How to use wfdb - 10 common examples

if file.endswith(".dat"):
                dat_files.append(file)
        
        mit_records = [w.replace(".dat", "") for w in dat_files]
        
        results = np.zeros((len(mit_records), 5), dtype=int)

        i = 0
        for record in mit_records:
            progress = int(i/float(len(mit_records))*100.0)
            print("MITDB progress: %i%%" % progress)

            sig, fields = wfdb.rdsamp(self.mitdb_dir+'/'+record)
            unfiltered_ecg = sig[:, 0]  

            ann = wfdb.rdann(str(self.mitdb_dir+'/'+record), 'atr')    
            anno = _tester_utils.sort_MIT_annotations(ann)    

            r_peaks = detector(unfiltered_ecg)

            delay = _tester_utils.calcMedianDelay(r_peaks, unfiltered_ecg, max_delay_in_samples)

            if delay > 1:

                TP, FP, FN = _tester_utils.evaluate_detector(r_peaks, anno, delay, tol=tolerance)
                TN = len(unfiltered_ecg)-(TP+FP+FN)
                
                results[i, 0] = int(record)    
                results[i, 1] = TP
                results[i, 2] = FP
                results[i, 3] = FN
                results[i, 4] = TN

max_delay_in_samples = 350 / 5
        dat_files = []
        for file in os.listdir(self.mitdb_dir):
            if file.endswith(".dat"):
                dat_files.append(file)
        
        mit_records = [w.replace(".dat", "") for w in dat_files]
        
        results = np.zeros((len(mit_records), 5), dtype=int)

        i = 0
        for record in mit_records:
            progress = int(i/float(len(mit_records))*100.0)
            print("MITDB progress: %i%%" % progress)

            sig, fields = wfdb.rdsamp(self.mitdb_dir+'/'+record)
            unfiltered_ecg = sig[:, 0]  

            ann = wfdb.rdann(str(self.mitdb_dir+'/'+record), 'atr')    
            anno = _tester_utils.sort_MIT_annotations(ann)    

            r_peaks = detector(unfiltered_ecg)

            delay = _tester_utils.calcMedianDelay(r_peaks, unfiltered_ecg, max_delay_in_samples)

            if delay > 1:

                TP, FP, FN = _tester_utils.evaluate_detector(r_peaks, anno, delay, tol=tolerance)
                TN = len(unfiltered_ecg)-(TP+FP+FN)
                
                results[i, 0] = int(record)    
                results[i, 1] = TP

- patient001/
    - patient002/
    - ...
"""


#==============================================================================
# Extracting
#==============================================================================
data={"Control": {}, "Patient": {}}
participants = [x for x in os.listdir("./data/") if 'patient' in x]
for participant in participants:
    files = os.listdir("./data/" + participant)
    if len([x for x in files if '.dat' in x]) > 0:
        file = [x for x in files if '.dat' in x][0].split(".")[0]
        signals, info = wfdb.rdsamp("data/" + participant + "/" + file)

        signals = pd.DataFrame(signals)
        signals.columns = info["sig_name"]

        data_participant = {}
        data_participant["Signals"] = signals
        data_participant["sampling_rate"] = info["fs"]


        for key in info["comments"]:
            try:
                data_participant[key.split(": ")[0]] = key.split(": ")[1]
            except IndexError:
                data_participant[key.split(":")[0]] = np.nan

        if data_participant["Reason for admission"] in ["n/a", "Healthy control"]:

def generate_raw_db(self):
        """Generate the raw version of the MIT-BIH Atrial Fibrillation database in the 'raw' folder."""
        print('Generating Raw MIT-BIH Atrial Fibrillation Database ...')
        # Download database
        wfdb.dl_database(self.db_name, self.raw_path)

        # Get list of recordings
        self.record_ids = [file.split('.')[0] for file in os.listdir(self.raw_path) if '.dat' in file]
        print('Complete!\n')

def generate_raw_db(self):
        """Generate the raw version of the MIT-BIH Normal Sinus Rhythm database in the 'raw' folder."""
        print('Generating Raw MIT-BIH Normal Sinus Rhythm Database ...')
        # Download database
        wfdb.dl_database(self.db_name, self.raw_path)

        # Get list of recordings
        self.record_ids = [file.split('.')[0] for file in os.listdir(self.raw_path) if '.dat' in file]
        print('Complete!\n')

sampto = None#int(60. * 1 * 360)
sampfrom = 0#int(15. * 360)

sensitivity = []
precision = []
   
for subject in zip(records, annotations):

#for i in selection:
#      
#    subject = zip(records, annotations)[i]

    print('processing subject {}'.format(subject[1][-7:-4]))

    data = wfdb.rdrecord(subject[0][:-4], sampto=sampto)
    annotation = wfdb.rdann(subject[1][:-4], 'atr',
                            sampfrom=sampfrom,
                            sampto=sampto)

    sfreq = data.fs
    ecg = data.p_signal[:, 0]

    manupeaks = annotation.sample
    #algopeaks = peaks_signal(ecg, sfreq)
    algopeaks = rpeaks(ecg, sfreq)

    # tolerance for match between algorithmic and manual annotation (in sec)
    tolerance = 0.05
    comparitor = compare_annotations(manupeaks, algopeaks,
                                     int(np.rint(tolerance * sfreq)))
    tp = comparitor.tp
    fp = comparitor.fp

def make_dataset(records, width, savepath):
    """ Inside an array """
    # Prepare containers
    signals, labels = [], []

    # Iterate files
    for path in records:
        print 'Processing file:', path
        record = wf.rdsamp(path)
        annotations = wf.rdann(path, 'atr')

        # Extract pure signals
        data = record.p_signals

        # Convert each channel into labeled fragments
        signal, label = convert_data(data, annotations, width)

        # Cumulate
        signals.append(signal)
        labels.append(label)

    # Convert to one huge numpy.array
    signals = np.vstack(signals)
    labels = np.vstack(labels)

    # Write to disk

def read_file(file, participant):
    """Utility function
    """
    # Get signal
    data = pd.DataFrame({"ECG": wfdb.rdsamp(file[:-4])[0][:, 0]})
    data["Participant"] = "MIT-Arrhythmia_%.2i" %(participant)
    data["Sample"] = range(len(data))
    data["Sampling_Rate"] = 360
    data["Database"] = "MIT-Arrhythmia-x" if "x_mitdb" in file else "MIT-Arrhythmia"

    # getting annotations
    anno = wfdb.rdann(file[:-4], 'atr')
    anno = np.unique(anno.sample[np.in1d(anno.symbol, ['N', 'L', 'R', 'B', 'A', 'a', 'J', 'S', 'V', 'r', 'F', 'e', 'j', 'n', 'E', '/', 'f', 'Q', '?'])])
    anno = pd.DataFrame({"Rpeaks": anno})
    anno["Participant"] = "MIT-Arrhythmia_%.2i" %(participant)
    anno["Sampling_Rate"] = 360
    anno["Database"] = "MIT-Arrhythmia-x" if "x_mitdb" in file else "MIT-Arrhythmia"

    return data, anno

def show_path(path):
    """ As a plot """
    # Read in the data
    record = wf.rdsamp(path)
    annotation = wf.rdann(path, 'atr')
    data = record.p_signals
    cha = data[:, 0]
    print 'Channel type:', record.signame[0]
    times = np.arange(len(cha), dtype = float)
    times /= record.fs
    plt.plot(times, cha)
    plt.xlabel('Time [s]')
    plt.show()

def _get_sections(self):
        """Collect continuous arrhythmia sections."""
        # Empty dictionary for arrhythmia sections
        sections = list()

        # Loop through records
        for record_id in self.record_ids:

            # Import recording
            record = wfdb.rdrecord(os.path.join(self.raw_path, record_id))

            # Import annotations
            annotation = wfdb.rdann(os.path.join(self.raw_path, record_id), 'atr')

            # Get sample frequency
            fs = record.__dict__['fs']

            # Get waveform
            waveform = record.__dict__['p_signal']

            # labels
            labels = [label[1:] for label in annotation.__dict__['aux_note']]

            # Samples
            sample = annotation.__dict__['sample']

            # Loop through labels and collect sections
            for idx, label in enumerate(labels):