Question

我是ML和图像分类的新手。我正在关注this tutorial

我无法计算混淆矩阵或准确性得分...

有人可以帮我吗？

我已经重塑并尝试将预测放入数组中，但是nothong可以工作...

testLabelsGlobal的形状为(16,)，文件夹Test中有114张图像。

# split the training and testing data
(trainDataGlobal, testDataGlobal, trainLabelsGlobal,testLabelsGlobal) = train_test_split(np.array(global_features),
                                                                        np.array(global_labels),
                                                                        test_size=test_size,
                                                                        random_state=seed)

# 10-fold cross validation
for name, model in models:
kfold = KFold(n_splits=10, random_state=seed)
cv_results = cross_val_score(model, trainDataGlobal, trainLabelsGlobal, cv=kfold, scoring=scoring)
results.append(cv_results)
names.append(name)
msg = "%s: %f (%f)" % (name, cv_results.mean(), cv_results.std())
print(msg)

print('testdataglobal ' , testDataGlobal.shape)
print('testlabelglobal ' , testLabelsGlobal.shape)
#testLabelsGlobal.reshape((16, 1))
#print('testlabelglobal reshape ' , testLabelsGlobal.shape)

clf  = KNeighborsClassifier(n_neighbors=3)
# fit the training data to the model
clf.fit(trainDataGlobal, trainLabelsGlobal)

predictions =[]
# loop through the test images
resultat=[]
for i, file in enumerate(glob.glob(test_path)):
    image = cv2.imread(file)

    fv_hu_moments = fd_hu_moments(image)
    fv_haralick   = fd_haralick(image)
    fv_histogram  = fd_histogram(image)

    global_feature = np.hstack([fv_hu_moments, fv_haralick,fv_histogram])
    print('GLOABAL FEATURE TEST ' , global_feature)

    # scale features in the range (0-1)
    #scaler = MinMaxScaler(feature_range=(-1, 1))
    #rescaled_feature = scaler.fit_transform(global_feature)

    # predict label of test image
    prediction = clf.predict(global_feature.reshape(1,-1))[0]
    print('prediction'  , prediction)
    prediction= np.array(prediction)

    accuracy = accuracy_score(testLabelsGlobal, np.argmax(prediction))

    #print(manually_calculate_tp_tn_fp_fn(testLabelsGlobal, prediction))
    #print(accuracy_score(testDataGlobal, prediction))
    #.append(prediction)
    #resultat.append(metrics.accuracy_score(testLabelsGlobal, prediction))
    #cm = confusion_matrix(testLabelsGlobal, prediction)
    # show predicted label on image

    cv2.putText(image, train_labels[prediction], (20,30), cv2.FONT_HERSHEY_SIMPLEX, 1.0, (0,255,255), 3)

    # display the output image
    #plt.imshow(cv2.cvtColor(image, cv2.COLOR_BGR2RGB))
    #plt.axis('off')
    plt.savefig("resultat"+str(i) + '.jpg')
    #plt.show()

计算每个图像的混淆矩阵

0 个答案: