Gabatarwa

A cikin shimfidar wuri mai sauri na dijital na yau, buƙatar ingantacciyar sarrafa kayan more rayuwa ba ta taɓa kasancewa mai matsi ba. Yayin da buƙatun sabis na tushen girgije ke ci gaba da haɓaka, ƙungiyoyi suna juyawa zuwa hanyoyin samar da ababen more rayuwa masu cin gashin kansu don daidaita ayyukansu na DevOps. A cikin wannan labarin, za mu shiga cikin duniyar ƙungiyar kayan aikin AI mai ƙarfi, bincika iyawar DSPy da LangChain, kayan aikin yanke-yanke guda biyu waɗanda ke juyi yadda muke sarrafawa da sarrafa kayan aikinmu.

Kalubalen DevOps Automation

DevOps aiki da kai aiki ne mai rikitarwa wanda ke buƙatar zurfin fahimtar fasahohi da yawa, daga ɗaukar kaya da ƙungiyar kade-kade zuwa saka idanu da shiga. Yayin da adadin aikace-aikace da ayyuka ke ƙaruwa, haka kuma ƙaƙƙarfan kayan aikin ke ƙaruwa, yana ƙara wahalar sarrafawa da kulawa. Hanyoyi na al'ada zuwa sarrafa kansa na DevOps galibi suna dogara ne akan rubutun hannu da daidaitawa, wanda zai haifar da kurakurai, rashin daidaituwa, da raunin tsaro.

Haɓaka Ƙungiyoyin Ƙarfafa Kayan Aiki na AI

Ƙaddamar da kayan aikin AI mai ƙarfi shine mai canza wasa a cikin duniyar sarrafa kansa ta DevOps. Ta hanyar ba da damar koyan na'ura da sarrafa harshe na halitta, wakilan AI na iya sarrafa ayyuka masu rikitarwa, tsinkaya da hana al'amura, da haɓaka amfani da albarkatu. DSPy da LangChain sune manyan kayan aiki guda biyu a cikin wannan sararin samaniya, suna ba da nau'ikan fasali da damar da za su iya taimaka wa ƙungiyoyi su hanzarta ayyukan DevOps.

DSPy: Tsarin Tsarin Kaya na AI na Python

DSPy tsari ne na AI na tushen Python wanda ke ba masu haɓaka damar ginawa da tura hanyoyin samar da kayan aikin AI mai ƙarfi. Tare da DSPy, zaku iya ƙirƙirar wakilan AI na al'ada waɗanda za su iya yin hulɗa tare da kayan aikin ku, sarrafa ayyuka, da yanke shawara dangane da bayanan ainihin-lokaci. Tsarin gine-gine na zamani na DSPy da babban ɗakin karatu na abubuwan da aka riga aka gina su sun sa ya zama kyakkyawan zaɓi don sarrafa manyan abubuwan more rayuwa.

Key Features na DSPy

** Gine-gine na Modular ***: Tsarin gine-gine na DSPy yana ba ku damar ginawa da tura wakilan AI na al'ada waɗanda za su iya yin hulɗa tare da kayan aikin ku. ** Babban Laburare ***: Babban ɗakin karatu na DSPy na abubuwan da aka riga aka gina shi yana sauƙaƙa sarrafa ayyuka masu rikitarwa da haɗawa tare da shahararrun kayan aikin DevOps.

** Gudanar da Bayanan Lokaci na Gaskiya ***: Ƙarfin sarrafa bayanai na lokaci-lokaci na DSPy yana ba wa wakilan AI damar yanke shawara dangane da bayanan-zuwa-minti.

Misalin Lambobin DSPy Mai Sauƙi

Ga misali mai sauƙi na toshe lambar DSPy wanda ke sarrafa tura sabon akwati: `` Python shigo da dspy

ayyana wakilin AI

wakili = dspy.Agent()

Ƙayyade aikin tura kwantena

aiki = dspy.Task( suna = "Tsarin Kwantena", description="Ar da sabon akwati zuwa yanayin samarwa", abubuwan shiga = ["container_image", "container_port"], abubuwan da aka fitar = ["container_id"] )

Ƙayyade rubutun jigilar kwantena

rubutun = dspy.Script( suna = "Ajiye Rubutun Kwantena", description="Ar da sabon akwati zuwa yanayin samarwa", code = "" #!/bin/bash docker run -d -p {container_port}: {container_port} {container_image} """ )

Haɗa rubutun zuwa aikin

task.attach_script (rubutun)

Ƙayyade aikin wakilin AI

aikin aiki = dspy. Gudun aiki ( suna = "Tsarin Aikin Aiki na Kwantena", description="Kwantar da tura sabon akwati ta atomatik", ayyuka = [aiki] )

Ƙayyade daidaitawar wakilin AI

config = dspy.Config( suna = "Kwantar da Kayan Aiki", description="Shigar da wakilin AI don jigilar akwati", abubuwan shiga = ["container_image", "container_port"] )

Ƙirƙiri wakilin AI

agent.create (gudun aiki, daidaitawa)

Wannan toshe lambar yana bayyana ma'anar wakili mai sauƙi na AI wanda ke sarrafa jigilar sabon akwati zuwa yanayin samarwa. Wakilin yana amfani da rubutun ƙaddamar da kwantena da aka riga aka gina don ƙaddamar da kwantena, kuma ana daidaita aikin aiki don karɓar sigogin shigarwa don hoton akwati da tashar jiragen ruwa.

## LangChain: Tsarin Tsatsa na tushen AI Orchestration

LangChain tsari ne na tushen tsatsa na AI wanda ke ba masu haɓaka damar ginawa da tura hanyoyin samar da kayan aikin AI mai ƙarfi. Tare da LangChain, zaku iya ƙirƙirar wakilan AI na al'ada waɗanda zasu iya hulɗa tare da kayan aikin ku, sarrafa ayyuka, da yanke shawara dangane da bayanan lokaci-lokaci. Tsarin gine-ginen na LangChain da ɗimbin ɗakin karatu na abubuwan da aka riga aka gina su sun sa ya zama kyakkyawan zaɓi don sarrafa manyan abubuwan more rayuwa.

### Mahimman Fasalolin LangChain

** Gine-gine na Modular ***: Tsarin gine-gine na LangChain yana ba ku damar ginawa da tura wakilan AI na al'ada waɗanda za su iya yin hulɗa tare da kayan aikin ku.
** Babban Laburare ***: Babban ɗakin karatu na LangChain na abubuwan da aka riga aka gina shi yana sauƙaƙa sarrafa ayyuka masu rikitarwa da haɗawa tare da shahararrun kayan aikin DevOps.
* ** Gudanar da Bayanan Lokaci na Gaskiya ***: Ƙarfin sarrafa bayanai na ainihin lokacin LangChain yana ba wa wakilan AI damar yanke shawara dangane da bayanan-zuwa-minti.

### Misalin LangChain Mai Sauƙi

Ga misali mai sauƙi na toshe lambar LangChain wanda ke sarrafa tura sabon akwati:
``tsatsa
yi amfani da langchain :: {Agent, Aiki, Rubutu, Gudun Aiki, Config};

// Ƙayyade wakilin AI
bari wakili = Wakili :: sabo();

// Ƙayyade aikin tura kwantena
bari task = Aiki :: sabo(
    "Tsarin Kwantena",
    "Aike da sabon akwati zuwa yanayin samarwa",
    vec!["container_image", "container_port"],
    vec!["container_id"]
);

// Ƙayyade rubutun jigilar kaya
bari rubutun = Rubutun :: sabo(
    "Aika Rubutun Kwantena",
    "Aike da sabon akwati zuwa yanayin samarwa",
    r#"
    #!/bin/bash
    docker run -d -p {container_port}: {container_port} {container_image}
    "#,
    vec!["container_image", "container_port"]
);

// Haɗa rubutun zuwa aikin
task.attach_script (rubutun);

// Ƙayyade aikin wakilin AI
bari aikin aiki = Gudun Aiki :: sabo(
    "Tsarin Ayyukan Aiki Na Kwantena",
    "Yi atomatik tura sabon akwati",
    vc![aiki]
);

// Ƙayyade daidaitawar wakilin AI
let config = Config :: sabo(
    "Tsarin Bayar da Kwantena",
    "Shigar da wakilin AI don jigilar akwati",
    vec!["container_image", "container_port"]
);

// Ƙirƙiri wakilin AI
agent.create (gudun aiki, saiti);

Wannan toshe lambar yana bayyana ma'anar wakili mai sauƙi na AI wanda ke sarrafa jigilar sabon akwati zuwa yanayin samarwa. Wakilin yana amfani da rubutun ƙaddamar da kwantena da aka riga aka gina don ƙaddamar da kwantena, kuma ana daidaita aikin aiki don karɓar sigogin shigarwa don hoton akwati da tashar jiragen ruwa.

Manyan Dokoki 5-10 da Akafi Amfani da su/Snippets

Anan akwai manyan 5-10 da aka fi amfani da umarni/snippets don DSPy da LangChain:

DSPy

`dspy.Agent () .ƙirƙira (aiki, daidaitawa): Yana ƙirƙira sabon wakili na AI tare da ƙayyadaddun aikin aiki da daidaitawa.
dspy.Task() haše_script(rubutun): Haɗa rubutun zuwa ɗawainiya a cikin aikin wakilin AI.
`dspy.Script() sabon(suna, kwatance, lamba): Yana ƙirƙira sabon rubutun tare da takamaiman suna, kwatance, da lamba.
`dspy.Workflow() sabon (suna, bayanin, ayyuka)': Yana ƙirƙira sabon tafiyar aiki tare da ƙayyadadden suna, bayanin, da ɗawainiya.
dspy.Config() sabon(suna, bayanin, bayanai): Yana ƙirƙira sabon tsari tare da ƙayyadadden suna, bayanin, da sigogin shigarwa.

LangChain

`langchain :: Wakili :: sabo()': Yana ƙirƙira sabon wakilin AI.
`langchain :: Aiki :: sabon(suna, bayanin, bayanai, abubuwan da aka fitar)': Yana ƙirƙira sabon ɗawainiya tare da ƙayyadaddun suna, bayanin, bayanai, da fitarwa.
`langchain :: Rubutun :: sabon(suna, bayanin, lamba, bayanai)': Ƙirƙirar sabon rubutun tare da ƙayyadadden suna, bayanin, lamba, da sigogin shigarwa.
`langchain :: Gudun Aiki :: sabon(suna, bayanin, ayyuka)': Ƙirƙirar sabon tsarin aiki tare da ƙayyadadden suna, bayanin, da ayyuka.
`langchain :: Config :: sabon(suna, bayanin, bayanai)': Ƙirƙirar sabon tsari tare da ƙayyadadden suna, bayanin, da sigogin shigarwa.

Kallon gani

Anan ga hangen nesa na aikin wakilin AI ta amfani da Chart.js: Json { "type": "bar", "data": { "labels": ["Tsarin Kwantena", "Task 1", "Task 2", "Task 3"], "datasets": [ { "label": "Tsarin aiki", "bayanai": [10, 5, 3, 2], "backgroundColor": "rgba(255, 99, 132, 0.2)", "BorderColor": "rgba(255, 99, 132, 1)", "BorderWidth": 1 } ] }, "zaɓi": { "take": { "nuna": gaskiya, "rubutu": "AI Agent Workflow" }, "ma'auni": { "yAxes": [ { "nuna": gaskiya, "Label": { "nuna": gaskiya, "labelString": "Ayyuka" } } ] } } }

Wannan hangen nesa yana nuna tafiyar aikin wakilin AI, tare da kowane ɗawainiya da mashaya ke wakilta. X-axis yana wakiltar ayyuka, kuma y-axis yana wakiltar adadin ayyuka.

## Kammalawa

A cikin wannan labarin, mun bincika duniyar ƙungiyar kayan aikin AI mai ƙarfi, mai da hankali kan DSPy da LangChain, manyan kayan aiki guda biyu a cikin wannan sarari. Mun ga yadda za a iya amfani da waɗannan kayan aikin don sarrafa hadaddun ayyuka, tsinkaya da hana al'amura, da haɓaka amfani da albarkatu. Tare da tsarin gine-ginen su na zamani da manyan ɗakunan karatu na abubuwan da aka riga aka gina, DSPy da LangChain zaɓi ne masu kyau don sarrafa manyan abubuwan more rayuwa. Ko kai ƙwararren ƙwararren DevOps ne ko kuma farawa, waɗannan kayan aikin tabbas za su taimaka maka haɓaka ayyukan DevOps da ɗaukar ayyukan sarrafa kayan aikin ku zuwa mataki na gaba.

Ƙirƙirar Kayan Aiki Mai Zaman Kanta Tare da Ingantattun Ƙwararrun Wakilin AI: Jagorar Fasaha don Haɓaka DevOps tare da DSPy da LangChain