Amalabhulali e-Python amangalisayo, kodwa awaziwa kancane

Umhleli wePython uhlale efuna imitapo yolwazi emisha, engathuthukisa umsebenzi kubunjiniyela bedatha namaphrojekthi wezobunhloli bebhizinisi.

Kulesi sihloko sibona imitapo yolwazi ye-python eyaziwa kancane, kodwa ewusizo kakhulu:

1. Ipipi

Nakuba imitapo yolwazi eminingi itholakala ku Python nge-DateTime, ngithola i-Pendulum kulula ukuyisebenzisa kunoma yikuphi ukusebenza kwedethi. I-pendulum yikhabethe lezincwadi engilithanda kakhulu engilisebenzisa nsuku zonke emsebenzini. Inweba imojula yedethi ye-Python eyakhelwe ngaphakathi, yengeza i-API enembile yokuphatha izindawo zesikhathi nokwenza imisebenzi yedethi nesikhathi njengokwengeza izikhawu zesikhathi, ukukhipha amadethi, nokuguqula phakathi kwezindawo zesikhathi. Ihlinzeka nge-API elula nenembile yokufometha izinsuku nezikhathi.

Ukufaka

!pip install pendulum

Ngokwesibonelo

# import library

import pendulum
dt = pendulum.datetime(2023, 1, 31)
print(dt)
 
#local() creates datetime instance with local timezone

local = pendulum.local(2023, 1, 31)
print("Local Time:", local)
print("Local Time Zone:", local.timezone.name)

# Printing UTC time

utc = pendulum.now('UTC')
print("Current UTC time:", utc)
 
# Converting UTC timezone into Europe/Paris time

europe = utc.in_timezone('Europe/Paris')
print("Current time in Paris:", europe)

Okukhiphayo

2. ftfi

Ingabe uke wahlangabezana lapho ulimi lwangaphandle kudatha lungaveli kahle? Lokhu kubizwa ngeMojibake. I-Mojibake igama elisetshenziselwa ukuchaza umbhalo ovalekile noma ocashile ovela ngenxa yezinkinga zombhalo wekhodi noma wokuqopha. Ngokuvamile kwenzeka lapho umbhalo obhalwe ngohlamvu olulodwa lombhalo wekhodi ukhishwa ikhodi ngokungalungile kusetshenziswa umbhalo wekhodi ohlukile. Umtapo wezincwadi we-ftfy python uzokusiza ukuthi ulungise i-Mojibake, ewusizo kakhulu ezimweni zokusebenzisa i-NLP.

Ukufaka

!pip faka i-ftfy

Ngokwesibonelo

print(ftfy.fix_text('Lungisa umusho usebenzisa “ftfyâ€\x9d.')) print(ftfy.fix_text('✔ Azikho izinkinga ngombhalo')) print(ftfy.fix_text('à perturber la réflexion '))

Okukhiphayo

Ngokungeziwe ku-Mojibake, i-ftfy izolungisa amakhodi amabi, iziphetho zemigqa ezimbi, nezingcaphuno ezimbi. ingase iqonde umbhalo okhishwe njengenye yamakhodi alandelayo:

• Isi-Latin-1 (ISO-8859–1)
• Windows-1252 (cp1252 — esetshenziswa emikhiqizweni ye-Microsoft)
• Windows-1251 (cp1251 — inguqulo yesiRashiya ye-cp1252)
• Windows-1250 (cp1250 — inguqulo yaseMpumalanga Yurophu ye-cp1252)
• I-ISO-8859–2 (engafani ncamashi ne-Windows-1250)
• I-MacRoman (esetshenziswa ku-Mac OS 9 nangaphambili)
• cp437 (esetshenziswa ku-MS-DOS nezinye izinguqulo ze-Windows command prompt)

3.Sketch

I-Sketch iwumsizi wekhodi we-AI oyingqayizivele oklanyelwe ngqo abasebenzisi abasebenza nomtapo wezincwadi we-pandas e-Python. Isebenzisa ama-algorithms okufunda komshini ukuze iqonde umongo wedatha yomsebenzisi futhi inikeze iziphakamiso zekhodi ezifanele ukwenza imisebenzi yokukhohlisa nokuhlaziya ibe lula futhi isebenze kahle kakhudlwana. I-Sketch ayidingi abasebenzisi ukuthi bafake noma yimaphi ama-plug-in engeziwe ku-IDE yabo, okwenza isheshe futhi ibe lula ukuyisebenzisa. Lokhu kunganciphisa kakhulu isikhathi nomzamo odingekayo wemisebenzi ehlobene nedatha futhi kusize abasebenzisi babhale ikhodi engcono, esebenza kahle kakhulu.

Ukufaka

!i-pip install sketch

Ngokwesibonelo

Sidinga ukungeza isandiso .sketch kuhlaka lwedatha ye-pandas ukuze sisebenzise le labhulali.

.sketch.buza

cela isici se-Sketch esivumela abasebenzisi ukuthi babuze imibuzo mayelana nedatha yabo ngefomethi yolimi lwemvelo. Inikeza impendulo esekelwe embhalweni embuzweni wabasebenzisi.

# Imitapo yolwazi ingenisa ama-panda edwetshiwe njenge-pd # Ukufunda idatha (usebenzisa idatha ye-twitter njengesibonelo) df = pd.read_csv("tweets.csv") print(df)

# Ukubuza ukuthi yimaphi amakholomu awuhlobo lwesigaba df.sketch.ask("Imaphi amakholomu awuhlobo lwesigaba?")

Okukhiphayo

# Ukuze uthole umumo we-dataframe df.sketch.ask("Yini ukuma kohlaka lwedatha")

.sketch.ukwenza kanjani

Kanjani isici esihlinzeka ngebhulokhi yekhodi engasetshenziswa njengendawo yokuqala noma yokugcina yemisebenzi ehlukahlukene ehlobene nedatha. Singacela amazwibela ekhodi ukwenza idatha yabo ibe yejwayelekile, sidale izici ezintsha, silandelele idatha, futhi sakhe amamodeli. Lokhu kuzokonga isikhathi futhi kwenze kube lula ukukopisha nokunamathisela ikhodi; awudingi ukubhala ikhodi mathupha kusukela ekuqaleni.

# Ukucela ukuhlinzeka ngekhodi eshunquliwe ukuze ubone ngeso lengqondo imizwa df.sketch.howto("Bona ngeso lengqondo imizwa")

Okukhiphayo

.sketch.sebenzisa

Umsebenzi .apply kusiza ukukhiqiza izici ezintsha, ukuhlaziya izinkambu, nokwenza okunye ukukhohlisa kwedatha. Ukuze sisebenzise lesi sici, sidinga ukuba ne-akhawunti ye-OpenAI futhi sisebenzise ukhiye we-API ukwenza imisebenzi. Angikazami lesi sici.

Ngikujabulele ukusebenzisa lo mtapo wolwazi, ikakhulukazi woza iyasebenza, futhi ngiyithola iwusizo.

4. ikhasikhodi

I-“pgeocode” iwumtapo wolwazi omuhle kakhulu engisanda kuwela kuwo obusebenziseke ngendlela emangalisayo kumaphrojekthi ami wokuhlaziya indawo. Isibonelo, ikuvumela ukuthi uthole ibanga phakathi kwamakhodi amabili eposi futhi inikeza ulwazi lwendawo ngokuthatha izwe kanye nekhodi yeposi njengokufaka.

Ukufaka

!pip faka i-pgeocode

Ngokwesibonelo

Thola imininingwane yendawo yamakhodi eposi athile

# Ihlola izwe elithi "India" nomi = pgeocode.Nominatim('In') # Ukuthola ulwazi lwe-geo ngokudlulisela amakhodi eposi nomi.query_postal_code(["620018", "620017", "620012"])

Okukhiphayo

I-“pgeocode” ibala ibanga eliphakathi kwamakhodi eposi amabili ngokuthatha izwe namakhodi eposi njengokufaka. Umphumela uvezwa ngamakhilomitha.

# Ukuthola ibanga phakathi kwamakhodi eposi amabili ibanga = pgeocode.GeoDistance('In') distance.query_postal_code("620018", "620012")

Okukhiphayo

5. rembg

i-rembg ingenye ilabhulali ewusizo esusa kalula ingemuva ezithombeni.

Ukufaka

!pip faka i-rembg

Ngokwesibonelo

# Ingenisa imitapo yolwazi

kusuka ku-rembg khipha ukungenisa kwe-cv2 # indlela yesithombe sokufaka (ifayela lami: image.jpeg) input_path = 'image.jpeg' # indlela yokugcina isithombe esiphumayo nokugcina njengokuphumayo.jpeg output_path = 'output.jpeg' # Ukufunda okokufaka okokufaka kwesithombe = cv2.imread(input_path) # Ukukhipha okuphumayo kwangemuva = khipha(okufakiwe) # Ukugcina ifayela cv2.imwrite(output_path, okukhiphayo)

Okukhiphayo

Kungenzeka ukuthi usuyazi kakade eminye yale mitapo yolwazi, kodwa kimina, i-Sketch, i-Pendulum, i-pgeocode, ne-ftfy ibalulekile emsebenzini wami wobunjiniyela bedatha. Ngithembele kubo kakhulu kumaphrojekthi ami.

6. Yenza ubuntu

I-Humanize” ihlinzeka ngefomethi yeyunithi yezinhlamvu elula, efundeka kalula yezinombolo, amadethi, nezikhathi. Umgomo welabhulali ukuthatha idatha futhi uyenze isebenziseke kalula, isibonelo ngokuguqula inani lamasekhondi libe iyunithi yezinhlamvu efundeka kakhudlwana njengokuthi "2 emizuzwini edlule". Ilabhulali ingafometha idatha ngezindlela ezahlukahlukene, okuhlanganisa izinombolo zokufometha ngokhefana, ukuguqula izitembu zesikhathi zibe izikhathi ezihlotshaniswayo, nokuningi.

Ngivame ukusebenzisa izinombolo kanye nezitembu zesikhathi kumaphrojekthi wami wobunjiniyela bedatha.

Ukufaka

!pip ukufaka humanize

Isibonelo (Izinombolo)

# Ingenisa umtapo wolwazi ngenisa ubuntu ngesikhathi sosuku lokungenisa njenge-dt # Izinombolo zokufometha ngokhefana a = humanize.intcomma(951009) # ukuguqula izinombolo zibe amagama b = humanize.intword(10046328394) #printing print(a) print(b)

Okukhiphayo

Isibonelo (usuku nesikhathi)

import humanize importation datetime as dt a = humanize.naturaldate(dt.date(2012, 6, 5)) b = humanize.naturalday(dt.date(2012, 6, 5)) phrinta(a) phrinta(b)

Ercole Palmeri