深層学習初心者の学生です
上記のコードを元にして、tflearnを用いたカラー画像のnetwork in networkによるクラス21個の画像分類を行おうとしたのですが途中でエラーが発生してしまい原因がいまいちつかめていません
アドバイスをいただけますと幸いです
画像枚数は各クラス100枚で合計2100枚のものをそれぞれテスト用30枚、トレイン用70枚に分けています
Ubuntu16.04,python3.6,jupyternotebookで実行しています
以下コードです
Python
1from __future__ import division, print_function, absolute_import 2import tflearn 3from tflearn.data_utils import shuffle, to_categorical 4from tflearn.layers.core import input_data, dropout, flatten 5from tflearn.layers.conv import conv_2d, max_pool_2d, avg_pool_2d 6from tflearn.layers.estimator import regression 7import numpy as np 8 9num_classes = 21 10#arrayの読み込み 11X_train, X_test, y_train, y_test = np.load("tf_correct.npy")#X_train.shape=(1470,32,32,3),X_test.shape=(630,32,32,3),y_train.shape=(1470,21),y_test.shape=(630,21) 12X_train = X_train.astype("float") / 256 13X_test = X_test.astype("float") / 256 14y_train = to_categorical(y_train, num_classes) 15y_test = to_categorical(y_test, num_classes) 16 17network = input_data(shape=[None, 32, 32, 3]) 18network = conv_2d(network, 192, 5, activation='relu') 19network = conv_2d(network, 160, 1, activation='relu') 20network = conv_2d(network, 96, 1, activation='relu') 21network = max_pool_2d(network, 3, strides=2) 22network = dropout(network, 0.5) 23network = conv_2d(network, 192, 5, activation='relu') 24network = conv_2d(network, 192, 1, activation='relu') 25network = conv_2d(network, 192, 1, activation='relu') 26network = avg_pool_2d(network, 3, strides=2) 27network = dropout(network, 0.5) 28network = conv_2d(network, 192, 3, activation='relu') 29network = conv_2d(network, 192, 1, activation='relu') 30network = conv_2d(network, 21, 1, activation='relu') 31network = avg_pool_2d(network, 8) 32network = flatten(network) 33network = regression(network, optimizer='adam', 34 loss='softmax_categorical_crossentropy', 35 learning_rate=0.001) 36 37model = tflearn.DNN(network) 38model.fit(X_train, y_train, n_epoch=50, shuffle=True, validation_set=(X_test, y_test),batch_size=32) 39
これを実行すると次のようなエラーが発生します
Python
1IndexError Traceback (most recent call last) 2<ipython-input-12-fd9b52414d33> in <module>() 3 45 # Training 4 46 model = tflearn.DNN(network) 5---> 47 model.fit(X_train, y_train, n_epoch=50, shuffle=True, validation_set=(X_test, y_test),batch_size=32) 6 7~/anaconda3/lib/python3.6/site-packages/tflearn/models/dnn.py in fit(self, X_inputs, Y_targets, n_epoch, validation_set, show_metric, batch_size, shuffle, snapshot_epoch, snapshot_step, excl_trainops, validation_batch_size, run_id, callbacks) 8 182 # TODO: check memory impact for large data and multiple optimizers 9 183 feed_dict = feed_dict_builder(X_inputs, Y_targets, self.inputs, 10--> 184 self.targets) 11 185 feed_dicts = [feed_dict for i in self.train_ops] 12 186 val_feed_dicts = None 13 14~/anaconda3/lib/python3.6/site-packages/tflearn/utils.py in feed_dict_builder(X, Y, net_inputs, net_targets) 15 281 X = [X] 16 282 for i, x in enumerate(X): 17--> 283 feed_dict[net_inputs[i]] = x 18 284 else: 19 285 # If a dict is provided 20 21IndexError: list index out of range
ここでのlist index out of rangeはどこのことを指しているのでしょうか?
よろしくお願いします
回答1件
あなたの回答
tips
プレビュー
バッドをするには、ログインかつ
こちらの条件を満たす必要があります。
2018/06/09 12:32