== pdb howto ==
[https://realpython.com/python-debugging-pdb/]

==start debugger automatically on error==
python -m pdb -c continue myscript.py
[https://stackoverflow.com/a/2438834/766330]

==print all variables==
locals()
[https://stackoverflow.com/questions/21961693/how-to-print-all-variables-values-when-debugging-python-with-pdb-without-specif]

==current time as str==
import datetime
cur = datetime.datetime.now()
cur.strftime('%m%d%H%M')
[https://tecadmin.net/get-current-date-time-python/]

==simple line plots==
plt.style.use('seaborn-whitegrid')
[https://jakevdp.github.io/PythonDataScienceHandbook/04.01-simple-line-plots.html]

==flatten all lists recursively==
<pre>
li=[[1,[[2]],[[[3]]]],[['4'],{5:5}]]
flatten=lambda l: sum(map(flatten,l),[]) if isinstance(l,list) else [l]
print flatten(li)
</pre>
[https://stackoverflow.com/a/12474246/766330]

다음과 같은 기묘한 속성 때문이다
<pre>
>>> [] + [1]
[1]
>>> sum([[1]], [2])
[2, 1]
>>> sum([[1]], [])
[1]
</pre>

다음을 실행하면,
<pre>
depth = 0
def flatten_(x):
global depth
depth += 1
r = None
if isinstance(x, list):
print('>'*depth, 'list:', x)
y = sum(map(flatten_,x), [])
r = y
else:
print('>'*depth, 'not list:',x)
r = [x]
print('<'*depth, 'returning', r)
depth -= 1
return r

li=[[1,[[2]],[[[3]]]],[['4'],{5:5}]]
flatten_(li)
</pre>

결과는,
<pre>
> list: [[1, [[2]], [[[3]]]], [['4'], {5: 5}]]
>> list: [1, [[2]], [[[3]]]]
>>> not list: 1
<<< returning [1]
>>> list: [[2]]
>>>> list: [2]
>>>>> not list: 2
<<<<< returning [2]
<<<< returning [2]
<<< returning [2]
>>> list: [[[3]]]
>>>> list: [[3]]
>>>>> list: [3]
>>>>>> not list: 3
<<<<<< returning [3]
<<<<< returning [3]
<<<< returning [3]
<<< returning [3]
<< returning [1, 2, 3]
>> list: [['4'], {5: 5}]
>>> list: ['4']
>>>> not list: 4
<<<< returning ['4']
<<< returning ['4']
>>> not list: {5: 5}
<<< returning [{5: 5}]
<< returning ['4', {5: 5}]
< returning [1, 2, 3, '4', {5: 5}]
</pre>

==재정의==
[https://docs.python.org/3/reference/datamodel.html#object.__int__ 이 페이지]에 거의 다 있는듯함.<br>
나열하면 다음과 같다.
<pre>
__abs__ __add__ __aenter__ __aexit__
__aiter__ __and__ __anext__ __annotations__
__await__ __bases__ __bool__ __bytes__
__call__ __ceil__ __class__ __classcell__
__closure__ __code__ __complex__ __contains__
__defaults__ __del__ __delattr__ __delete__
__delitem__ __dict__ __dir__ __divmod__
__doc__ __enter__ __eq__ __exit__
__file__ __float__ __floor__ __floordiv__
__format__ __func__ __future__ __ge__
__get__ __getattr__ __getattribute__ __getitem__
__globals__ __gt__ __hash__ __iadd__
__iand__ __ifloordiv__ __ilshift__ __imatmul__
__imod__ __import__ __imul__ __index__
__init__ __instancecheck__ __int__ __invert__
__ior__ __ipow__ __irshift__ __isub__
__iter__ __itruediv__ __ixor__ __kwdefaults__
__le__ __len__ __lshift__ __lt__
__matmul__ __missing__ __mod__ __module__
__mro__ __mul__ __name__ __ne__
__neg__ __new__ __next__ __objclass__
__or__ __pos__ __pow__ __prepare__
__qualname__ __radd__ __rand__ __rdivmod__
__repr__ __reversed__ __rfloordiv__ __rlshift__
__rmatmul__ __rmod__ __rmul__ __ror__
__round__ __rpow__ __rrshift__ __rshift__
__rsub__ __rtruediv__ __rxor__ __self__
__set__ __setattr__ __setitem__ __slots__
__str__ __sub__ __subclasscheck__ __traceback__
__truediv__ __trunc__ __weakref__ __xor__
</pre>
<code>grep -ho -e '__[a-z]*__' x.txt | sort -u | xargs -n 4 printf '%-20s%-20s%-20s%-20s\n'</code>

==2.7과 3.6 동시에 쓰기==
[https://conda.io/docs/user-guide/tasks/manage-python.html 참고링크] [https://conda.io/docs/user-guide/tasks/manage-environments.html#activate-env 참고링크2]

아래로 만들고,
conda create -n py36 python=3.6 anaconda
아래로 activate
source activate py36

환경 리스트
conda env list

==<code>__call__</code> 클래스로도, 함수로도 쓸 수 있음.==
<pre>The first is used to initialise newly created object, and receives arguments used to do that:

class Foo:
def __init__(self, a, b, c):
# ...

x = Foo(1, 2, 3) # __init__

The second implements function call operator.

class Foo:
def __call__(self, a, b, c):
# ...

x = Foo()
x(1, 2, 3) # __call__</pre>
[https://stackoverflow.com/a/9663601/766330]

==jupyter notebook matplotlib on OSX==
Insert below code at the head of the note.
import matplotlib as mpl
mpl.use('TkAgg')
[https://stackoverflow.com/questions/21784641/installation-issue-with-matplotlib-python#comment56913201_21789908]

==한글외 문자 제거==
<pre>hangul = re.compile('[^ ㄱ-ㅣ가-힣]+') # 한글과 띄어쓰기를 제외한 모든 글자
# hangul = re.compile('[^ \u3131-\u3163\uac00-\ud7a3]+') # 위와 동일
result = hangul.sub('', target_text) # 한글과 띄어쓰기를 제외한 모든 부분을 제거</pre>
http://jokergt.tistory.com/52

==ternary operator==
a if condition else b
[https://stackoverflow.com/a/394814/766330]

==print color to terminal stdout==
print('\x1b[6;30;42m' + 'Success!' + '\x1b[0m')
[https://stackoverflow.com/a/21786287/766330]

ansi
<pre>RED = '\033[91m'
GREEN = '\033[92m'
YELLOW = '\033[93m'
LIGHT_PURPLE = '\033[94m'
PURPLE = '\033[95m'
END = '\033[0m'</pre>
코드로도 된다고 하는데 그럴필요까지야.
<pre>from __future__ import print_function
from colorprint import *

print('Hello', 'world', color='blue', end='', sep=', ')
print('!', color='red', format=['bold', ‘blink'])</pre>
[https://stackoverflow.com/a/31310228/766330]

==filter==
filter(None, list)
None이나 $0$을 제외하게 된다.[https://stackoverflow.com/q/16096754/766330]

==ipython (jupyter notebook) ==
===load a file===
%load f.py
===save current cell to a file===
%%writefile f.py
===shell command===
!ls
!cat “file”
[https://stackoverflow.com/questions/21034373/how-to-load-edit-run-save-text-files-py-into-an-ipython-notebook-cell]

==Multiple input stream==
import fileinput
for line in fileinput.input():
process(line)
https://docs.python.org/2/library/fileinput.html#module-fileinput

==file position==
def skip_comments(f):
while True:
pos = f.tell()
line = f.readline().strip()
if not line.startswith('#'):
f.seek(pos) #rewind
break
{{c|tell}}로 현재 위치를 얻고, {{c|seek}}로 set.
>>> f = open('workfile', 'r+')
>>> f.write('0123456789abcdef')
>>> f.seek(5) # Go to the 6th byte in the file
>>> f.read(1)
'5'
>>> f.seek(-3, 2) # Go to the 3rd byte before the end
>>> f.read(1)
'd'
https://docs.python.org/2/tutorial/inputoutput.html#methods-of-file-objects

==Interactive session detection==
sys.stdin.isatty()
http://stackoverflow.com/a/7576567/766330

==Temporary file or directory==
tempfile.TemporaryFile(<nowiki>[mode='w+b'[, bufsize=-1[, suffix=''[, prefix='tmp'[, dir=None]]]]]</nowiki>) [https://docs.python.org/2/library/tempfile.html#tempfile.TemporaryFile]
임시파일을 만든다. 아래를 쓸것.
tempfile.NamedTemporaryFile(<nowiki>[mode='w+b'[, bufsize=-1[, suffix=''[, prefix='tmp'[, dir=None[, delete=True]]]]]]</nowiki>) [https://docs.python.org/2/library/tempfile.html#tempfile.NamedTemporaryFile]
위와 동일하나 visible name의 파일을 만든다. 이름은 반환값의 name속성으로 받을 수 있다.
<br>주의할점은, {{c|delete}}속성의 기본값이 {{c|True}}여서 쓰여진 후 곧바로 지워진다. 계속 쓰려면 해당속성을 {{c|False}}로 해야 한다.
tempfile.mkdtemp(<nowiki>[suffix=''[, prefix='tmp'[, dir=None]]]</nowiki>) [https://docs.python.org/2/library/tempfile.html#tempfile.mkdtemp]
임시디렉토리를 만들고, 자동으로 지우지 않는다.

==Check syntax without running==
python -m py_compile script.py
http://stackoverflow.com/a/8437597/766330

==Time==
===current time===
>>> import time
>>> time.time()
1491792653.410371

>>> import datetime
>>> datetime.datetime.fromtimestamp(time.time()).strftime('%Y-%m-%d %H:%M:%S')
'2017-04-10 11:51:17'
>>> datetime.datetime.utcnow()
datetime.datetime(2017, 4, 10, 2, 51, 34, 356682)
>>> datetime.datetime.now()
datetime.datetime(2017, 4, 10, 11, 51, 36, 572681)
>>> datetime.datetime.now().strftime("%A, %d. %B %Y %I:%M%p")
'Monday, 10. April 2017 11:51AM'
[http://stackoverflow.com/questions/13890935/does-pythons-time-time-return-the-local-or-utc-timestamp]<br>
if you want UTC, use <c>utcfromtimestamp</c> instead of <c>fromtimestamp</c> [https://stackoverflow.com/a/37188257/766330]

===yesterday===
>>> from datetime import date, timedelta
>>> yesterday = date.today() - timedelta(1)
>>> print yesterday.strftime('%m%d%y')
'110909'
[http://stackoverflow.com/a/1712127/766330]

[https://docs.python.org/2/library/datetime.html#strftime-and-strptime-behavior strftime format manual]

==String replace==
==={{c|str.replace}}===
str.replace(old, new[, max]) [https://www.tutorialspoint.com/python/string_replace.htm]
==={{c|re.sub}}===
>>><nowiki> re.sub(r'def\s+([a-zA-Z_][a-zA-Z_0-9]*)\s*$\s*$:',
... r'static PyObject*\npy_\1(void)\n{',
... 'def myfunc():')
'static PyObject*\npy_myfunc(void)\n{'</nowiki> [https://docs.python.org/2/library/re.html#re.sub ]
commented version example
line = re.sub(r"""
(?x) # Use free-spacing mode.
< # Match a literal '<'
/? # Optionally match a '/'
\[ # Match a literal '['
\d+ # Match one or more digits
> # Match a literal '>'
""", "", line) [http://stackoverflow.com/a/5658439/766330]

=={{c|is}} and {{c|<nowiki>==</nowiki>}}==
{{c|is}} is identity testing, {{c|<nowiki>==</nowiki>}} is equality testing. what happens in your code would be emulated in the interpreter like this:
>>> a = 'pub'
>>> b = <nowiki>''</nowiki>.join(['p', 'u', 'b'])
>>> a == b
True
>>> a is b
False
[http://stackoverflow.com/a/1504742/766330]

==Access index in {{c|for}} loop==
{{c|enumerate}}
for idx, val in enumerate(ints):
print(idx, val)
[http://stackoverflow.com/a/522578/766330]

==Max integer==
sys.maxint
[http://stackoverflow.com/a/7604981/766330]

python 3에서는 안된다고 한다. <del>어차피 안쓰니 상관없지만</del> 그래서 다음과 같이 한다고.
<pre>import sys
max = sys.maxsize
min = -sys.maxsize</pre>
[https://stackoverflow.com/a/42544543/766330]

다음도 간단하고 좋은 것 같다. 2,3에서 모두 동작
float('inf')
float('-inf')
[https://stackoverflow.com/a/37300370/766330]

==Argument parsing==
use [https://docs.python.org/2.7/library/argparse.html#action argparse]

=={{c|range}} by step==
>>> step = .1
>>> N = 10 # number of data points
>>> [ x / pow(step, -1) for x in range(0, N + 1) ]
[0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0]
[http://stackoverflow.com/a/41079203/766330]

==File existence==
os.path.exists(file_path)
[http://stackoverflow.com/a/82846/766330] [https://docs.python.org/2/library/os.path.html#os.path.exists]

==Define source code encoding==
첫째나 둘째줄에,
# coding=<encoding name>
보통 에디터들은 다음 형식을 이해함.
#!/usr/bin/python
# -*- coding: <encoding name> -*-
혹은
#!/usr/bin/python
# vim: set fileencoding=<encoding name> :
정확히는, 첫째나 둘째줄이 아래 정규식에 맞아야 함.
<nowiki>^[ \t\v]*#.*?coding[:=][ \t]*([-_.a-zA-Z0-9]+)</nowiki>
https://www.python.org/dev/peps/pep-0263/

==encoding==
기본적으로 파이썬은 유니코드를 사용한다. encode/decode할 때, encode하면 유니코드로부터 다른 인코딩으로 가고, decode하면 다른 인코딩으로부터 유니코드로 온다는 얘기. 그래서 unicode().decode()는 매우 쓸데가 없다. [https://stackoverflow.com/a/449281/766330] 보통은 기본 인코딩을 utf8로 쓰므로 파이썬 코드상에서 가끔 한글이 이상한 동작을 보이면 다 <code>decode('utf-8')</code>붙여주면 된다. 하이픈 없이 utf8이라고 써도 된다.

주의할점은, encode를 부르면 암묵적으로 decode가 먼저 불린다. 따라서 (쓸일도 거의 없지만) encode할 때는 원래 문자열을 decode해주는 것이 안전하다. 안해주면 ascii codec으로 디코딩을 먼저 시도하는데, 한글의 경우는 잘 안되므로 <code>UnicodeEncodeError: 'ascii' codec can't encode characters in position blabla: ordinal not in range(128)</code> 요런 친숙한(?) 에러가 난다. 일단 유니코드로 다 바꾸고 다른 인코딩으로 넘어간다고 생각하면 됨. [https://stackoverflow.com/questions/10561923/unicodedecodeerror-ascii-codec-cant-decode-byte-0xef-in-position-1]

스트링format에서, %s로 유니코드를 받으려면 format string도 unicode여야 한다. <code>u'%s'</code> 이렇게. 그래서 섞어 써야 할 때는, ①format string을 놔두고, 유니코드를 utf8로 디코드 해서 넘기든지, ② format string을 unicode로 주고 utf8문자열에 죄다 <code>decode('utf-8')</code>을 붙인다.

시간이 너무 남아돌면 기본 매뉴얼[https://docs.python.org/2/howto/unicode.html]을 훑어본다.

[https://stackoverflow.com/a/492711/766330 stackoverflow]에 훌륭한 규칙이 나옴: A rule of thumb is: Always use Unicode internally. Decode what you receive, and encode what you send. 받는건 무조건 decode, 내보내는건 무조건 encode. 내부는 유니코드로 고정이기 때문.

==Reading another encoding from a file==
<shl>
import codecs
f = codecs.open('unicode.rst', encoding='utf-8')
for line in f:
print repr(line)
</shl>
https://stackoverflow.com/a/147756/766330

=={{c|randint}}==
from random import randint
print(randint(0,9))
0~9 inclusive<br>
https://docs.python.org/2/library/random.html#random.randint

=={{c|groupby}} example==
from itertools import groupby

things = [("animal", "bear"), ("animal", "duck"), ("plant", "cactus"), ("vehicle", "speed boat"), ("vehicle", "school bus")]

for key, group in groupby(things, lambda x: x[0]):
for thing in group:
print "A %s is a %s." % (thing[1], key)
print " "
output
A bear is a animal.
A duck is a animal.

A cactus is a plant.

A speed boat is a vehicle.
A school bus is a vehicle.

http://stackoverflow.com/a/7286/766330

==Suppress scientific notation==
use {{c|%f}}
'%f' % (x/y)
http://stackoverflow.com/a/658777/766330

==To ignore exception==
except:
pass
http://stackoverflow.com/a/574734/766330

==Delete items from a dictionary while iterating over it==
iteration중에는 지울 수 없다. 두번 돌아야 함.
remove = [k for k in mydict if k == val]
for k in remove: del mydict[k]
http://stackoverflow.com/a/5385196/766330

==Print without newline==
import sys
sys.stdout.write('.')
http://stackoverflow.com/a/493399/766330

=={{c|if else}} in list comprehension==
<nowiki>[ unicode(x.strip()) if x is not None else '' for x in row ]</nowiki>
http://stackoverflow.com/a/4260304/766330

==filtering a dictionary according to an arbitrary condition==
use dict comprehension
{k: v for k, v in points.iteritems() if v[0] < 5 and v[1] < 5}
https://stackoverflow.com/a/16589453/766330

==Executing multi-line statements within PDB==
(pdb) <span style='color:red'>!import code; code.interact(local=vars()) </span>
Python 2.6.5 (r265:79063, Apr 16 2010, 13:57:41)
[GCC 4.4.3] on linux2
Type "help", "copyright", "credits" or "license" for more information.
(InteractiveConsole)
>>>
https://stackoverflow.com/a/8387484/766330

==Converting a IPython Notebook into a Python file via commandline==
$ jupyter nbconvert --to script [YOUR_NOTEBOOK].ipynb
or in ipython notebook
!jupyter nbconvert --to script [YOUR_NOTEBOOK].ipynb
https://stackoverflow.com/a/19779226/766330

==Write float(or int) to a binary file==
<pre>import struct
s = struct.pack('f'*len(floats), *floats)
f = open('file','wb')
f.write(s)
f.close()</pre>
To read this from [[C]], refer [http://iph.kr/wiki/index.php/C#Read.2Fwrite_float_binary here]

https://stackoverflow.com/a/807881/766330

==cross reference between files==
Use following statements according to appropriate order
from __main__ import *

https://stackoverflow.com/a/16068078/766330

==pdf file merge==
from PyPDF2 import PdfMerger

pdfs = ['file1.pdf', 'file2.pdf', 'file3.pdf', 'file4.pdf']

merger = PdfMerger()

for pdf in pdfs:
merger.append(pdf)

merger.write("result.pdf")
merger.close()

page range를 줄 수도 있다고 함. 별로 직관적이지는 않아보임

그냥 파일 여러개를 합하는것을 명령어로 할 수 있는 것:
python -m fitz join -o result.pdf file1.pdf file2.pdf file3.pdf

PyMuPdf[https://github.com/pymupdf/PyMuPDF]참고

[https://stackoverflow.com/a/37945454/766330]

==natural sort==
use <kbd>natsorted</kbd>
<pre>
>>> from natsort import natsorted, ns
>>> x = ['Elm11', 'Elm12', 'Elm2', 'elm0', 'elm1', 'elm10', 'elm13', 'elm9']
>>> natsorted(x, key=lambda y: y.lower())
['elm0', 'elm1', 'Elm2', 'elm9', 'elm10', 'Elm11', 'Elm12', 'elm13']
>>> natsorted(x, alg=ns.IGNORECASE) # or alg=ns.IC
['elm0', 'elm1', 'Elm2', 'elm9', 'elm10', 'Elm11', 'Elm12', 'elm13']
</pre>
[https://stackoverflow.com/a/18415320/766330]

Python

2023-02-20T04:06:13Z

Admin: /* cross reference between files */

Machine Learning

2021-12-10T06:25:53Z

Admin:

==by themes==
* [[Recommendation]]
* [[Face]]
* [[Person re-identification]]
* [[segmentation]]

==posts==
* [[conv1d]]
* [[naive gradient descent]]
* [https://www.notion.so/nll-loss-57987c7b5f7342e4b6bc929c9b3be587 nll loss]

==ril==
* https://medium.com/technologymadeeasy/the-best-explanation-of-convolutional-neural-networks-on-the-internet-fbb8b1ad5df8
* http://nmhkahn.github.io/Casestudy-CNN
* https://stats.stackexchange.com/questions/205150/how-do-bottleneck-architectures-work-in-neural-networks
* https://www.quora.com/What-exactly-is-the-degradation-problem-that-Deep-Residual-Networks-try-to-alleviate
* dl with torch
* bias 붙여버리면 inv는 어케 구하나? D=0되지 않나? 구할필요 없나?
* [https://github.com/pytorch/examples pytorch examples ]
* [https://arxiv.org/abs/1603.05027 Identity Mappings in Deep Residual Networks] arXiv:1603.05027
* [http://www.fast.ai/2017/07/28/deep-learning-part-two-launch/ cutting edge deeplearning for coders]

==nets==
* [[1118_Convolutional_Neural_Network|CNN]]
* [[AlexNet]]
* [[dilated cnn]]
* [[pathnet]]
* [[ResNet]]
* [[Fast RCNN]]
* [[Faster RCNN]]
* [[R-FCN]]
* [[GoogLeNet|Inception]] (GoogLeNet)
* [[fully convolutional networks]]
* [[FractalNets]]
* [[highway networks]]
* [[Memory networks]]
* [[DenseNet]]
* [[Network in Network|NIN]]
* [[Deeply Supervised Network|DSN]]
* [[Ladder Networks]]
* [[Deeply-Fused Nets|DFNs]]
* [[YOLO]]

==general==
* [[0811 Affinity Propagation|Affinity Propagation]], [[Apcluster sparse]]
* <del>[[CUDA기타설치]]</del>
* [[Learning to learn by GD by GD]]
* [[Generative Models]] (GAN, VAE, etc)
* [[Batch Normalization]]
* [[Mean Average Precision]]
* [https://medium.com/@kailashahirwar/essential-cheat-sheets-for-machine-learning-and-deep-learning-researchers-efb6a8ebd2e5 Essential Cheat Sheets for Machine Learning and Deep Learning Engineers]
* [http://fa.bianp.net/blog/2014/surrogate-loss-functions-in-machine-learning/ What is surrogate loss?]
* [[Exponential Linear Unit]]
* [[Neural net이 working하지 않는 37가지 이유]]
* [[deconvolution]]
* [[Sparse coding]]
* [http://mxnet.io/model_zoo/index.html MXNet Model Zoo]
* [[logistic regression]]
*[[0926 information bottleneck|information bottleneck]]
*[[0928 Artificial Curiosity|Artificial Curiosity]]
* sklearn examples
** [[sklearn preprocessing]]

==x==
* [[Isolation Forest]]

2020-03-03T09:21:18Z

Admin: /* Regexp with magic */

Matplotlib,seaborn

2020-02-28T09:46:31Z

Admin: /* Default style for more than 6 data series in seaborn */

== xticks on top ==
ref. https://matplotlib.org/3.1.3/gallery/ticks_and_spines/tick_xlabel_top.html

<source>
import matplotlib.pyplot as plt
import numpy as np

plt.rcParams['xtick.bottom'] = plt.rcParams['xtick.labelbottom'] = False
plt.rcParams['xtick.top'] = plt.rcParams['xtick.labeltop'] = True
</source>

== use of ttf with seaborn ==
ref. https://matplotlib.org/gallery/api/font_file.html , https://seaborn.pydata.org/generated/seaborn.set.html
<source>
import os
from matplotlib import font_manager as fm, rcParams
import matplotlib.pyplot as plt

fig, ax = plt.subplots()

fpath = os.path.join(rcParams["datapath"], "fonts/ttf/cmr10.ttf")
prop = fm.FontProperties(fname=fpath)
</source>
<source>
p = sns.lineplot(ax=ax, data=data, palette='tab10', linewidth=1, dashes=dash_styles)
plt.setp(p.get_xticklabels(), rotation=90, fontproperties=prop)
</source>

== rotate xticks ==
ref. https://www.drawingfromdata.com/how-to-rotate-axis-labels-in-seaborn-and-matplotlib , https://stackoverflow.com/questions/44954123/rotate-xtick-labels-in-seaborn-boxplot
<source>
ax = sns.boxplot(x='categories', y='oxygen', hue='target', data=df)
plt.setp(ax.get_xticklabels(), rotation=45)
</source>

== Default style for more than 6 data series in seaborn ==
ref. https://github.com/mwaskom/seaborn/issues/1513

<source>
dash_styles = ["",
(4, 1.5),
(1, 1),
(3, 1, 1.5, 1),
(5, 1, 1, 1),
(5, 1, 2, 1, 2, 1),
(2, 2, 3, 1.5),
(1, 2.5, 3, 1.2)]

sns.relplot(..., dashes=dash_styles,...)
</source>

== set title size ==
ref. https://stackoverflow.com/a/36222162/766330
<source>
p = sns.lineplot(ax=ax, data=data, palette='tab10', linewidth=1, dashes=dash_styles)
p.axes.set_title("EN",fontsize=50)
</source>

Matplotlib,seaborn

2020-02-28T09:42:23Z

Admin: /* rotate xticks */

== xticks on top ==
ref. https://matplotlib.org/3.1.3/gallery/ticks_and_spines/tick_xlabel_top.html

<source>
import matplotlib.pyplot as plt
import numpy as np

plt.rcParams['xtick.bottom'] = plt.rcParams['xtick.labelbottom'] = False
plt.rcParams['xtick.top'] = plt.rcParams['xtick.labeltop'] = True
</source>

== use of ttf with seaborn ==
ref. https://matplotlib.org/gallery/api/font_file.html , https://seaborn.pydata.org/generated/seaborn.set.html
<source>
import os
from matplotlib import font_manager as fm, rcParams
import matplotlib.pyplot as plt

fig, ax = plt.subplots()

fpath = os.path.join(rcParams["datapath"], "fonts/ttf/cmr10.ttf")
prop = fm.FontProperties(fname=fpath)
</source>
<source>
p = sns.lineplot(ax=ax, data=data, palette='tab10', linewidth=1, dashes=dash_styles)
plt.setp(p.get_xticklabels(), rotation=90, fontproperties=prop)
</source>

== rotate xticks ==
ref. https://www.drawingfromdata.com/how-to-rotate-axis-labels-in-seaborn-and-matplotlib , https://stackoverflow.com/questions/44954123/rotate-xtick-labels-in-seaborn-boxplot
<source>
ax = sns.boxplot(x='categories', y='oxygen', hue='target', data=df)
plt.setp(ax.get_xticklabels(), rotation=45)
</source>

== Default style for more than 6 data series in seaborn ==
ref. https://github.com/mwaskom/seaborn/issues/1513

<source>
dash_styles = ["",
(4, 1.5),
(1, 1),
(3, 1, 1.5, 1),
(5, 1, 1, 1),
(5, 1, 2, 1, 2, 1),
(2, 2, 3, 1.5),
(1, 2.5, 3, 1.2)]

sns.relplot(..., dashes=dash_styles,...)
</source>

Matplotlib,seaborn

2020-02-28T09:41:35Z

Admin: /* use of ttf with seaborn */

Matplotlib,seaborn

2020-02-28T09:40:00Z

Admin:

2020-01-21T06:00:14Z

Admin: 새 문서: ==2020== :0121 unfold

==2020==
:[[0121 unfold]]