Gabatarwa

Filin DevOps ya sami sauye-sauye masu mahimmanci a cikin 'yan shekarun nan, tare da haɗin gwiwar Artificial Intelligence (AI) da Koyon Injin (ML) yana ƙara zama sananne. Ɗaya daga cikin mahimman wuraren da AI ke yin tasiri mai mahimmanci shine a fagen sarrafa kansa. Ta hanyar yin amfani da wakilai na AI, ƙungiyoyi za su iya daidaita ayyukan haɓaka software na su, rage kurakuran hannu, da haɓaka haɓaka gabaɗaya.

A cikin wannan labarin, za mu shiga cikin duniyar AI-kore DevOps aiki da kai, mai da hankali kan tsarin LangChain da DSPy. Za mu bincika yadda za a iya amfani da waɗannan kayan aikin don ƙirƙirar tsarin ƙudurin al'amura da sarrafa sarrafa ayyukan haɓaka software. Bugu da ƙari, za mu ba da misali mai amfani na yadda ake haɗa Manyan Samfuran Harshe (LLMs) cikin bututun ku na DevOps.

Menene LangChain da DSPy?

Kafin mu nutse cikin cikakkun bayanai na aiki da kai na DevOps na AI, bari mu ɗan kalli tsarin LangChain da DSPy.

LangChain

LangChain shine tsarin buɗe tushen tushen da ke ba masu haɓaka damar gina aikace-aikacen AI na tattaunawa ta amfani da Manyan Harshe Model (LLMs). Yana ba da API mai sauƙi kuma mai fahimta don hulɗa tare da LLMs, yana barin masu haɓakawa su ƙirƙira musaya na tattaunawa na al'ada don aikace-aikacen su.

LangChain yana goyan bayan kewayon LLMs, gami da shahararrun samfura kamar LLAMA, BERT, da RoBERTa. Hakanan yana ba da kewayon kayan aikin don daidaitawa da daidaita LLMs don dacewa da takamaiman yanayin amfani.

DSPy

DSPy ɗakin karatu ne na Python wanda ke ba da hanya mai sauƙi da inganci don haɗa samfuran AI cikin bututun ku na DevOps. Yana goyan bayan tsarin tsarin AI mai yawa, ciki har da LangChain, kuma yana ba da kayan aiki da yawa don sarrafa ayyukan aiki na AI.

An tsara DSPy don yin aiki ba tare da matsala ba tare da shahararrun kayan aikin DevOps kamar Jenkins, GitLab, da Docker, yana mai da shi kyakkyawan zaɓi ga ƙungiyoyin da ke neman haɗa AI cikin kayan aikin DevOps na yanzu.

Wakilan AI na Orchestrating don DevOps Automation

Yanzu da muke da taƙaitaccen bayani game da LangChain da DSPy, bari mu bincika yadda waɗannan tsarin za a iya amfani da su don tsara wakilan AI don sarrafa kansa na DevOps.

Ƙaddamar Batutuwa Mai Sauƙi

Ɗaya daga cikin mahimman fa'idodin aikin DevOps na AI-kore shine ikon magance al'amura da gaske kafin su zama manyan matsaloli. Ana iya amfani da LangChain da DSPy don ƙirƙirar hanyoyin tattaunawa na al'ada waɗanda ke hulɗa tare da LLMs don gano abubuwan da za su iya yiwuwa da kuma samar da shawarwari don ƙuduri.

Ga misali mai sauƙi na yadda wannan zai iya aiki:

`` Python shigo da langchain daga langchain.chat_models shigo da LLMA

Ƙirƙiri samfurin taɗi na LangChain

chat_model = LLMA()

Ƙayyade aiki don hulɗa tare da ƙirar taɗi

def interact_with_chat_model(da sauri): amsa = chat_model.run (gaggawa) mayar da martani

Ƙayyade aiki don gano abubuwan da za su iya faruwa

def detect_dissues(lambar): m = f"Gano matsaloli masu yuwuwa a cikin lambar mai zuwa: {code}" amsa = interact_with_chat_model (da sauri) mayar da martani

Ƙayyade aiki don samar da shawarwari don ƙuduri

def provide_recommendations(masu magana): m = f"Ba da shawarwari don warware waɗannan batutuwa: {al'amura}" amsa = interact_with_chat_model (da sauri) mayar da martani

Misalin amfani

code = "" def add(x, y): dawo x + y """ al'amurran da suka shafi = gano_al'amura (lambar) shawarwari = bayar da shawarwari (masu ba da shawara) buga (shawarwari)

A cikin wannan misalin, mun ayyana aiki mai sauƙi 'mu'amala da_chat_model' wanda ke hulɗa tare da ƙirar taɗi na LangChain don ba da amsa ga abin da aka bayar. Sannan muna ayyana ƙarin ayyuka guda biyu 'gano_al'amurra' da 'bayar da_shawarwari' waɗanda ke amfani da ƙirar taɗi don gano abubuwan da za su yuwu a cikin wasu lambar samfurin da bayar da shawarwari don ƙuduri.

### Ayyukan Ci gaban Software ta atomatik

Hakanan za'a iya amfani da LangChain da DSPy don sarrafa ayyukan haɓaka software ta hanyar haɗa samfuran AI cikin bututun DevOps ɗin ku.

Ga misalin yadda wannan zai iya aiki:

`` Python
shigo da dspy
daga dspy shigo da Gudun Ayyuka

# Ƙayyade aikin DSPy
aiki = Gudun aiki ()

# Ƙayyade aiki don gudanar da kayan aikin bincike na lamba
def run_code_analysis(ladi):
    # Gudanar da kayan aikin bincike na lamba don gano abubuwan da za su iya faruwa
    matsalolin = []
    #...
    matsalolin dawowa

# Ƙayyade aiki don ba da shawarwari don ƙuduri
def provide_recommendations(masu magana):
    # Yi amfani da samfurin taɗi na LangChain don ba da shawarwari
    m = f"Ba da shawarwari don warware waɗannan batutuwa: {al'amura}"
    amsa = langchain.chat_model.run (gaggawa)
    mayar da martani

# Ƙayyade aiki don tura lambar
def deploy_code(code):
    # Sanya lambar zuwa yanayin samarwa
    #...

# Bayyana tsarin aiki
workflow.add_task(run_code_analysis)
workflow.add_task(bayar da_shawarwari)
workflow.add_task(deploy_code)

# Gudanar da aikin
aikin aiki.run ()

A cikin wannan misali, mun ayyana aiki mai sauƙi na DSPy wanda ya ƙunshi ayyuka uku: gudanar da kayan aikin bincike na lamba, samar da shawarwari don ƙuduri, da ƙaddamar da lambar. Sa'an nan kuma mu ayyana tsarin aiki ta hanyar ƙara kowane ɗawainiya da gudanar da shi.

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

Anan akwai wasu umarni da snippets da aka fi amfani da su don LangChain da DSPy:

LangChain

• langchain.chat_models.LLaMA(): Ƙirƙiri samfurin taɗi na LangChain ta amfani da ƙirar LLAMA.
• `langchain.chat_model.run(gaggauta)': Guda samfurin taɗi tare da faɗakarwa.
• langchain.chat_model.interact_with(gaggauta): Yi hulɗa tare da ƙirar taɗi ta amfani da faɗakarwa.

DSPy

• `dspy.Tsarin aiki()': Ƙirƙiri aikin DSPy.
• 'Gudun aiki.add_task(aiki)': Ƙara ɗawainiya zuwa aikin aiki.
• workflow.run(): Gudanar da aikin.

Kallon gani

Anan ga sauƙin gani na tafiyar aiki ta amfani da Chart.js:

Json { "type": "bar", "data": { "labels": ["Label Analysis", "Shawarwari", "Tsarin aiki"], "datasets": [ { "label": "Tsarin aiki", "bayanai": [10, 20, 30], "BackgroundLauni": [ "rgba (255, 99, 132, 0.2)", "rgba (54, 162, 235, 0.2)", "rgba (255, 206, 86, 0.2)" ], "Launi Border": [ "rgba (255, 99, 132, 1)", "rgba (54, 162, 235, 1)", "rgba (255, 206, 86, 1)" ], "BorderWidth": 1 } ] }, "zaɓi": { "take": { "nuna": gaskiya, "text": "Tsarin aiki" }, "ma'auni": { "yAxes": [ { "zama": { "beginAtZero": gaskiya } } ] } } }

Wannan hangen nesa yana nuna ginshiƙi mai sauƙi wanda ke wakiltar tafiyar aiki, tare da sanduna uku masu wakiltar ƙididdigar lamba, shawarwari, da ayyukan turawa.

## Kammalawa

A cikin wannan labarin, mun bincika duniyar AI-kore DevOps aiki da kai ta amfani da LangChain da DSPy. Mun ga yadda za a iya amfani da waɗannan tsare-tsare don ƙirƙirar tsarin ƙudurin al'amura da sarrafa sarrafa ayyukan haɓaka software. Mun kuma bayar da misalai masu amfani na yadda ake haɗa LLMs a cikin bututun DevOps ɗinku kuma mu hango yadda aikin ke gudana ta amfani da Chart.js.

Ta hanyar amfani da ikon AI da aiki da kai, ƙungiyoyi za su iya daidaita ayyukan haɓaka software na su, rage kurakuran hannu, da haɓaka haɓaka gabaɗaya. Ko kai injiniyan DevOps ne, mai haɓaka software, ko jagoran kasuwanci, LangChain da DSPy kayan aiki ne masu ƙarfi waɗanda zasu iya taimaka muku cimma burin ku.