def draw_precision_recall_curve(recall_domain, p_interp, title_pref=None, fnum=1, pnum=None, color=None):
import plottool as pt
if color is None:
color = (0.4, 1.0, 0.4) if pt.is_default_dark_bg() else (0.1, 0.4, 0.4)
if recall_domain is None:
recall_domain = np.array([])
p_interp = np.array([])
if recall_domain is None:
ave_p = -1.0 # np.nan
else:
ave_p = p_interp.sum() / p_interp.size
pt.plot2(
recall_domain,
p_interp,
marker="o--",
x_label="recall",
y_label="precision",
unitbox=True,
flipx=False,
color=color,
fnum=fnum,
pnum=pnum,
title="Interplated Precision Vs Recall\n" + "avep = %.3f" % ave_p,
)
def draw_precision_recall_curve(recall_domain,
p_interp,
title_pref=None,
fnum=1,
pnum=None,
color=None):
import plottool as pt
if color is None:
color = (0.4, 1.0, 0.4) if pt.is_default_dark_bg() else (0.1, 0.4, 0.4)
if recall_domain is None:
recall_domain = np.array([])
p_interp = np.array([])
if recall_domain is None:
ave_p = -1.0 # np.nan
else:
ave_p = p_interp.sum() / p_interp.size
pt.plot2(recall_domain,
p_interp,
marker='o--',
x_label='recall',
y_label='precision',
unitbox=True,
flipx=False,
color=color,
fnum=fnum,
pnum=pnum,
title='Interplated Precision Vs Recall\n' + 'avep = %.3f' % ave_p)
def distinct_colors(N, brightness=.878, randomize=True, hue_range=(0.0, 1.0), cmap_seed=None):
r"""
Args:
N (int):
brightness (float):
Returns:
list: RGB_tuples
CommandLine:
python -m plottool.color_funcs --test-distinct_colors --N 2 --show --hue-range=0.05,.95
python -m plottool.color_funcs --test-distinct_colors --N 3 --show --hue-range=0.05,.95
python -m plottool.color_funcs --test-distinct_colors --N 4 --show --hue-range=0.05,.95
python -m plottool.color_funcs --test-distinct_colors --N 3 --show --no-randomize
python -m plottool.color_funcs --test-distinct_colors --N 4 --show --no-randomize
python -m plottool.color_funcs --test-distinct_colors --N 20 --show
References:
http://blog.jianhuashao.com/2011/09/generate-n-distinct-colors.html
CommandLine:
python -m plottool.color_funcs --exec-distinct_colors --show
python -m plottool.color_funcs --exec-distinct_colors --show --no-randomize --N 50
python -m plottool.color_funcs --exec-distinct_colors --show --cmap_seed=foobar
Example:
>>> # ENABLE_DOCTEST
>>> from plottool.color_funcs import * # NOQA
>>> # build test data
>>> N = ut.get_argval('--N', int, 2)
>>> randomize = not ut.get_argflag('--no-randomize')
>>> brightness = 0.878
>>> # execute function
>>> cmap_seed = ut.get_argval('--cmap_seed', str, default=None)
>>> hue_range = ut.get_argval('--hue-range', list, default=(0.00, 1.0))
>>> RGB_tuples = distinct_colors(N, brightness, randomize, hue_range, cmap_seed=cmap_seed)
>>> # verify results
>>> assert len(RGB_tuples) == N
>>> result = str(RGB_tuples)
>>> print(result)
>>> ut.quit_if_noshow()
>>> color_list = RGB_tuples
>>> testshow_colors(color_list)
>>> ut.show_if_requested()
"""
# TODO: Add sin wave modulation to the sat and value
#import plottool as pt
if True:
import plottool as pt
# HACK for white figures
remove_yellow = not pt.is_default_dark_bg()
#if not pt.is_default_dark_bg():
# brightness = .8
use_jet = False
if use_jet:
import plottool as pt
cmap = pt.plt.cm.jet
RGB_tuples = list(map(tuple, cmap(np.linspace(0, 1, N))))
elif cmap_seed is not None:
# Randomized map based on a seed
#cmap_ = 'Set1'
#cmap_ = 'Dark2'
choices = [
#'Set1', 'Dark2',
'jet',
#'gist_rainbow',
#'rainbow',
#'gnuplot',
#'Accent'
]
cmap_hack = ut.get_argval('--cmap-hack', type_=str, default=None)
ncolor_hack = ut.get_argval('--ncolor-hack', type_=int, default=None)
if cmap_hack is not None:
choices = [cmap_hack]
if ncolor_hack is not None:
N = ncolor_hack
N_ = N
seed = sum(list(map(ord, ut.hashstr27(cmap_seed))))
rng = np.random.RandomState(seed + 48930)
cmap_str = rng.choice(choices, 1)[0]
#print('cmap_str = %r' % (cmap_str,))
cmap = pt.plt.cm.get_cmap(cmap_str)
#ut.hashstr27(cmap_seed)
#cmap_seed = 0
#pass
jitter = (rng.randn(N) / (rng.randn(100).max() / 2)).clip(-1, 1) * ((1 / (N ** 2)))
range_ = np.linspace(0, 1, N, endpoint=False)
#print('range_ = %r' % (range_,))
range_ = range_ + jitter
#print('range_ = %r' % (range_,))
while not (np.all(range_ >= 0) and np.all(range_ <= 1)):
range_[range_ < 0] = np.abs(range_[range_ < 0] )
range_[range_ > 1] = 2 - range_[range_ > 1]
#print('range_ = %r' % (range_,))
shift = rng.rand()
range_ = (range_ + shift) % 1
#print('jitter = %r' % (jitter,))
#print('shift = %r' % (shift,))
#print('range_ = %r' % (range_,))
if ncolor_hack is not None:
range_ = range_[0:N_]
RGB_tuples = list(map(tuple, cmap(range_)))
else:
sat = brightness
val = brightness
hmin, hmax = hue_range
if remove_yellow:
hue_skips = [(.13, .24)]
else:
hue_skips = []
hue_skip_ranges = [_[1] - _[0] for _ in hue_skips]
total_skip = sum(hue_skip_ranges)
hmax_ = hmax - total_skip
hue_list = np.linspace(hmin, hmax_, N, endpoint=False, dtype=np.float)
# Remove colors (like hard to see yellows) in specified ranges
for skip, range_ in zip(hue_skips, hue_skip_ranges):
hue_list = [hue if hue <= skip[0] else hue + range_ for hue in hue_list]
HSV_tuples = [(hue, sat, val) for hue in hue_list]
RGB_tuples = [colorsys.hsv_to_rgb(*x) for x in HSV_tuples]
if randomize:
ut.deterministic_shuffle(RGB_tuples)
return RGB_tuples
def draw_roc_curve(
fpr,
tpr,
fnum=None,
pnum=None,
marker="-",
target_tpr=None,
target_fpr=None,
thresholds=None,
color=None,
show_operating_point=False,
):
r"""
Args:
fpr (?):
tpr (?):
fnum (int): figure number(default = None)
pnum (tuple): plot number(default = None)
marker (str): (default = '-x')
target_tpr (None): (default = None)
target_fpr (None): (default = None)
thresholds (None): (default = None)
color (None): (default = None)
show_operating_point (bool): (default = False)
CommandLine:
python -m vtool.confusion --exec-draw_roc_curve --show --lightbg
Example:
>>> # DISABLE_DOCTEST
>>> from vtool.confusion import * # NOQA
>>> scores, labels = testdata_scores_labels()
>>> confusions = get_confusion_metrics(scores, labels)
>>> fpr = confusions.fpr
>>> tpr = confusions.tpr
>>> thresholds = confusions.thresholds
>>> fnum = None
>>> pnum = None
>>> marker = '-x'
>>> target_tpr = .85
>>> target_fpr = None
>>> color = None
>>> show_operating_point = True
>>> draw_roc_curve(fpr, tpr, fnum, pnum, marker, target_tpr, target_fpr,
>>> thresholds, color, show_operating_point)
>>> ut.show_if_requested()
"""
import plottool as pt
if fnum is None:
fnum = pt.next_fnum()
if color is None:
color = (0.4, 1.0, 0.4) if pt.is_default_dark_bg() else (0.1, 0.4, 0.4)
roc_auc = sklearn.metrics.auc(fpr, tpr)
title_suffix = ""
if target_fpr is not None:
# func = scipy.interpolate.interp1d(fpr, tpr, kind='linear', assume_sorted=False)
# func = scipy.interpolate.interp1d(xdata, ydata, kind='nearest', assume_sorted=False)
# interp_vals[interp_mask] = func(pt[interp_mask])
target_fpr = np.clip(target_fpr, 0, 1)
interp_tpr = interpolate_replbounds(fpr, tpr, target_fpr)
choice_tpr = interp_tpr
choice_fpr = target_fpr
elif target_tpr is not None:
target_tpr = np.clip(target_tpr, 0, 1)
interp_fpr = interpolate_replbounds(tpr, fpr, target_tpr)
choice_tpr = target_tpr
choice_fpr = interp_fpr
else:
choice_tpr = None
choice_fpr = None
if choice_fpr is not None:
choice_thresh = 0
if thresholds is not None:
try:
index = np.nonzero(tpr >= choice_tpr)[0][0]
except IndexError:
index = len(thresholds) - 1
choice_thresh = thresholds[index]
# percent = ut.scalar_str(choice_tpr * 100).split('.')[0]
# title_suffix = ', FPR%s=%05.2f%%' % (percent, choice_fpr)
title_suffix = ""
if show_operating_point:
title_suffix = ", fpr=%.2f, tpr=%.2f, thresh=%.2f" % (choice_fpr, choice_tpr, choice_thresh)
else:
title_suffix = ""
# if recall_domain is None:
# ave_p = np.nan
# else:
# ave_p = p_interp.sum() / p_interp.size
title = "Receiver operating characteristic\n" + "AUC=%.3f" % (roc_auc,)
title += title_suffix
pt.plot2(
fpr,
tpr,
marker=marker,
x_label="False Positive Rate",
y_label="True Positive Rate",
unitbox=True,
flipx=False,
color=color,
fnum=fnum,
pnum=pnum,
title=title,
)
if False:
# Interp does not work right because of duplicate values
# in xdomain
line_ = np.linspace(0.11, 0.9, 20)
# np.append([np.inf], np.diff(fpr)) > 0
# np.append([np.inf], np.diff(tpr)) > 0
unique_tpr_idxs = np.nonzero(np.append([np.inf], np.diff(tpr)) > 0)[0]
unique_fpr_idxs = np.nonzero(np.append([np.inf], np.diff(fpr)) > 0)[0]
pt.plt.plot(line_, interpolate_replbounds(fpr[unique_fpr_idxs], tpr[unique_fpr_idxs], line_), "b-x")
pt.plt.plot(interpolate_replbounds(tpr[unique_tpr_idxs], fpr[unique_tpr_idxs], line_), line_, "r-x")
if choice_fpr is not None:
pt.plot(choice_fpr, choice_tpr, "o", color=pt.PINK)
def draw_roc_curve(fpr,
tpr,
fnum=None,
pnum=None,
marker='-',
target_tpr=None,
target_fpr=None,
thresholds=None,
color=None,
show_operating_point=False):
r"""
Args:
fpr (?):
tpr (?):
fnum (int): figure number(default = None)
pnum (tuple): plot number(default = None)
marker (str): (default = '-x')
target_tpr (None): (default = None)
target_fpr (None): (default = None)
thresholds (None): (default = None)
color (None): (default = None)
show_operating_point (bool): (default = False)
CommandLine:
python -m vtool.confusion --exec-draw_roc_curve --show --lightbg
Example:
>>> # DISABLE_DOCTEST
>>> from vtool.confusion import * # NOQA
>>> scores, labels = testdata_scores_labels()
>>> confusions = get_confusion_metrics(scores, labels)
>>> fpr = confusions.fpr
>>> tpr = confusions.tpr
>>> thresholds = confusions.thresholds
>>> fnum = None
>>> pnum = None
>>> marker = '-x'
>>> target_tpr = .85
>>> target_fpr = None
>>> color = None
>>> show_operating_point = True
>>> draw_roc_curve(fpr, tpr, fnum, pnum, marker, target_tpr, target_fpr,
>>> thresholds, color, show_operating_point)
>>> ut.show_if_requested()
"""
import plottool as pt
if fnum is None:
fnum = pt.next_fnum()
if color is None:
color = (0.4, 1.0, 0.4) if pt.is_default_dark_bg() else (0.1, 0.4, 0.4)
roc_auc = sklearn.metrics.auc(fpr, tpr)
title_suffix = ''
if target_fpr is not None:
#func = scipy.interpolate.interp1d(fpr, tpr, kind='linear', assume_sorted=False)
#func = scipy.interpolate.interp1d(xdata, ydata, kind='nearest', assume_sorted=False)
#interp_vals[interp_mask] = func(pt[interp_mask])
target_fpr = np.clip(target_fpr, 0, 1)
interp_tpr = interpolate_replbounds(fpr, tpr, target_fpr)
choice_tpr = interp_tpr
choice_fpr = target_fpr
elif target_tpr is not None:
target_tpr = np.clip(target_tpr, 0, 1)
interp_fpr = interpolate_replbounds(tpr, fpr, target_tpr)
choice_tpr = target_tpr
choice_fpr = interp_fpr
else:
choice_tpr = None
choice_fpr = None
if choice_fpr is not None:
choice_thresh = 0
if thresholds is not None:
try:
index = np.nonzero(tpr >= choice_tpr)[0][0]
except IndexError:
index = len(thresholds) - 1
choice_thresh = thresholds[index]
#percent = ut.scalar_str(choice_tpr * 100).split('.')[0]
#title_suffix = ', FPR%s=%05.2f%%' % (percent, choice_fpr)
title_suffix = ''
if show_operating_point:
title_suffix = ', fpr=%.2f, tpr=%.2f, thresh=%.2f' % (
choice_fpr, choice_tpr, choice_thresh)
else:
title_suffix = ''
#if recall_domain is None:
# ave_p = np.nan
#else:
# ave_p = p_interp.sum() / p_interp.size
title = 'Receiver operating characteristic\n' + 'AUC=%.3f' % (roc_auc, )
title += title_suffix
pt.plot2(fpr,
tpr,
marker=marker,
x_label='False Positive Rate',
y_label='True Positive Rate',
unitbox=True,
flipx=False,
color=color,
fnum=fnum,
pnum=pnum,
title=title)
if False:
# Interp does not work right because of duplicate values
# in xdomain
line_ = np.linspace(.11, .9, 20)
#np.append([np.inf], np.diff(fpr)) > 0
#np.append([np.inf], np.diff(tpr)) > 0
unique_tpr_idxs = np.nonzero(np.append([np.inf], np.diff(tpr)) > 0)[0]
unique_fpr_idxs = np.nonzero(np.append([np.inf], np.diff(fpr)) > 0)[0]
pt.plt.plot(
line_,
interpolate_replbounds(fpr[unique_fpr_idxs], tpr[unique_fpr_idxs],
line_), 'b-x')
pt.plt.plot(
interpolate_replbounds(tpr[unique_tpr_idxs], fpr[unique_tpr_idxs],
line_), line_, 'r-x')
if choice_fpr is not None:
pt.plot(choice_fpr, choice_tpr, 'o', color=pt.PINK)
