0.840000 time:  27.858200311660767 ----------epoch 185 is finished--------------- the loss is 0.464019  the accuracy is 0.920000 time:  27.4760000705719 ----------epoch 186 is finished--------------- the loss is 0.520933  the accuracy is 0.840000 time:  27.736000299453735 ----------epoch 187 is finished--------------- the loss is 0.458529  the accuracy is 0.840000 time:  27.768400192260742 ----------epoch 188 is finished--------------- the loss is 0.493793  the accuracy is 0.920000 time:  27.51039981842041 ----------epoch 189 is finished--------------- the loss is 0.537836  the accuracy is 0.800000 time:  27.89340043067932 ----------epoch 190 is finished--------------- the loss is 0.416713  the accuracy is 0.840000 time:  27.73680019378662 ----------epoch 191 is finished--------------- the loss is 0.442437  the accuracy is 0.840000 time:  27.68840003013611 ----------epoch 192 is finished--------------- the loss is 0.361428  the accuracy is 0.920000 time:  27.536800146102905 ----------epoch 193 is finished--------------- the loss is 0.466083  the accuracy is 0.880000 time:  27.438400268554688 ----------epoch 194 is finished--------------- the loss is 0.394836  the accuracy is 0.840000 time:  27.5518000125885 ----------epoch 195 is finished--------------- the loss is 0.459927  the accuracy is 0.960000 time:  27.627800226211548 ----------epoch 196 is finished--------------- the loss is 0.481451  the accuracy is 0.920000 time:  27.813600063323975 ----------epoch 197 is finished--------------- the loss is 0.451922  the accuracy is 0.920000 time:  27.564800024032593 ----------epoch 198 is finished--------------- the loss is 0.508634  the accuracy is 0.920000 time:  27.714200258255005 ----------epoch 199 is finished--------------- Optimization Finished!

Process finished with exit code 0