A ranar 9 ga Janairu, Pandas 1.0.0rc ya fito. Sigar da ta gabata na ɗakin karatu shine 0.25.

Babban fitowar farko ta ƙunshi sabbin abubuwa masu girma da yawa, gami da ingantattun bayanan bayanan atomatik, ƙarin tsarin fitarwa, sabbin nau'ikan bayanai, har ma da sabon rukunin yanar gizo.

Ana iya duba duk canje-canje a nan, a cikin labarin za mu iyakance kanmu ga ƙananan ƙananan fasaha na fasaha mafi mahimmanci.

Kuna iya shigar da ɗakin karatu kamar yadda kuka saba amfani da shi pip, amma tun a lokacin rubuta Pandas 1.0 yana nan saki dan takara, kuna buƙatar bayyana sigar a sarari:

pip install --upgrade pandas==1.0.0rc0

Yi hankali: tunda wannan babban saki ne, sabuntawar na iya karya tsohuwar lambar!

Af, an daina goyon bayan Python 2 gaba ɗaya tun wannan sigar (me zai iya zama kyakkyawan dalili sabunta - kimanin. fassarar). Pandas 1.0 yana buƙatar aƙalla Python 3.6+, don haka idan ba ku da tabbas, duba wacce kuka shigar:

$ pip --version
pip 19.3.1 from /usr/local/lib/python3.7/site-packages/pip (python 3.7)

$ python --version
Python 3.7.5

Hanya mafi sauƙi don bincika sigar Pandas ita ce:

>>> import pandas as pd
>>> pd.__version__
1.0.0rc0

Ingantacciyar taƙaitawa ta atomatik tare da DataFrame.info

Ƙirƙirar da na fi so ita ce sabunta hanyar DataFrame.info. Aikin ya zama abin karantawa sosai, yana sa tsarin binciken bayanai ya fi sauƙi:

>>> df = pd.DataFrame({
...:   'A': [1,2,3], 
...:   'B': ["goodbye", "cruel", "world"], 
...:   'C': [False, True, False]
...:})
>>> df.info()
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 3 entries, 0 to 2
Data columns (total 3 columns):
 #   Column  Non-Null Count  Dtype
---  ------  --------------  -----
 0   A       3 non-null      int64
 1   B       3 non-null      object
 2   C       3 non-null      object
dtypes: int64(1), object(2)
memory usage: 200.0+ bytes

Fitar tebur a cikin tsarin Markdown

Daidaitaccen bidi'a mai daɗi shine ikon fitar da firam ɗin bayanai zuwa teburin Markdown ta amfani da DataFrame.to_markdown.

>>> df.to_markdown()
|    |   A | B       | C     |
|---:|----:|:--------|:------|
|  0 |   1 | goodbye | False |
|  1 |   2 | cruel   | True  |
|  2 |   3 | world   | False |

Wannan ya sa ya fi sauƙi don buga tebur akan shafuka kamar Medium ta amfani da github gists.

Sabbin nau'ikan kirtani da booleans

Sakin Pandas 1.0 shima ya kara sabo na gwaji iri. API ɗin su na iya canzawa har yanzu, don haka yi amfani da shi da taka tsantsan. Amma gabaɗaya, Pandas yana ba da shawarar yin amfani da sabbin nau'ikan a duk inda yake da ma'ana.

A yanzu, simintin gyare-gyare na buƙatar yin shi a sarari:

>>> B = pd.Series(["goodbye", "cruel", "world"], dtype="string")
>>> C = pd.Series([False, True, False], dtype="bool")
>>> df.B = B, df.C = C
>>> df.info()
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 3 entries, 0 to 2
Data columns (total 3 columns):
 #   Column  Non-Null Count  Dtype
---  ------  --------------  -----
 0   A       3 non-null      int64
 1   B       3 non-null      string
 2   C       3 non-null      bool
dtypes: int64(1), object(1), string(1)
memory usage: 200.0+ bytes

Lura yadda shafi Dtype yana nuna sabbin nau'ikan - kirtani и barkono.

Mafi amfani fasalin sabon nau'in kirtani shine ikon zaɓi ginshiƙan jere kawai daga dataframes. Wannan na iya sa tantance bayanan rubutu ya fi sauƙi:

df.select_dtypes("string")

A baya can, ba za a iya zaɓar ginshiƙan jere ba tare da ƙayyadadden sunaye ba.

Kuna iya karanta ƙarin game da sababbin nau'ikan a nan.

Na gode don karantawa! Cikakken jerin canje-canje, kamar yadda aka ambata, ana iya duba su a nan.

source: www.habr.com