tensorflow 用自己的資料集訓練CNN模型
阿新 • • 發佈:2019-01-01
最近用tensorflow訓練自己的模型的時候發現,tensorflow官網上所給的例子,都是用處理好資料格式的mnist資料或者其他格式的資料,所以在訓練自己的模型的時候的第一步就卡住了。所以上網搜尋了相關的資料之後便得出了相關的解決方案(有好幾種,這裡只說明一種,另外有TFRecord的格式的網上很多教程,將不在這敘述)….
import os import glob import time import numpy as np import tensorflow as tf from skimage import io, transform # os.environ["TF_CPP_MIN_LOG_LEVEL"] = '1' # 這是預設的顯示等級,顯示所有資訊 # os.environ["TF_CPP_MIN_LOG_LEVEL"] = '2' # 只顯示 warning 和 Error os.environ["TF_CPP_MIN_LOG_LEVEL"] = '3' # 只顯示 Error # 讀取圖片 def read_img(path, w, h): cate = [path + x for x in os.listdir(path) if os.path.isdir(path + x)] # print(cate) imgs = [] labels = [] print('Start read the image ...') for index, folder in enumerate(cate): # print(index, folder) for im in glob.glob(folder + '/*.jpg'): # print('Reading The Image: %s' % im) img = io.imread(im) img = transform.resize(img, (w, h)) imgs.append(img) labels.append(index) print('Finished ...') return np.asarray(imgs, np.float32), np.asarray(labels, np.float32) # 打亂順序 def messUpOrder(data, label): num_example = data.shape[0] arr = np.arange(num_example) np.random.shuffle(arr) data = data[arr] label = label[arr] return data, label # 將所有資料分為訓練集和驗證集 def segmentation(data, label, ratio=0.8): num_example = data.shape[0] s = np.int(num_example * ratio) x_train = data[:s] y_train = label[:s] x_val = data[s:] y_val = label[s:] return x_train, y_train, x_val, y_val # 構建網路 def buildCNN(w, h, c): # 佔位符 x = tf.placeholder(tf.float32, shape=[None, w, h, c], name='x') y_ = tf.placeholder(tf.int32, shape=[None, ], name='y_') # 第一個卷積層 + 池化層(100——>50) conv1 = tf.layers.conv2d( inputs=x, filters=32, kernel_size=[5, 5], padding="same", activation=tf.nn.relu, kernel_initializer=tf.truncated_normal_initializer(stddev=0.01)) pool1 = tf.layers.max_pooling2d(inputs=conv1, pool_size=[2, 2], strides=2) # 第二個卷積層 + 池化層 (50->25) conv2 = tf.layers.conv2d( inputs=pool1, filters=64, kernel_size=[5, 5], padding="same", activation=tf.nn.relu, kernel_initializer=tf.truncated_normal_initializer(stddev=0.01)) pool2 = tf.layers.max_pooling2d(inputs=conv2, pool_size=[2, 2], strides=2) # 第三個卷積層 + 池化層 (25->12) conv3 = tf.layers.conv2d( inputs=pool2, filters=128, kernel_size=[3, 3], padding="same", activation=tf.nn.relu, kernel_initializer=tf.truncated_normal_initializer(stddev=0.01)) pool3 = tf.layers.max_pooling2d(inputs=conv3, pool_size=[2, 2], strides=2) # 第四個卷積層 + 池化層 (12->6) conv4 = tf.layers.conv2d( inputs=pool3, filters=128, kernel_size=[3, 3], padding="same", activation=tf.nn.relu, kernel_initializer=tf.truncated_normal_initializer(stddev=0.01)) pool4 = tf.layers.max_pooling2d(inputs=conv4, pool_size=[2, 2], strides=2) re1 = tf.reshape(pool4, [-1, 6 * 6 * 128]) # 全連線層 dense1 = tf.layers.dense(inputs=re1, units=1024, activation=tf.nn.relu, kernel_initializer=tf.truncated_normal_initializer(stddev=0.01), kernel_regularizer=tf.contrib.layers.l2_regularizer(0.003)) dense2 = tf.layers.dense(inputs=dense1, units=512, activation=tf.nn.relu, kernel_initializer=tf.truncated_normal_initializer(stddev=0.01), kernel_regularizer=tf.contrib.layers.l2_regularizer(0.003)) logits = tf.layers.dense(inputs=dense2, units=20, activation=None, kernel_initializer=tf.truncated_normal_initializer(stddev=0.01), kernel_regularizer=tf.contrib.layers.l2_regularizer(0.003)) return logits, x, y_ # 返回損失函式的值,準確值等引數 def accCNN(logits, y_): loss = tf.losses.sparse_softmax_cross_entropy(labels=y_, logits=logits) train_op = tf.train.AdamOptimizer(learning_rate=0.001).minimize(loss) correct_prediction = tf.equal(tf.cast(tf.argmax(logits, 1), tf.int32), y_) acc = tf.reduce_mean(tf.cast(correct_prediction, tf.float32)) return loss, train_op, correct_prediction, acc # 定義一個函式,按批次取資料 def minibatches(inputs=None, targets=None, batch_size=None, shuffle=False): assert len(inputs) == len(targets) if shuffle: indices = np.arange(len(inputs)) np.random.shuffle(indices) for start_idx in range(0, len(inputs) - batch_size + 1, batch_size): if shuffle: excerpt = indices[start_idx:start_idx + batch_size] else: excerpt = slice(start_idx, start_idx + batch_size) yield inputs[excerpt], targets[excerpt] def runable(x_train, y_train, train_op, loss, acc, x, y_, x_val, y_val): # 訓練和測試資料,可將n_epoch設定更大一些 n_epoch = 50 batch_size = 64 sess = tf.InteractiveSession() sess.run(tf.global_variables_initializer()) for epoch in range(n_epoch): # training train_loss, train_acc, n_batch = 0, 0, 0 for x_train_a, y_train_a in minibatches(x_train, y_train, batch_size, shuffle=True): _, err, ac = sess.run([train_op, loss, acc], feed_dict={x: x_train_a, y_: y_train_a}) train_loss += err train_acc += ac n_batch += 1 print("train loss: %f" % (train_loss / n_batch)) print("train acc: %f" % (train_acc / n_batch)) # validation val_loss, val_acc, n_batch = 0, 0, 0 for x_val_a, y_val_a in minibatches(x_val, y_val, batch_size, shuffle=False): err, ac = sess.run([loss, acc], feed_dict={x: x_val_a, y_: y_val_a}) val_loss += err val_acc += ac n_batch += 1 print("validation loss: %f" % (val_loss / n_batch)) print("validation acc: %f" % (val_acc / n_batch)) print('*' * 50) sess.close() if __name__ == '__main__': imgpath = '../dataset/classify/' w = 100 h = 100 c = 3 ratio = 0.8 # 選取訓練集的比例 data, label = read_img(path=imgpath, w=w, h=h) data, label = messUpOrder(data=data, label=label) x_train, y_train, x_val, y_val = segmentation(data=data, label=label, ratio=ratio) logits, x, y_ = buildCNN(w=w, h=h, c=c) loss, train_op, correct_prediction, acc = accCNN(logits=logits, y_=y_) runable(x_train=x_train, y_train=y_train, train_op=train_op, loss=loss, acc=acc, x=x, y_=y_, x_val=x_val, y_val=y_val)