Skip to content

Commit 23ed61b

Browse files
timhoffmtimhoffm
committed
Remove usage of numpy recarray
Structured numpy arrays are more fundamental than recarrays and sufficient in all cases. Supersedes #26664.
1 parent 42336be commit 23ed61b

File tree

7 files changed

+34
-30
lines changed

7 files changed

+34
-30
lines changed

‎galleries/examples/lines_bars_and_markers/fill_between_alpha.py

+4-4
Original file line numberDiff line numberDiff line change
@@ -18,14 +18,14 @@
1818
import matplotlib.cbook as cbook
1919

2020
# load up some sample financial data
21-
r = cbook.get_sample_data('goog.npz')['price_data'].view(np.recarray)
21+
r = cbook.get_sample_data('goog.npz')['price_data']
2222
# create two subplots with the shared x and y axes
2323
fig, (ax1, ax2) = plt.subplots(1, 2, sharex=True, sharey=True)
2424

25-
pricemin = r.close.min()
25+
pricemin = r["close"].min()
2626

27-
ax1.plot(r.date, r.close, lw=2)
28-
ax2.fill_between(r.date, pricemin, r.close, alpha=0.7)
27+
ax1.plot(r["date"], r["close"], lw=2)
28+
ax2.fill_between(r["date"], pricemin, r["close"], alpha=0.7)
2929

3030
for ax in ax1, ax2:
3131
ax.grid(True)

‎galleries/examples/lines_bars_and_markers/scatter_demo2.py

+4-4
Original file line numberDiff line numberDiff line change
@@ -14,14 +14,14 @@
1414
# low, close, volume, adj_close from the mpl-data/sample_data directory. The
1515
# record array stores the date as an np.datetime64 with a day unit ('D') in
1616
# the date column.
17-
price_data = cbook.get_sample_data('goog.npz')['price_data'].view(np.recarray)
17+
price_data = cbook.get_sample_data('goog.npz')['price_data']
1818
price_data = price_data[-250:] # get the most recent 250 trading days
1919

20-
delta1 = np.diff(price_data.adj_close) / price_data.adj_close[:-1]
20+
delta1 = np.diff(price_data["adj_close"]) / price_data["adj_close"][:-1]
2121

2222
# Marker size in units of points^2
23-
volume = (15 * price_data.volume[:-2] / price_data.volume[0])**2
24-
close = 0.003 * price_data.close[:-2] / 0.003 * price_data.open[:-2]
23+
volume = (15 * price_data["volume"][:-2] / price_data["volume"][0])**2
24+
close = 0.003 * price_data["close"][:-2] / 0.003 * price_data["open"][:-2]
2525

2626
fig, ax = plt.subplots()
2727
ax.scatter(delta1[:-1], delta1[1:], c=close, s=volume, alpha=0.5)

‎galleries/examples/misc/keyword_plotting.py

+7-6
Original file line numberDiff line numberDiff line change
@@ -3,13 +3,14 @@
33
Plotting with keywords
44
======================
55
6-
There are some instances where you have data in a format that lets you
7-
access particular variables with strings: for example, with
8-
`numpy.recarray` or `pandas.DataFrame`.
6+
Some data structures, like dict, `structured numpy array
7+
<https://numpy.org/doc/stable/user/basics.rec.html#structured-arrays>`_
8+
or `pandas.DataFrame` provide access to labelled data via string index access
9+
``data[key]``.
910
10-
Matplotlib allows you to provide such an object with the ``data`` keyword
11-
argument. If provided, you may generate plots with the strings
12-
corresponding to these variables.
11+
For these data types, Matplotlib supports passing the whole datastructure via the
12+
``data`` keyword argument, and using the string names as plot function parameters,
13+
where you'd normally pass in your data.
1314
"""
1415

1516
import matplotlib.pyplot as plt

‎galleries/examples/ticks/centered_ticklabels.py

+3-4
Original file line numberDiff line numberDiff line change
@@ -18,18 +18,17 @@
1818
"""
1919

2020
import matplotlib.pyplot as plt
21-
import numpy as np
2221

2322
import matplotlib.cbook as cbook
2423
import matplotlib.dates as dates
2524
import matplotlib.ticker as ticker
2625

2726
# Load some financial data; Google's stock price
28-
r = cbook.get_sample_data('goog.npz')['price_data'].view(np.recarray)
27+
r = cbook.get_sample_data('goog.npz')['price_data']
2928
r = r[-250:] # get the last 250 days
3029

3130
fig, ax = plt.subplots()
32-
ax.plot(r.date, r.adj_close)
31+
ax.plot(r["date"], r["adj_close"])
3332

3433
ax.xaxis.set_major_locator(dates.MonthLocator())
3534
# 16 is a slight approximation since months differ in number of days.
@@ -45,5 +44,5 @@
4544
for label in ax.get_xticklabels(minor=True):
4645
label.set_horizontalalignment('center')
4746
imid = len(r) // 2
48-
ax.set_xlabel(str(r.date[imid].item().year))
47+
ax.set_xlabel(str(r["date"][imid].item().year))
4948
plt.show()

‎galleries/examples/ticks/date_index_formatter.py

+6-6
Original file line numberDiff line numberDiff line change
@@ -25,17 +25,17 @@
2525
# low, close, volume, adj_close from the mpl-data/sample_data directory. The
2626
# record array stores the date as an np.datetime64 with a day unit ('D') in
2727
# the date column (``r.date``).
28-
r = cbook.get_sample_data('goog.npz')['price_data'].view(np.recarray)
28+
r = cbook.get_sample_data('goog.npz')['price_data']
2929
r = r[:9] # get the first 9 days
3030

3131
fig, (ax1, ax2) = plt.subplots(nrows=2, figsize=(6, 6), layout='constrained')
3232
fig.get_layout_engine().set(hspace=0.15)
3333

3434
# First we'll do it the default way, with gaps on weekends
35-
ax1.plot(r.date, r.adj_close, 'o-')
35+
ax1.plot(r["date"], r["adj_close"], 'o-')
3636

3737
# Highlight gaps in daily data
38-
gaps = np.flatnonzero(np.diff(r.date) > np.timedelta64(1, 'D'))
38+
gaps = np.flatnonzero(np.diff(r["date"]) > np.timedelta64(1, 'D'))
3939
for gap in r[['date', 'adj_close']][np.stack((gaps, gaps + 1)).T]:
4040
ax1.plot(gap.date, gap.adj_close, 'w--', lw=2)
4141
ax1.legend(handles=[ml.Line2D([], [], ls='--', label='Gaps in daily data')])
@@ -51,12 +51,12 @@
5151
def format_date(x, _):
5252
try:
5353
# convert datetime64 to datetime, and use datetime's strftime:
54-
return r.date[round(x)].item().strftime('%a')
54+
return r["date"][round(x)].item().strftime('%a')
5555
except IndexError:
5656
pass
5757

5858
# Create an index plot (x defaults to range(len(y)) if omitted)
59-
ax2.plot(r.adj_close, 'o-')
59+
ax2.plot(r["adj_close"], 'o-')
6060

6161
ax2.set_title("Plot y at Index Coordinates Using Custom Formatter")
6262
ax2.xaxis.set_major_formatter(format_date) # internally creates FuncFormatter
@@ -79,6 +79,6 @@ def __call__(self, x, pos=0):
7979
pass
8080

8181

82-
ax2.xaxis.set_major_formatter(MyFormatter(r.date, '%a'))
82+
ax2.xaxis.set_major_formatter(MyFormatter(r["date"], '%a'))
8383

8484
plt.show()

‎galleries/tutorials/pyplot.py

+4-2
Original file line numberDiff line numberDiff line change
@@ -106,8 +106,10 @@
106106
# =============================
107107
#
108108
# There are some instances where you have data in a format that lets you
109-
# access particular variables with strings. For example, with
110-
# `numpy.recarray` or `pandas.DataFrame`.
109+
# access particular variables with strings. For example, with `structured arrays`_
110+
# or `pandas.DataFrame`.
111+
#
112+
# .. _structured arrays: https://numpy.org/doc/stable/user/basics.rec.html#structured-arrays
111113
#
112114
# Matplotlib allows you to provide such an object with
113115
# the ``data`` keyword argument. If provided, then you may generate plots with

‎galleries/users_explain/quick_start.py

+6-4
Original file line numberDiff line numberDiff line change
@@ -126,10 +126,12 @@
126126
# b = np.matrix([[1, 2], [3, 4]])
127127
# b_asarray = np.asarray(b)
128128
#
129-
# Most methods will also parse an addressable object like a *dict*, a
130-
# `numpy.recarray`, or a `pandas.DataFrame`. Matplotlib allows you to
131-
# provide the ``data`` keyword argument and generate plots passing the
132-
# strings corresponding to the *x* and *y* variables.
129+
# Most methods will also parse a string-indexable object like a *dict*, a
130+
# `structured numpy array
131+
# <https://numpy.org/doc/stable/user/basics.rec.html#structured-arrays>`_, or a
132+
# `pandas.DataFrame`. Matplotlib allows you to provide the ``data`` keyword argument
133+
# and generate plots passing the strings corresponding to the *x* and *y* variables.
134+
133135
np.random.seed(19680801) # seed the random number generator.
134136
data = {'a': np.arange(50),
135137
'c': np.random.randint(0, 50, 50),

0 commit comments

Comments
 (0)