我正在尝试训练CNN(Sklearn神经网络)。我有4张128 x 128像素的图像。形状 - >(4,128,128)我正在阅读像 -
in1 = misc.imread('../data/Train_Data/train-1.jpg', mode='L', flatten=True)/255. in2 = misc.imread('../data/Train_Data/train-2.jpg', mode='L', flatten=True)/255. in3 = misc.imread('../data/Train_Data/train-3.jpg', mode='L', flatten=True)/255. in4 = misc.imread('../data/Train_Data/train-4.jpg', mode='L', flatten=True)/255.
然后像这样创建numpy数组 -
X_train = [in1,in2,in3,in4]
X_train = np.array(X_train)
与标签和测试集相同。
然后我正在训练我的CNN -
nn = Classifier(
layers=[
Convolution('Rectifier', channels=12, kernel_shape=(3, 3), border_mode='full'),
Convolution('Rectifier', channels=8, kernel_shape=(3, 3), border_mode='valid'),
Layer('Rectifier', units=64),
Layer('Softmax')],
learning_rate=0.002,
valid_size=0.2,
n_stable=10,
verbose=True)
nn.fit(X_train, y_train)
它抛出错误 -
回溯(最近一次调用最后一次):文件“/home/zaverichintan/PycharmProjects/WBC_identification/neural/trial.py”,第91行,在nn.fit(X_train,y_train)文件“/ home / zaverichintan / miniconda2 / lib / python2 .7 / site-packages / sknn / mlp.py“,第383行,in ys = [lb.fit_transform(y [:,i])for i,lb in enumerate(self.label_binarizers)] File”/ home / zaverichintan / miniconda2 / lib / python2.7 / site-packages / sklearn / base.py“,第494行,在fit_transform中返回self.fit(X,** fit_params).transform(X)文件”/ home / zaverichintan / miniconda2 /lib/python2.7/site-packages/sklearn/preprocessing/label.py“,第335行,在transform sparse_output = self.sparse_output中)文件”/home/zaverichintan/miniconda2/lib/python2.7/site-packages/ sklearn / preprocessing / label.py“,第497行,在label_binarize y = column_or_1d(y)文件”/home/zaverichintan/miniconda2/lib/python2.7/site-packages/sklearn/utils/validation.py“,第563行,在column_or_1d中引发ValueError(“输入形状错误{0}”。格式(形状))ValueError:输入形状错误(4,128)
你的in1, in2,....inN
是二维数组,是128x128
你必须将它们全部转换为16384
的一维数组。 in1.shape
应该打印(16384,)
和X_train.shape
应该打印(4,16384)
。您可以使用numpy数组并应用[reshape
] [1]函数。 https://docs.scipy.org/doc/numpy/reference/generated/numpy.reshape.html