Ukuvula iZeroOps ngeeAgents ze-AI kunye nobuNjineli boMxholo: Indlela eSekwe kwiPateni kwi-AI DevOps Automation usebenzisa iStackGen kunye neLangChain

Ukuvula iZeroOps ngeeAgents ze-AI kunye nobuNjineli boMxholo: Indlela eSekwe kwiPattern kwi-AI DevOps Automation usebenzisa iStackGen kunye neLangChain

Intshayelelo

Kumhlaba ovela ngokukhawuleza wophuhliso lwesoftware, ingcinga yeZeroOps iye yavela njengendlela yoguqulo kwiDevOps automation. I-ZeroOps ijolise ekupheliseni imfuno yongenelelo lwezandla ekuhanjisweni nasekulawuleni iinkqubo zesoftware, okwenza abaphuhlisi bagxile kwimisebenzi ekumgangatho ophezulu efana nokutsha kunye nobuchule. Kweli nqaku, siza kuphonononga kwihlabathi leZeroOps kwaye siphonononge ukuba iiarhente ze-AI kunye nobunjineli bomxholo zinokuxhaswa njani ukufezekisa le njongo. Siza kuxubusha kwakhona ukuphunyezwa kwendlela esekelwe kwipatheni kwi-AI DevOps automation usebenzisa i-StackGen kunye neLangChain.

Yintoni iZeroOps?

I-ZeroOps yi-paradigm ye-DevOps efuna ukuzenzekelayo yonke i-software yophuhliso lwe-software, ukusuka kuphuhliso ukuya ekuhanjisweni nasekulawuleni. Eyona njongo iphambili ye-ZeroOps kukuphelisa imfuno yongenelelo ngoncedo lwesandla ekuhanjisweni nasekulawuleni iinkqubo zesoftware, ngaloo ndlela kuncitshiswa ukuba nokwenzeka kwempazamo yomntu kunye nokwandisa isantya kunye nokusebenza kakuhle kokuhanjiswa kwesoftware.

Iiarhente ze-AI kunye nobuNjineli boMxholo

Iiarhente ze-AI kunye nobunjineli bomxholo zidlala indima ebalulekileyo ekuphunyezweni kweZeroOps. Iiarhente ze-AI ziinkqubo zesoftware ezinokubona indawo yazo kwaye zithathe amanyathelo okuphumeza iinjongo ezithile. Ubunjineli bomxholo, kwelinye icala, lubandakanya ukuyila kunye nokuphumeza iinkqubo ezinokuqonda kunye nokuphendula kumxholo ezisebenza kuwo.

Kumxholo weZeroOps, ii-arhente ze-AI zingasetyenziselwa ukwenza imisebenzi eyahlukeneyo efana nokuthunyelwa, uvavanyo kunye nokubeka iliso. Ubunjineli bomxholo bunokusetyenziswa ukuyila iinkqubo ezinokuqonda umxholo ezisebenza kuwo, njengokusingqongileyo, umsebenzisi, kunye nomsebenzi okhoyo.

I-StackGen: Indlela esekwe kwiPateni kwi-AI DevOps Automation

I-StackGen sisikhokelo somthombo ovulekileyo esenza abaphuhlisi basebenzise indlela esekelwe kwipatheni kwi-AI DevOps automation. I-StackGen ibonelela ngeseti yeepatheni ezichazwe kwangaphambili kunye neetemplates ezinokuthi zisetyenziswe ukwenza ngokuzenzekelayo imisebenzi eyahlukeneyo efana nokusasazwa, uvavanyo kunye nokubeka iliso.

I-StackGen yakhiwe phezu kweLangChain, isakhelo esenza abaphuhlisi bakwazi ukwakha iinkqubo ze-AI zokuxoxa. I-LangChain inikeza isethi ye-API kunye nezixhobo ezingasetyenziselwa ukwakha iinkqubo ze-AI zengxoxo ezinokuqonda nokuphendula kwiingeniso zolwimi lwendalo.

LangChain: Ulwakhiwo lweeNkqubo zeNgxoxo ze-AI

I-LangChain yinkqubo-sikhokelo eyenza abaphuhlisi bakwazi ukwakha iinkqubo ze-AI zencoko. I-LangChain inikeza isethi ye-API kunye nezixhobo ezingasetyenziselwa ukwakha iinkqubo ze-AI zengxoxo ezinokuqonda nokuphendula kwiingeniso zolwimi lwendalo.

ILangChain yakhiwe phezu kwenani letekhnoloji yomthombo ovulekileyo, kubandakanya iTensorFlow kunye nePyTorch. I-LangChain inikeza isethi ye-API kunye nezixhobo ezingasetyenziselwa ukwakha iinkqubo ze-AI zengxoxo ezinokuqonda nokuphendula kwiingeniso zolwimi lwendalo.

Ukuzalisekisa iZeroOps ngeStackGen kunye neLangChain

Ukuphumeza iZeroOps ngeStackGen kunye neLangChain, kufuneka silandele inani lamanyathelo:

Chaza i-Agent ye-AI: Isinyathelo sokuqala ekuphunyezweni kweZeroOps kukuchaza i-arhente ye-AI eya kusetyenziselwa ukuzenzekelayo imisebenzi eyahlukeneyo. I-arhente ye-AI inokuchazwa kusetyenziswa inani leepateni ezichazwe kwangaphambili kunye neetemplates ezibonelelwe ngu-StackGen.
**Yila i-Context Engine **: Isinyathelo esilandelayo kukuyila injini yomxholo eya kusetyenziswa ukuqonda nokuphendula kwimeko apho i-arhente ye-AI isebenza khona. Injini yomxholo inokuqulunqwa kusetyenziswa inani le-API kunye nezixhobo ezibonelelwa nguLangChain.
** Ukuphumeza i-Agent ye-AI ** : Emva kokuba i-arhente ye-AI kunye ne-injini yomxholo ichazwe, isinyathelo esilandelayo kukuphumeza i-arhente ye-AI usebenzisa inani le-API kunye nezixhobo ezibonelelwa nguStackGen kunye neLangChain.
** Uvavanyo kunye nokusabalalisa **: Isinyathelo sokugqibela kukuvavanya nokuthumela i-agent ye-AI kwindawo yokuvelisa.

Ikhowudi yoMzekelo: Ukuphumeza i-Agent ye-AI elula usebenzisa i-StackGen kunye neLangChain

Nantsi ibhloko yekhowudi yomzekelo ebonisa indlela yokuphumeza i-arhente ye-AI elula usebenzisa i-StackGen kunye neLangChain:

ngenisa i-langchain
ukusuka langchain.chains ukungenisa LLMChain
ukusuka langchain.llms ukungenisa LLM


i-arhente = LLMChain(
    llm=LLM(
        imodeli_igama = "davinci:3-batch",
        iqondo lobushushu=0.5,
        max_tokens=100,
    ),
    input_format="text",
    output_format="text",
)

# Chaza i-injini yomxholo
umxholo_injini = langchain.ContextEngine(
    llm=LLM(
        imodeli_igama = "davinci:3-batch",
        iqondo lobushushu=0.5,
        max_tokens=100,
    ),
    input_format="text",
    output_format="text",
)

# Sebenzisa i-arhente ye-AI
ai_agent.zalisekisa(
    umxholo_injini=injini_yomxholo,
    input_text="Molo, unjani?",
)

# Vavanya kwaye uthumele iarhente ye-AI
ai_agent.test_and_deploy(
    input_text="Molo, unjani?",
)

Le khowudi yekhowudi ibonisa indlela yokuphumeza i-arhente ye-AI elula usebenzisa i-StackGen kunye neLangChain. I-arhente ye-AI ichazwa ngokusebenzisa inani leepateni ezichazwe kwangaphambili kunye neetemplates ezinikezwe ngu-StackGen. Injini yomxholo yenzelwe ukusebenzisa inani le-API kunye nezixhobo ezibonelelwa nguLangChain. I-arhente ye-AI iphunyezwa ngokusebenzisa inani le-API kunye nezixhobo ezibonelelwa nguStackGen kunye neLangChain. Ekugqibeleni, i-arhente ye-AI ivavanywa kwaye isetyenziswe kwindawo yokuvelisa.

Ephezulu 5-10 Eyeyona Isetyenziswayo Imiyalelo/iziqwengana

Nantsi eyona miyalelo/iziqwengana eziphezulu ezi-5-10 zeStackGen kunye neLangChain:

langchain.LLMChain(): Yenza into entsha yetsheyina leLLM.
langchain.ContextEngine(): Yenza into entsha yenjini yomxholo.
LLM(igama_lemodeli="davinci:3-batch", ubushushu=0.5, max_tokens=100): Yenza into entsha yeLLM enegama lomfuziselo elikhankanyiweyo, ubushushu, kunye nobuninzi beempawu.
ai_agent.implement(context_engine=context_engine, input_text="Molo, unjani?"): Iphumeza i-arhente ye-AI isebenzisa i-injini yomxholo ekhankanyiweyo kunye nombhalo ofakiweyo.
ai_agent.test_and_deploy(input_text="Molo, unjani?"): Iimvavanyo kunye nokusebenzisa i-arhente ye-AI isebenzisa umbhalo ofakiweyo okhankanyiweyo.
langchain.llms.LLM(): Yenza into entsha yeLLM.
langchain.chains.LLMChain(): Yenza into entsha yekhonkco leLLM.
langchain.ContextEngine(llm=LLM(model_name="davinci:3-batch", temperature=0.5, max_tokens=100)): Yenza into entsha yenjini yomxholo usebenzisa into ekhankanyiweyo yeLLM.
ai_agent.context_engine = langchain.ContextEngine(llm=LLM(model_name="davinci:3-batch", temperature=0.5, max_tokens=100)): Iseta i-injini yomxholo we-arhente ye-AI kwinto ekhankanyiweyo yenjini yomxholo.
ai_agent.input_text = "Molo, unjani?": Misela igalelo lombhalo we-arhente ye-AI kumbhalo wegalelo okhankanyiweyo.

Umbono

Nantsi imbonakalo yoyilo lweZeroOps usebenzisa iChart.js:

{
  "type": "bar",
  "idatha": {
    "labels": ["I-Agent ye-AI", "Injini Yomxholo", "StackGen", "LangChain"],
    "iiseti zedatha": [
      {
        "label": "Uyilo",
        "idatha": [10, 8, 6, 4],
        "backgroundColor": ["#FF6384", "#36A2EB", "#FFCE56", "#4BC0C0"]
      }
    ]
  },
  "iinketho": {
    "isihloko": {
      "bonisa": yinyani,
      "umbhalo": "ZeroOps Architecture"
    },
    "legend": {
      "bonisa": yinyani
    },
    "izikali": {
      "yAxes": [
        {
          "amanqaku": {
            "beginAtZero": yinyani
          }
        }
      ]
    }
  }
}

Lo mboniso ubonisa ukwakhiwa kwenkqubo yeZeroOps, kuquka i-arhente ye-AI, injini yomxholo, i-StackGen, kunye neLangChain.

Isiphelo

I-ZeroOps yindlela yokuguqula i-automation ye-DevOps efuna ukuphelisa imfuno yokungenelela ngesandla ekuhanjisweni nasekulawuleni iinkqubo zesoftware. Iiarhente ze-AI kunye nobunjineli bomxholo zidlala indima ebalulekileyo ekuphunyezweni kweZeroOps, kwaye iStackGen kunye neLangChain zibonelela ngendlela esekelwe kwipatheni kwi-AI DevOps automation. Ngokuphumeza i-arhente ye-AI elula usebenzisa i-StackGen kunye ne-LangChain, sinokwenza ngokuzenzekelayo imisebenzi eyahlukeneyo efana nokuthunyelwa, ukuvavanywa, kunye nokubeka iliso. I-5-10 ephezulu kakhulu imiyalelo / iziqwenga ezisetyenzisiweyo ze-StackGen kunye neLangChain nazo zinikezelwe, kunye nokubonwa kwe-architecture ye-ZeroOps usebenzisa i-Chart.js.

Ukuvula iZeroOps ngeeAgents ze-AI kunye nobuNjineli boMxholo: Indlela eSekwe kwiPateni kwi-AI DevOps Automation usebenzisa iStackGen kunye neLangChain

Have Questions?