Python sklearn.metrics.jaccard_similarity_score() Examples

def multi_class_classification(data_X,data_Y):
    '''
    calculate multi-class classification and return related evaluation metrics
    '''

    svc = svm.SVC(C=1, kernel='linear')
    # X_train, X_test, y_train, y_test = train_test_split( data_X, data_Y, test_size=0.4, random_state=0) 
    clf = svc.fit(data_X, data_Y) #svm
    # array = svc.coef_
    # print array
    predicted = cross_val_predict(clf, data_X, data_Y, cv=2)
    print "accuracy",metrics.accuracy_score(data_Y, predicted)
    print "f1 score macro",metrics.f1_score(data_Y, predicted, average='macro') 
    print "f1 score micro",metrics.f1_score(data_Y, predicted, average='micro') 
    print "precision score",metrics.precision_score(data_Y, predicted, average='macro') 
    print "recall score",metrics.recall_score(data_Y, predicted, average='macro') 
    print "hamming_loss",metrics.hamming_loss(data_Y, predicted)
    print "classification_report", metrics.classification_report(data_Y, predicted)
    print "jaccard_similarity_score", metrics.jaccard_similarity_score(data_Y, predicted)
    # print "log_loss", metrics.log_loss(data_Y, predicted)
    print "zero_one_loss", metrics.zero_one_loss(data_Y, predicted)
    # print "AUC&ROC",metrics.roc_auc_score(data_Y, predicted)
    # print "matthews_corrcoef", metrics.matthews_corrcoef(data_Y, predicted) 

def evaluation_analysis(true_label,predicted): 
    '''
    return all metrics results
    '''
    print "accuracy",metrics.accuracy_score(true_label, predicted)
    print "f1 score macro",metrics.f1_score(true_label, predicted, average='macro')     
    print "f1 score micro",metrics.f1_score(true_label, predicted, average='micro') 
    print "precision score",metrics.precision_score(true_label, predicted, average='macro') 
    print "recall score",metrics.recall_score(true_label, predicted, average='macro') 
    print "hamming_loss",metrics.hamming_loss(true_label, predicted)
    print "classification_report", metrics.classification_report(true_label, predicted)
    print "jaccard_similarity_score", metrics.jaccard_similarity_score(true_label, predicted)
    print "log_loss", metrics.log_loss(true_label, predicted)
    print "zero_one_loss", metrics.zero_one_loss(true_label, predicted)
    print "AUC&ROC",metrics.roc_auc_score(true_label, predicted)
    print "matthews_corrcoef", metrics.matthews_corrcoef(true_label, predicted) 

def metric_report(y_pred, y_true, therhold=0.5):
    y_prob = y_pred.copy()
    y_pred[y_pred > therhold] = 1
    y_pred[y_pred <= therhold] = 0

    acc_container = {}
    ja, prauc, avg_p, avg_r, avg_f1 = multi_label_metric(
        y_true, y_pred, y_prob)
    acc_container['jaccard'] = ja
    acc_container['f1'] = avg_f1
    acc_container['prauc'] = prauc

    # acc_container['jaccard'] = jaccard_similarity_score(y_true, y_pred)
    # acc_container['f1'] = f1(y_true, y_pred)
    # acc_container['auc'] = roc_auc(y_true, y_prob)
    # acc_container['prauc'] = precision_auc(y_true, y_prob)

    for k, v in acc_container.items():
        logger.info('%-10s : %-10.4f' % (k, v))

    return acc_container 

def test_multilabel_jaccard_similarity_score_deprecation():
    # Dense label indicator matrix format
    y1 = np.array([[0, 1, 1], [1, 0, 1]])
    y2 = np.array([[0, 0, 1], [1, 0, 1]])

    # size(y1 \inter y2) = [1, 2]
    # size(y1 \union y2) = [2, 2]

    jss = partial(assert_warns, DeprecationWarning, jaccard_similarity_score)
    assert_equal(jss(y1, y2), 0.75)
    assert_equal(jss(y1, y1), 1)
    assert_equal(jss(y2, y2), 1)
    assert_equal(jss(y2, np.logical_not(y2)), 0)
    assert_equal(jss(y1, np.logical_not(y1)), 0)
    assert_equal(jss(y1, np.zeros(y1.shape)), 0)
    assert_equal(jss(y2, np.zeros(y1.shape)), 0) 

def _get_similarity_values(self, q1_csc, q2_csc):
        cosine_sim = []
        manhattan_dis = []
        eucledian_dis = []
        jaccard_dis = []
        minkowsk_dis = []
        
        for i,j in zip(q1_csc, q2_csc):
            sim = cs(i, j)
            cosine_sim.append(sim[0][0])
            sim = md(i, j)
            manhattan_dis.append(sim[0][0])
            sim = ed(i, j)
            eucledian_dis.append(sim[0][0])
            i_ = i.toarray()
            j_ = j.toarray()
            try:
                sim = jsc(i_, j_)
                jaccard_dis.append(sim)
            except:
                jaccard_dis.append(0)
                
            sim = minkowski_dis.pairwise(i_, j_)
            minkowsk_dis.append(sim[0][0])
        return cosine_sim, manhattan_dis, eucledian_dis, jaccard_dis, minkowsk_dis 

def test_classifier_chain_crossval_fit_and_predict():
    # Fit classifier chain with cross_val_predict and verify predict
    # performance
    X, Y = generate_multilabel_dataset_with_correlations()
    classifier_chain_cv = ClassifierChain(LogisticRegression(), cv=3)
    classifier_chain_cv.fit(X, Y)

    classifier_chain = ClassifierChain(LogisticRegression())
    classifier_chain.fit(X, Y)

    Y_pred_cv = classifier_chain_cv.predict(X)
    Y_pred = classifier_chain.predict(X)

    assert_equal(Y_pred_cv.shape, Y.shape)
    assert_greater(jaccard_similarity_score(Y, Y_pred_cv), 0.4)

    assert_not_equal(jaccard_similarity_score(Y, Y_pred_cv),
                     jaccard_similarity_score(Y, Y_pred)) 

def test_classifier_chain_vs_independent_models():
    # Verify that an ensemble of classifier chains (each of length
    # N) can achieve a higher Jaccard similarity score than N independent
    # models
    X, Y = generate_multilabel_dataset_with_correlations()
    X_train = X[:600, :]
    X_test = X[600:, :]
    Y_train = Y[:600, :]
    Y_test = Y[600:, :]

    ovr = OneVsRestClassifier(LogisticRegression())
    ovr.fit(X_train, Y_train)
    Y_pred_ovr = ovr.predict(X_test)

    chain = ClassifierChain(LogisticRegression())
    chain.fit(X_train, Y_train)
    Y_pred_chain = chain.predict(X_test)

    assert_greater(jaccard_similarity_score(Y_test, Y_pred_chain),
                   jaccard_similarity_score(Y_test, Y_pred_ovr)) 

def multi_iou(a,b):
    jk = 0.0
    vals = np.unique(a)[1:]
    for v in vals:
        ac = a.copy()
        ac[ac != v] = 0
        bc = b.copy()
        bc[bc != v] = 0
        jk += iou(ac,bc)
    return jk/len(vals) 

def calc_jacc_img_msk(model, img, msk, batch_size, n_classes):
    prd = model.predict(img, batch_size= batch_size)
    #print("prd.shape {0}, msk.shape {1}". format(prd.shape, msk.shape))
     #prd.shape, msk.shape (16, 2, 256, 256) (16, 2, 256, 256)
    avg, trs = [], []

    for i in range(n_classes):
        t_msk = msk[:, i, :, :] # t_mask shape is (Npredictions, H, W)
        t_prd = prd[:, i, :, :]
        t_msk = t_msk.reshape(msk.shape[0] * msk.shape[2], msk.shape[3]) # shape is Npredictions*W, H
        t_prd = t_prd.reshape(msk.shape[0] * msk.shape[2], msk.shape[3])

        m, b_tr = 0, 0
        for j in range(10):
            tr = j / 10
            pred_binary_mask = t_prd > tr

            jk = jaccard_similarity_score(t_msk, pred_binary_mask)
            if jk > m:
                m = jk
                b_tr = tr

        print("i, m, b_tr", i, m, b_tr)
        avg.append(m)
        trs.append(b_tr)

    score = sum(avg) / n_classes
    return score, trs 

def display_pred(pred_res, true_masks, config, modelName, amt_pred, trs, min_pred_sum, output_folder):
    os.makedirs(output_folder, exist_ok=True)
    trs_str = "trs_none"
    if trs != None:
        trs_str = '_'.join([str(x) for x in trs])
        trs_str = "trs_" + trs_str

    #print("Saving predictions when np.sum(pred) >", min_pred_sum)
    nothing_saved = True
    for p in range(amt_pred):
        for i in range(config.NUM_CLASSES):
            pred = pred_res[p,:, :, i]
            true_mask = true_masks[p, :, :, i]

            sum_pred = np.sum(pred)
            sum_mask = np.sum(true_mask)
            if sum_pred > min_pred_sum :
                    #and sum_mask > min_pred_sum:

                jk = jaccard_similarity_score(true_mask, pred)
                #print("Calc jaccard", jk)
                fn = os.path.join(output_folder,"{4}{0}_p{1}_cl{2}_{3}.png".format(modelName, p, i, trs_str, jk))
                #print("Saving  predictions with np.sum {0} to  {1}".format(sum, fn))
                plt.imsave(fn, pred, cmap='hot')

                fn_tr= os.path.join(output_folder,"{4}{0}_p{1}_TRUE_cl{2}_{3}.png".format(modelName, p, i, trs_str, jk))
                plt.imsave(fn_tr, true_mask, cmap='hot')

                nothing_saved = False

    if (nothing_saved):
        print("All predictions did not satisfy: sum_pred > min_pred_sum, nothing saved. Min_pred_sum:", min_pred_sum) 

def similarity(self, query, type):
        assert self.corpus != None, "self.corpus can't be None"
        ret = []
        if type == 'cosine':
            query = self.get_vector(query)
            for item in self.corpus_vec:
                sim = cosine_similarity(item, query)
                ret.append(sim[0][0])
        elif type == 'manhattan':
            query = self.get_vector(query)
            for item in self.corpus_vec:
                sim = manhattan_distances(item, query)
                ret.append(sim[0][0])
        elif type == 'euclidean':
            query = self.get_vector(query)
            for item in self.corpus_vec:
                sim = euclidean_distances (item, query)
                ret.append(sim[0][0])
        #elif type == 'jaccard':
        #    #query = query.split()
        #    query = self.get_vector(query)
        #    for item in self.corpus_vec:
        #        pdb.set_trace()
        #        sim = jaccard_similarity_score(item, query)
        #        ret.append(sim)
        elif type == 'bm25':
            query = query.split()
            ret = self.bm25_model.get_scores(query)
        else:
            raise ValueError('similarity type error:%s'%type)
        return ret 

def jaccard_coefficient(left, right):
    jaccard_scores = jaccard_similarity_score(left,right)
    return jaccard_scores 

def ComputeMetrics(prob, batch_labels, p1, p2, rgb=None, save_path=None, ind=0):
    """
    Computes all metrics between probability map and corresponding label.
    If you give also an rgb image it will save many extra meta data image.
    """
    GT = label(batch_labels.copy())
    PRED = PostProcess(prob, p1, p2)
    # PRED = label((prob > 0.5).astype('uint8'))
    lbl = GT.copy()
    pred = PRED.copy()
    aji = AJI_fast(lbl, pred)
    lbl[lbl > 0] = 1
    pred[pred > 0] = 1 
    l, p = lbl.flatten(), pred.flatten()
    acc = accuracy_score(l, p)
    roc = roc_auc_score(l, p)
    jac = jaccard_similarity_score(l, p)
    f1 = f1_score(l, p)
    recall = recall_score(l, p)
    precision = precision_score(l, p)
    if rgb is not None:
        xval_n = join(save_path, "xval_{}.png").format(ind)
        yval_n = join(save_path, "yval_{}.png").format(ind)
        prob_n = join(save_path, "prob_{}.png").format(ind)
        pred_n = join(save_path, "pred_{}.png").format(ind)
        c_gt_n = join(save_path, "C_gt_{}.png").format(ind)
        c_pr_n = join(save_path, "C_pr_{}.png").format(ind)

        imsave(xval_n, rgb)
        imsave(yval_n, color_bin(GT))
        imsave(prob_n, prob)
        imsave(pred_n, color_bin(PRED))
        imsave(c_gt_n, add_contours(rgb, GT))
        imsave(c_pr_n, add_contours(rgb, PRED))

    return acc, roc, jac, recall, precision, f1, aji 

def test_jaccard_similarity_score(self):
        result = self.df.metrics.jaccard_similarity_score()
        expected = metrics.jaccard_similarity_score(self.target, self.pred)
        self.assertEqual(result, expected)

        result = self.df.metrics.jaccard_similarity_score(normalize=False)
        expected = metrics.jaccard_similarity_score(self.target, self.pred, normalize=False)
        self.assertEqual(result, expected) 

def test_multilabel_jaccard_similarity_score():
    # Dense label indicator matrix format
    y1 = np.array([[0, 1, 1], [1, 0, 1]])
    y2 = np.array([[0, 0, 1], [1, 0, 1]])

    # size(y1 \inter y2) = [1, 2]
    # size(y1 \union y2) = [2, 2]

    assert_equal(jaccard_similarity_score(y1, y2), 0.75)
    assert_equal(jaccard_similarity_score(y1, y1), 1)
    assert_equal(jaccard_similarity_score(y2, y2), 1)
    assert_equal(jaccard_similarity_score(y2, np.logical_not(y2)), 0)
    assert_equal(jaccard_similarity_score(y1, np.logical_not(y1)), 0)
    assert_equal(jaccard_similarity_score(y1, np.zeros(y1.shape)), 0)
    assert_equal(jaccard_similarity_score(y2, np.zeros(y1.shape)), 0)