ADGEfficiency

$ python data.py --nrows 1000

ADGEfficiency

data = pd.read_csv('data.csv', nrows=1000)

ADGEfficiency

#  data.py
parser.add_argument('--nrows', nargs='?')
args = parser.parse_args()
data = pd.read_csv('data.csv', nrows=args.nrows)
print(f'loaded {data.shape[0]} rows')

ADGEfficiency

nrows = 1000
data = pd.read_csv('data.csv', nrows=nrows)

ADGEfficiency

import os

home = os.environ['HOME']
path = os.path.join(home, 'adg'))
os.makedirs(path, exist_ok=True)
np.save(path, data)

ADGEfficiency

#  bad
data.drop('target', axis=1)

#  good
data = data.drop('target', axis=1)j

ADGEfficiency

import itertools

def calc_num_combinations(data):
    return len(list(itertools.permutations(data, len(data))))
  
def test_calc_num_combinations():
    test_data = (
        ((0, ), 1), ((0, 1), 2), ((0, 1, 2), 6)
    )
    for data, length in test_data:

ADGEfficiency

import pandas as pd

data = ['awake'] * 1000 + ['asleep'] * 500 + ['dreaming'] * 50
pd.Series(data).value_counts().plot(kind='bar')

ADGEfficiency

standardized = (data - np.mean(data)) / np.std(data)

ADGEfficiency

data = data[:1000]