New action: "All-column" vis when only few columns in dataframe

[dorisjlee]

When there are only two columns in a dataframe involving a quantitative and a temporal attribute, we display separate univariate charts. The better chart to display here would be a line chart with both attributes. We need to change the action logic to account for this use case.

tmax = pd.read_csv('https://raw.githubusercontent.com/koldunovn/python_for_geosciences/master/DelhiTmax.txt', delimiter=r"\s+", parse_dates=[[0,1,2]], header=None)
tmax.columns = ['Date', 'Temp']

This notebook has a good use case example for time series.

[dorisjlee]

A similar use case is when there are 1 dimension and 2 quantitative variables, the visualization shows as two separate charts.


A single visualization like this would be more appropriate

Example code:

df = pd.DataFrame({'nPts': {0: 49999, 1: 71174, 2: 101317, 3: 144224, 4: 205303, 5: 292249, 6: 416016, 7: 592198, 8: 842993, 9: 1200000}, 't_heatmap': {0: 0.15121674537658691, 1: 0.1811518669128418, 2: 0.2179429531097412, 3: 0.2787730693817139, 4: 0.3973350524902344, 5: 0.4233138561248779, 6: 0.580251932144165, 7: 0.7928099632263184, 8: 1.0876789093017578, 9: 1.6242818832397459}, 't_color_heatmap': {0: 0.14242982864379886, 1: 0.18866705894470212, 2: 0.1566781997680664, 3: 0.16737699508666992, 4: 0.19900894165039065, 5: 0.2701129913330078, 6: 0.2533812522888184, 7: 0.37183785438537603, 8: 0.3830866813659668, 9: 0.39321017265319824}, 't_bar': {0: 0.01940608024597168, 1: 0.02618718147277832, 2: 0.024693727493286133, 3: 0.029685020446777344, 4: 0.03471803665161133, 5: 0.04173588752746582, 6: 0.04706382751464844, 7: 0.0667569637298584, 8: 0.08260798454284668, 9: 0.1006929874420166}, 't_cbar': {0: 0.04185795783996582, 1: 0.050965070724487305, 2: 0.052091121673583984, 3: 0.0610501766204834, 4: 0.07504606246948242, 5: 0.09924101829528807, 6: 0.10392117500305176, 7: 0.13044309616088867, 8: 0.18137121200561526, 9: 0.2069528102874756}, 't_hist': {0: 0.01846003532409668, 1: 0.018136978149414062, 2: 0.018748998641967773, 3: 0.01876473426818848, 4: 0.02434706687927246, 5: 0.025368213653564453, 6: 0.02530217170715332, 7: 0.02823114395141602, 8: 0.029034852981567383, 9: 0.034781217575073235}, 't_scatter': {0: 0.8179378509521484, 1: 1.0982017517089844, 2: 1.4875690937042236, 3: 2.146117925643921, 4: 3.1123709678649902, 5: 3.92416787147522, 6: 6.097048044204713, 7: 8.803220987319945, 8: 12.054464101791382, 9: 17.740845680236816}, 't_color_scatter': {0: 1.3474130630493164, 1: 1.7953579425811768, 2: 2.6268541812896733, 3: 3.646371126174927, 4: 4.901940107345581, 5: 6.974237203598023, 6: 10.197483777999876, 7: 14.134275913238525, 8: 21.13951110839844, 9: 27.127063989639282}})
pdf = df.melt(id_vars=["nPts"],value_vars=['t_heatmap', 't_color_heatmap', 't_bar', 't_cbar', 't_hist','t_scatter', 't_color_scatter'], var_name="type", value_name='time')