KaneAI - World’s First E2E Software Testing Agent.Book a Demo Join TestMu Conference

How to use t_full method in fMBT

Best Python code snippet using fMBT_python

waveform.py

Source:waveform.py

1import json2import numpy as np3import pylab as pl4def waveform(f, A, b, t0, tend, d_end_t=None, gamma=0.0, phi0=0.0, 5             N=1000, verbose=False, seed_number=None, project_name=None):6    """7    METHOD8    ======9    Takes input parameters of a wave and the strength and duration of10    noise, and returns the data.11    PARAMETERS12    ==========13    f : (Float) Frequency of the signal14    A : (Float) Amplitude of the signal15    b : (Float) Amplitude of the noise16    t0 : (Float) Timestamp of the beginning of the signal17    tend : (Float) Time stamp of the end of the signal18    d_end_t : (Float) Time stamp of the end time of the data. Default = None19    gamma : (Float) Attenuation factor of the signal. Default = 0.020    phi0 : (Float) Initial phase of the signal. Default = 0.021    N : (Int) Total number of time stamps. Default = 100022    verbose: (Bool) Set True to get diagnostic stdout. Default = False23    seed_number: (Int) Number set to seed noise. Default = None24    project_name: (String) Name given to png and json file created. Default = None25    OUTPUT26    ======27    A tuple of a float and two numpy arrays (dt, T_full, d), where dt is 28    the resolution of the time series. T_full is the full list of time stamps29    of the data starting at 0 and ending and d_end_t, and d is the 30    corresponding displacement values in the data.31    """32    33    # Conditional for noise duration34    # If the data-end time is supplied to be too small:35    if verbose:36        print("Making sure that the stretch of data is longer than signal")37    assert t0 > 0, "Signal should start later than t=0"38    if (d_end_t is None) or (tend > d_end_t - 10):39        d_end_t = tend + 1040        if verbose:41            print("data end time is set at {}".format(d_end_t))42    43    T = np.linspace(t0, tend, N) # Time stamps of signal44    dt = np.mean(np.diff(T)) # figuring out the resolution of the series45    if verbose:46        print("Mean value of timing resolution = {}".format(dt))47    48    t = t0 # Initializing the time series at the start time49    t_minus = [] # To populate time stamps prior to the signal start50    while t >= 0: # Making sure that we reach all the way back to zero.51        t = t - dt52        t_minus.append(t)  # Create time spamps from (t0-dt) to 053    t_minus = np.array(t_minus)[::-1]  # Reverse to be from 0 to t054    t_minus = t_minus[t_minus >= 0]  # Eliminate numbers less than 055    56    t_plus = np.arange(tend+dt, d_end_t, dt)  # Time stamps from (tend+dt) to d_end_t, in dt's57    58    T_full = np.hstack((t_minus, T, t_plus))  # Connect time stamps59    60    dev = np.std(np.diff(T_full))  # Standard deviation in dt's of T_full61    if verbose:62        print("Standard deviation of the resolution of time = {}".format(dev))63    if verbose:64        print("Creating time series of the signal...")65    w = 2 * np.pi * f  66    y = A*np.sin(w*T + phi0)*np.exp(-gamma*(T-t0))67    68    # Padding of signal data69    if verbose:70        print("Creating the zero-padded signal...")71    y_minus = np.zeros_like(t_minus)72    y_plus = np.zeros_like(t_plus)73    y_full = np.hstack((y_minus, y, y_plus))74    75    if verbose:76        print("Creating random noise...")77    if seed_number is None:78        seed_number = 179    np.random.seed(seed = seed_number)80    noise = -b+2*b*np.random.random(len(T_full))  # Noise!81    82    if verbose:83        print("Creating final data")84    d = noise + y_full  # Complete Data!85    86    # Graphing   87    pl.rcParams.update({'font.size': 18})88    pl.figure(figsize=(20,15))89    pl.plot(T_full, noise, color = 'green', linewidth=2)  # Noise90    pl.plot(T_full, d, color = 'black', linewidth=2)  # Combined91    pl.plot(T, y, color = 'orange', linewidth=2)  # Signal92    pl.xlabel("Time")93    pl.ylabel("displacement")94    text = "f={}; A={}; b={}; t0={}; tend={}; gamma={}; N={}"95    pl.title(text.format(f, A, b, t0, tend, gamma, N))96    #if project_name is None:97    #    project_name = "test"98    #pl.savefig("figures/{}-waveform_plot-f_{}-A_{}-b_{}-t0_{}-tend_{}-gamma_{}-seed_{}.png".format(project_name, f, A, b, t0, tend, gamma, seed_number))99    100    T_full = T_full101    d = d102    #data = {"dt" : dt, "t_full" : T_full, "d" : d}103    #outputfile = "data/{}-waveform_data-f_{}-A_{}-b_{}-t0_{}-tend_{}-gamma_{}-seed_{}.json".format(project_name, f, A, b, t0, tend, gamma, seed_number)104    #with open(outputfile, "w") as f:105    #    json.dump(data, f, indent=2, sort_keys=True)...

calcSimilarity.py

Source:calcSimilarity.py

1'''2File name: calcSimilarity.py3Author: Ningshan Zhang, Zheyuan Xie4Date created: 2018-12-195'''6import cv27import numpy as np8def calcSimilarity(landmarks1, landmarks2):9    T = cv2.estimateRigidTransform(landmarks1, landmarks2, False)10    if T is None:11        return np.inf12    T_full = np.vstack((T,np.array([0,0,1])))13    landmarks1_full = np.vstack((landmarks1.T,np.ones((1,landmarks1.shape[0]))))14    landmarks1_trans = np.dot(T_full,landmarks1_full)15    landmarks1_trans = landmarks1_trans[0:2,:].T16    dist = np.sum(np.sum((landmarks1_trans-landmarks2)**2,axis=1))17    return dist18# match a single face for all target frames19def findMinDistFace_static(landmarks1, landmarks2):20    faceind = 021    mindist = np.inf22    for i in range(len(landmarks1)):23        if landmarks1[i] is None:24            continue25        dist = 026        for j in range(len(landmarks2)):27            if landmarks2[j] is None:28                continue29            T = cv2.estimateRigidTransform(landmarks1[i], landmarks2[j], False)30            if T is None:31                continue32            T_full = np.vstack((T,np.array([0,0,1])))33            landmarks1_full = np.vstack((landmarks1[i].T,np.ones((1,landmarks1[i].shape[0]))))34            landmarks1_trans = np.dot(T_full,landmarks1_full)35            landmarks1_trans = landmarks1_trans[0:2,:].T36            dist = dist + calcSimilarity(landmarks1_trans,landmarks2[j])37        if dist < mindist:38            faceind = i39            mindist = dist40    return (faceind * np.ones((len(landmarks2),))).astype(int)41# match a source face for each target frames42def findMinDistFace(landmarks1, landmarks2):43    faceind = np.zeros((len(landmarks2),))44    for i in range(len(landmarks2)):45        if landmarks2[i] is None:46            continue47        mindist = np.inf48        for j in range(len(landmarks1)):49            if landmarks1[j] is None:50                continue51            T = cv2.estimateRigidTransform(landmarks1[j], landmarks2[i], False)52            if T is None:53                continue54            T_full = np.vstack((T,np.array([0,0,1])))55            landmarks1_full = np.vstack((landmarks1[j].T,np.ones((1,landmarks1[j].shape[0]))))56            landmarks1_trans = np.dot(T_full,landmarks1_full)57            landmarks1_trans = landmarks1_trans[0:2,:].T58            dist = calcSimilarity(landmarks1_trans,landmarks2[i])59            if dist < mindist:60                faceind[i] = j61                mindist = dist62    return faceind.astype(int)63if __name__ == "__main__":64    from loader import loadlandmarks_facepp, loadvideo65    import time66    easy1 = 'Datasets/Easy/FrankUnderwood.mp4'67    easy2 = 'Datasets/Easy/MrRobot.mp4'68    lm1 = loadlandmarks_facepp(easy2)69    lm2 = loadlandmarks_facepp(easy1)70    video2 = loadvideo(easy1)71    print(len(lm1))72    t0 = time.time()73    ind = findMinDistFace(lm1, lm2)74    ind_s = findMinDistFace_static(lm1,lm2)75    t1 = time.time()76    print(t1-t0)77    T = cv2.estimateRigidTransform(lm1[ind[0]], lm2[0], False)78    T_full = np.vstack((T,np.array([0,0,1])))79    landmarks1_full = np.vstack((lm1[ind[0]].T,np.ones((1,lm1[ind[0]].shape[0]))))80    landmarks1_trans = np.dot(T_full,landmarks1_full)81    landmarks1_trans = landmarks1_trans[0:2,:].T82    for groups in landmarks1_trans.astype(int):83        cv2.circle(video2[0], (groups[0],groups[1]), 1, (0, 255, 255), 2)84    for groups in lm2[0].astype(int):85        cv2.circle(video2[0], (groups[0],groups[1]), 1, (0, 0, 255), 2)86    for i in range(83):87        cv2.line(video2[0],88            (lm2[0].astype(int)[i,0],lm2[0].astype(int)[i,1]),89            (landmarks1_trans.astype(int)[i,0],landmarks1_trans.astype(int)[i,1]),(0,255,255),2)90    cv2.imshow('frame',video2[0])...

7_4_hints.py

Source:7_4_hints.py

1# 7-4 prep2dummies = pd.get_dummies(t[['sex', 'pclass']])3dummies.head()4t_full = pd.concat([t, dummies], axis=1)5t_full.head()6t_full = t_full.dropna(subset=['age', 'sex_male', 'fare', 'pclass_2nd', 'pclass_3rd', 'survived'])7t_full.shape8X = t_full[['age', 'sex_male', 'fare', 'pclass_2nd', 'pclass_3rd']]9y = t_full['survived']10X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=777)11sk_logit = LogisticRegression(penalty='none').fit(X_train, y_train)12sk_logit.coef_13prob_pred = sk_logit.predict_proba(X_test)[:, 0]14confusion_matrix(y_test, prob_pred > 0.45)15sns.heatmap(confusion_matrix(y_test, prob_pred > 0.4), 16annot=True, fmt='d', cmap='Blues');17plot_roc_curve(sk_logit, X_test, y_test);18FPR = 1 -  specificity = FP / (FP + TN) = 19= FP / cond Negative20TPR = Sencitivity = TP / (TP + FN) = 21= TP / cond Positive22plot_precision_recall_curve(sk_logit, X_test, y_test);23recall = TPR = Sencitivity = TP / (TP + FN) = 24= TP / cond Positive25precision = TP / (TP + FP) =...

Automation Testing Tutorials

Learn to execute automation testing from scratch with LambdaTest Learning Hub. Right from setting up the prerequisites to run your first automation test, to following best practices and diving deeper into advanced test scenarios. LambdaTest Learning Hubs compile a list of step-by-step guides to help you be proficient with different test automation frameworks i.e. Selenium, Cypress, TestNG etc.