[英]KNN classifier gives different results when using predict and kneighbors from sklearn.neighbors.KNeighborsClassifier
I want to classify the extracted features from a CNN with k-nearest neighbors classifier from sklearn.neighbors.KNeighborsClassifier.我想使用来自 sklearn.neighbors.KNeighborsClassifier 的 k 最近邻分类器对从 CNN 提取的特征进行分类。 But when I used predict() function on test data it gives a class different than the majority votes that can be found by kneighbors().
但是当我在测试数据上使用 predict() 函数时,它给出的类与 kneighbors() 可以找到的多数票不同。 I am using the following Resnet50 pretrained model to extract the features which is a branch of a siamese network.
我正在使用以下 Resnet50 预训练模型来提取作为连体网络分支的特征。 Details of the siamese network can be found here .
可以在此处找到连体网络的详细信息。
def embedding_model():
baseModel = ResNet50(weights="imagenet", include_top=False,input_tensor=Input(shape=(IMAGE_SIZE, IMAGE_SIZE, 3)))
for layer in baseModel.layers[:165]:
layer.trainable = False
headModel = baseModel.output
headModel = GlobalAveragePooling2D()(headModel)
model = Model(inputs=baseModel.input, outputs=headModel, name = 'embedding_model')
return model
#get embedding model weights from saved weights
embeddings_weights = siamese_test.get_layer('embedding_model').get_weights()
embeddings_branch = siamese_test.get_layer('embedding_model')
input_shape = (224,224,3)
input = Input(shape=input_shape)
x = embeddings_branch(input)
model = Model(input, x)
model.set_weights(embeddings_weights )
out_shape = model.layers[-1].output_shape
Model summary can be found here .模型摘要可以在这里找到。 I used the following function to extract the features using the model.
我使用以下函数来使用模型提取特征。
def create_features(dataset, pre_model,out_shape,batchSize=16):
features = pre_model.predict(dataset, batchSize)
features_flatten = features.reshape((features.shape[0], out_shape[1] ))
return features, features_flatten
train_features, train_features_flatten = create_features(x_train,model,out_shape, batchSize)
test_features, test_features_flatten = create_features(x_test,model,out_shape, batchSize)
Then I used KNN classifier to predict on test features然后我使用 KNN 分类器来预测测试特征
from sklearn.neighbors import KNeighborsClassifier
KNN_classifier = KNeighborsClassifier(n_neighbors=3)
KNN_classifier.fit(train_features_flatten, y_train)
y_pred = KNN_classifier.predict(test_features_flatten)
I used keighbors() function to find the nearest neighbors distance and their corresponding index.我使用 keighbors() 函数来查找最近邻居的距离及其相应的索引。 But it gives me different results than the predicted one.
但它给了我与预期不同的结果。
neighbors_dist, neighbors_index = KNN_classifier.kneighbors(test_features_flatten)
#replace the index with actual class
data2 = np.zeros(neighbors_index.shape, dtype=object)
for i in range(neighbors_index.shape[0]):
for j in range(neighbors_index.shape[1]):
data2[i,j] = str(y_test[neighbors_index[i][j]])
#get the majority class
from collections import Counter
majority_class = np.array([Counter(sorted(row, reverse=True)).most_common(1)[0][0] for row in data2])
As you can see the predicted class is not same as the majority class for first 10 samples如您所见,预测的类与前 10 个样本的多数类不同
for i, pred in enumerate(y_pred):
print(i,pred)
for i, c in enumerate(majority_class):
print(i,c)
Predicted output for first 10 samples: 0 corduroy 1 wool 2 wool 3 brown_bread 4 wood 5 corduroy 6 corduroy 7 corduroy 8 wool 9 wood 10 corduroy前 10 个样品的预测输出: 0 灯芯绒 1 羊毛 2 羊毛 3 brown_bread 4 木材 5 灯芯绒 6 灯芯绒 7 灯芯绒 8 羊毛 9 木材 10 灯芯绒
Majority class for first 10 samples: 0 corduroy 1 cork 2 cork 3 lettuce_leaf 4 linen 5 corduroy 6 wool 7 corduroy 8 brown_bread 9 linen 10 wool前 10 个样品的多数类: 0 灯芯绒 1 软木 2 软木 3 lettuce_leaf 4 亚麻 5 灯芯绒 6 羊毛 7 灯芯绒 8 brown_bread 9 亚麻 10 羊毛
Is there anything I am doing wrong ?有什么我做错了吗? Any help would be appreciated.
任何帮助,将不胜感激。 Thank you.
谢谢你。
This is incorrect:这是不正确的:
data2[i,j] = str(y_test[neighbors_index[i][j]])
The kneighbors
method (and also predict
) finds the nearest training points to the inputs, so you should reference y_train
here. kneighbors
方法(以及predict
)找到最接近输入的训练点,因此您应该在此处引用y_train
。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.