wavelet_test.py

import pywt
import csv
import pandas as pd
import numpy as np

def read_csv(filename):
  columns = defaultdict(list) # each value in each column is appended to a list

  with open(filename) as f:
    reader = csv.DictReader(f) # read rows into a dictionary format

    for row in reader: # read a row as {column1: value1, column2: value2,...}
      for (k,v) in row.items(): # go over each column name and value 
        columns[k].append(v) # append the value into the appropriate list
                             # based on column name k

  return columns

def csv_read_fast(filename, col_header_name, start_row, end_row):
  if end_row == -1:
    df = pd.read_csv(filename, sep=',', skipinitialspace=True, names=col_header_name, dtype=np.float32, skiprows=range(0, start_row)) #dtype={'a': np.float32}
  else:
    df = pd.read_csv(filename, sep=',', skipinitialspace=True, names=col_header_name, dtype=np.float32, skiprows=range(0, start_row), nrows=end_row-start_row) #dtype={'a': np.float32}
  
  return df

def waveFn(wavelet):
  if not isinstance(wavelet, pywt.Wavelet):
    return pywt.Wavelet(wavelet)
  else:
    return wavelet

# http://stackoverflow.com/questions/5707353/how-to-extend-pywavelets-to-work-with-n-dimensional-data
# given a single dimensional array ... returns the coefficients.
def wavedec(data, wavelet, mode='periodic'):
  wavelet = waveFn(wavelet)

  dLen = len(data)
  level = pywt.dwt_max_level(dLen, wavelet.dec_len)
  coeffs = np.zeros((1,level))

  a = data
  for idx in xrange(level):
    a, d = pywt.dwt(a, wavelet, mode)
    coeffs[0,idx] = a[0]
  return coeffs

def main():
  col_header_name = ['1', '2', '3']
  df = csv_read_fast('../data/wavelet_test.csv', col_header_name, 0, -1)
  res = wavedec(df['2'], 'haar', 'periodic')
  print res

if __name__ == "__main__":
  main()