settings.py

USERNAME = "[email protected]"
PASSWORD = "yourgooglepassword"

utilities.py

from __future__ import division
from pylab import *
import gspread
import pandas
from pint import UnitRegistry, UndefinedUnitError
from settings import USERNAME, PASSWORD
ureg = UnitRegistry()
Q_ = ureg.Quantity


def get_data(ss_name):
    gc = gspread.login(USERNAME, PASSWORD)
    wks = gc.open(ss_name).get_worksheet(0)
    list_of_lists = wks.get_all_values()
    head = list_of_lists[0]
    data = list_of_lists[1:]
    df = pandas.DataFrame(dict(zip(head, zip(*data))), columns=head)
    try:
        df['Timestamp'] = pandas.to_datetime(df['Timestamp'], dayfirst=True)
    except:
        pass
    return df


def subset_cols(df, q):
    return df[[c for c in df.columns if c.startswith(q)]]
    

def subset_units(df, column, unit):
    return df[df[column] == unit]
    

def get_mag(x):
    return Q_(x).to_base_units().magnitude


def get_units(x):
    return '{}'.format(Q_(x).to_base_units().units)


def normalise_units(df):
    for col in df.columns[1:]:
        try:
            df[col] = df[col].astype(float)
            continue
        except ValueError:
            pass
        try:
            df['{} - Converted'.format(col)] = df[col].map(get_mag)
            df['{} - Units'.format(col)] = df[col].map(get_units)
        except AttributeError:
            continue
        except UndefinedUnitError:
            continue
    return df


def w1_q3_get_data():
    return get_data("SCIE1000 Week 1: Estimates (Responses)")


def w1_q3():
    return normalise_units(w1_q3_get_data())


def _bwplot(ax, data):
    ax.boxplot(data, vert=False, labels=[""], widths=0.9)
    ax.axis('off')


def _kde(ax, data, **kwargs):
    x = data.plot(kind='kde', ax=ax, grid=False, c='red', **kwargs)
    ymax = x.lines[0].get_data()[1].max()
    ymin = x.lines[0].get_data()[1].min()
    ax.set_ylim([ymin,ymax*1.1])
    ax.set_ylim([ymin,ymax*1.1])
    ax.axis('off')


def _hist(ax, data, xlab):
    x = ax.hist(data, bins=10)
    xmax = x[0].max()
    ax.set_ylim([0,xmax])
    ax.set_xlim([x[1].min(),x[1].max()])
    xlabel(xlab)


def hist_bwplot(data, ptitle, log_data=False, xlab=''):
    data = log10(data) if log_data else data
    xlab = 'log10(%s)' % xlab if log_data else xlab

    figure(0, (7,5))

    #ax1 = subplot2grid((8,8), (0,0), colspan=8, rowspan=1)
    ax1 = subplot2grid((8,3), (0,0), colspan=3, rowspan=1)
    ax2 = subplot2grid((8,3), (1,0), colspan=3, rowspan=7, sharex=ax1)
    ax3 = ax2.twinx()

    _bwplot(ax1, data)
    _hist(ax2, data, xlab)
    _kde(ax3, data)

    ax1.set_title(ptitle)
    ax2.set_xlabel(xlab)
    ax2.xaxis.get_major_formatter().set_powerlimits((-3, 5))

    subplots_adjust(hspace=0, wspace=0)
    #tight_layout()


def w1_q3ai_plot(df, log_data=False):
    hist_bwplot(df['3(a)(i)'], 
                'Babies born in the world during the two hour SCIE1000 tutorial class', 
                log_data=log_data,
                xlab = 'Estimate of babies born / 2h')
    

def w1_q3aii_plot(df, log_data=False):
    data = subset_units(df, '3(a)(ii) - Units', 'gram')['3(a)(ii) - Converted']
    hist_bwplot(data, 
                'Total amount of pure alcohol consumed in Australia in a year',
                log_data=log_data,
                xlab = 'Estimate of Alcohol consumed per year in Aus (grams)')

    

def w2_7d_get_data():
    gc = gspread.login(USERNAME, PASSWORD)
    wks = gc.open("SCIE1000 Week2: BSA Estimates (Responses)").get_worksheet(0)
    list_of_lists = wks.get_all_values()
    head = list_of_lists[0]
    data = list_of_lists[1:]
    df = pandas.DataFrame(dict(zip(head, zip(*data))), columns=head)
    df[head[3:]] = df[head[3:]].astype(float)
    return df
    

def w2_7d_plot(df):
    r"^\d+(?:\.\d+) ?(?:L|mL|(?:(?:km|m|cm|mm)(?:^|\*\*)3)) ?/ ?(?:km|m|cm|mm)(?:^|\*\*)2$"

Week1.ipynb