def test_get_class_parts_limits(self):
l = model.get_class_parts_limits('class1')
self.assertSequenceEqual(l, [589.5])
l = model.get_class_parts_limits('class2')
self.assertSequenceEqual(l, [589.5, 1022.5, 1477.5])
l = model.get_class_parts_limits('class3a')
self.assertSequenceEqual(l, [589.5, 1022.5, 1477.5])
l = model.get_class_parts_limits('class3b')
self.assertSequenceEqual(l, [589.5, 1022.5, 1477.5])
def test_get_class_parts_limits(self):
l = model.get_class_parts_limits('class1')
self.assertSequenceEqual(l, [589.5])
l = model.get_class_parts_limits('class2')
self.assertSequenceEqual(l, [589.5, 1022.5, 1477.5])
l = model.get_class_parts_limits('class3a')
self.assertSequenceEqual(l, [589.5, 1022.5, 1477.5])
l = model.get_class_parts_limits('class3b')
self.assertSequenceEqual(l, [589.5, 1022.5, 1477.5])
def test_get_class_parts_limits_with_edges(self):
classes = model._get_wltc_data()['classes']
class_limits = {cls: model.get_class_parts_limits(cls, edges=True) for cls in classes.keys()}
for (cls, l) in class_limits.items():
self.assertEqual(l[0], 0, 'Class(%s): Left-edge not 0!'%cls)
for (cls, l) in class_limits.items():
self.assertEqual(l[-1], len(classes[cls]['cycle'])-0.5, 'Class(%s): Section Right-edge not len(cycle)!'%cls)
def test_get_class_parts_limits_with_edges(self):
classes = model._get_wltc_data()['classes']
class_limits = {cls: model.get_class_parts_limits(cls, edges=True) for cls in classes.keys()}
for (cls, l) in class_limits.items():
self.assertEqual(l[0], 0, 'Class(%s): Left-edge not 0!'%cls)
for (cls, l) in class_limits.items():
self.assertEqual(l[-1], len(classes[cls]['cycle'])-0.5, 'Class(%s): Section Right-edge not len(cycle)!'%cls)
def test_get_class_parts_limits_sorted(self):
classes = model._get_wltc_data()['classes']
class_limits = {
cls: model.get_class_parts_limits(cls, edges=True)
for cls in classes.keys()
}
for (cls, l) in class_limits.items():
self.assertSequenceEqual(l, sorted(l),
'Class(%s): Unsorted!' % cls)
def data_meanN_gears(cls_num):
"""
Constructs meanN vs gears, hierarchically grouped by: vehicle, python/heinz, class_part
:return: df like that::
python heinz
gear n gear n
1154 Low 2.020374 1078.312480 2.039049 1105.891341
Medium 3.577367 1364.552656 3.618938 1353.300231
High 4.028571 1694.902241 4.160440 1618.457143
ExtraHigh 4.266254 2532.802858 4.424149 2514.894737
1155 ...
"""
def means_by_parts(df, cols_in, cols_out):
df = df.ix[:, cols_in]
data = df.groupby(pd.cut(df_g.index, bins=parts)).mean()
data.index = part_names[:len(data.index)]
data.columns = cols_out
return data
cls_name = class_name_from_num(cls_num)
vehdata = wltpdb._run_the_experiments(transplant_original_gears=False, compare_results=False)
vehdata = vehdata[vehdata['class'] == cls_num] ## Filter only the Class.
cols_out = ['gear', 'n']
data = {}
for (veh_num, df_g, df_h) in wltpdb._file_pairs(wltpdb.gened_fname_glob):
try:
class_num = vehdata.loc[veh_num, 'class']
except KeyError:
continue
parts = model.get_class_parts_limits(cls_name, edges=True)
g_data = means_by_parts(df_g, ['gears', 'rpm'], cols_out)
h_data = means_by_parts(df_h, ['gear', 'n'], cols_out)
veh_data = pd.concat( (g_data, h_data), axis=1, keys=['python', 'heinz'])
data[veh_num] = veh_data
df = pd.concat(list(data.values()), axis=0, keys=data.keys())
return df
def test_get_class_parts_limits_sorted(self):
classes = model._get_wltc_data()['classes']
class_limits = {cls: model.get_class_parts_limits(cls, edges=True) for cls in classes.keys()}
for (cls, l) in class_limits.items():
self.assertSequenceEqual(l, sorted(l), 'Class(%s): Unsorted!'%cls)
