** Aiwatar da Tsararren Tsararren Gine-gine don Aikace-aikacen Yanar Gizo Mai Karfin AI**

** Gabatarwa ***

Yayin da aikace-aikacen gidan yanar gizo masu ƙarfin AI ke ƙara yaɗuwa, buƙatar ƙaƙƙarfan gine-ginen gine-gine bai taɓa kasancewa mai matsi ba. Tare da haɓakar hare-haren AI, yana da mahimmanci don aiwatar da amintacciyar hanya ta tsohuwa wacce za ta iya jure yuwuwar barazanar. A cikin wannan labarin, za mu bincika yadda ake haɗa amintattun bayanan tushen tushen bayanan Percona tare da amintaccen tsari na Chainguard da haɗawa tare da LangChain don aikace-aikacen yanar gizo masu ƙarfi na AI.

** Barazana ga Aikace-aikacen Yanar Gizo Mai Karfin AI**

Aikace-aikacen gidan yanar gizo masu ƙarfin AI suna da rauni ga barazana iri-iri, gami da:

** Guba bayanan ***: Ana iya sarrafa samfuran AI ta hanyar ciyar da su bayanan ƙeta, wanda ke haifar da tsinkaya ko yanke shawara mara kyau.

** Juyin Samfura ***: Mahara na iya ba da bayanai masu mahimmanci game da ƙirar AI ko bayanan horo. ** Hare-Hare-hare ***: Ana iya yaudarar samfuran AI ta hanyar abubuwan da aka ƙera a hankali waɗanda aka tsara don ɓatar da ƙirar.

Tsaffin Tsarin Gine-gine na Tsare-tsare

Don rage waɗannan barazanar, za mu aiwatar da ingantaccen tsarin gine-ginen da ya haɗu da amintattun bayanan buɗaɗɗen tushen bayanan Percona tare da amintaccen tsari na Chainguard da kuma haɗawa tare da LangChain don aikace-aikacen yanar gizo masu ƙarfin AI.

Tabbatattun Bayanan Bayanai na Buɗe-Source na Percona

Percona yana ba da kewayon amintattun bayanan tushen tushen bayanai, gami da Percona Server don MySQL da Percona Server don PostgreSQL. Waɗannan ma'ajin bayanai suna samar da ingantaccen fasali na tsaro, kamar:

Rufewa: Ana rufaffen bayanai a lokacin hutawa da wucewa, tabbatar da cewa mahimman bayanai sun kasance sirri.
Ikon Samun damar: Kyakkyawan ikon sarrafa damar shiga yana bawa masu gudanarwa damar iyakance damar yin amfani da bayanai masu mahimmanci kuma tabbatar da cewa ma'aikata masu izini kawai zasu iya gyara ko share bayanai.
Auditing: Cikakken ikon tantancewa yana ba da cikakken rikodin duk ayyukan bayanai, yana sauƙaƙa ganowa da amsa abubuwan tsaro.

Hanyar Tsare-tsare ta Tsare-tsare

Chainguard yana ba da amintaccen tsari na tsoho wanda ke tabbatar da cewa an gina aikace-aikacen tare da tsaro tun daga farko. Wannan tsarin ya haɗa da:

**Ƙaramar Gata ***: An ƙirƙira aikace-aikacen don gudanar da mafi ƙarancin gata da ake buƙata don aiwatar da ayyukansu, rage tasirin harin.
Kamfanin da ba za a iya canzawa ba: An ƙera kayan more rayuwa don su zama marasa canzawa, don sauƙaƙe ganowa da amsa abubuwan da suka faru na tsaro.
** Amintaccen Kanfigareshan ***: An saita aikace-aikacen tare da tsaro a zuciya, gami da amintattun abubuwan da ba za a iya ba da kuma ƙaramin tsari.

Haɗin LangChain

LangChain ɗakin karatu ne na Python wanda ke ba da API mai sauƙi da fahimta don gina aikace-aikacen yanar gizo masu ƙarfin AI. Za mu haɗa LangChain tare da amintattun bayanan buɗe tushen bayanai na Percona da amintaccen hanyar Chainguard don ƙirƙirar ƙaƙƙarfan tsarin gine-gine.

** Misalin Koda ***

Ga misali mai sauƙi na lamba wanda ke nuna yadda ake haɗa LangChain tare da amintattun bayanan tushen tushen bayanan Percona da tsarin tsaro na Chainguard: `` Python shigo da langchain daga langchain shigo da LLM

LLM = LLM model_name = "davinci: 2-large", na'ura = "cuda", max_tokens=1024, )

Ƙirƙiri haɗin bayanan Percona

shigo da mysql.connector cnx = mysql.connector.connect( mai amfani = 'username', kalmar sirri = 'password', Mai watsa shiri = '127.0.0.1', database = 'database' )

Ƙirƙiri amintaccen tsari na Chainguard

shigo da sarƙoƙi cfg = chainguard.Config( app_name = "my_app", app_version = "1.0", env_vars={ "DB_HOST": "127.0.0.1", "DB_USER": "sunan mai amfani", "DB_PASSWORD": "Password", }, )

Yi amfani da LangChain don bincika bayanan

tambaya = "Zabi * DAGA tebur na" sakamako = llm.query (tambaya, db=cnx, cfg=cfg)

Buga sakamakon

buga(sakamako)

Wannan misalin lambar yana nuna yadda ake ƙirƙirar misalin LangChain LLM, haɗa zuwa bayanan Percona, da ƙirƙirar amintaccen tsari na Chainguard. Sannan yana amfani da LangChain don bincika bayanan bayanai da buga sakamakon.

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

Anan akwai manyan umarni 5-10 da aka fi amfani da su don wannan fasaha:

1. `langchain.LLM (model_name = "davinci: 2-large", na'urar = "cuda", max_tokens = 1024) ` - Ƙirƙiri misali na LangChain LLM.
2. `mysql.connector.connect(user='username', password='password', host='127.0.0.1', database='database')` - Ƙirƙiri haɗin bayanan Percona.
3. `chainguard.Config(app_name="my_app", app_version="1.0", env_vars={"DB_HOST": "127.0.0.1", "DB_USER": "username", "DB_PASSWORD": "password"})` - Ƙirƙiri amintaccen tsari na Chainguard.
4. `llm.query(tambaya, db=cnx, cfg=cfg)` - Yi amfani da LangChain don neman bayanai.
5. `print(sakamakon)` - Buga sakamakon tambayar.
6. `shigo da langchain` - Shigo da ɗakin karatu na LangChain.
7. `import mysql.connector` - Shigo da bayanan Percona.
8. `shigo da sarƙoƙi` - Shigo da ɗakin karatu na Chainguard.
9. `llm = LLM()` - Ƙirƙiri misalin LangChain LLM ba tare da fayyace samfuri ba.
10. `cnx = mysql.connector.connect()` - Ƙirƙiri haɗin bayanai na Percona ba tare da ƙididdige bayanan bayanai ba.

**Kallon gani**

Anan ga hangen nesa na amintattun gine-gine ta hanyar amfani da Chart.js:
Json
{
  "type": "bar",
  "data": {
    "labels": ["Database Percona", "Chainguard Secure Kanfigareshan", "LangChain LLM"],
    "datasets": [
      {
        "label": "Tsarin Tsaro",
        "bayanai": [10, 8, 12],
        "Bayan launi": ["#FF6384", "#36A2EB", "#FFCE56"]
      }
    ]
  },
  "zaɓi": {
    "take": {
      "nuna": gaskiya,
      "text": "Tsarin Gine-gine na Tsare-tsare"
    },
    "ma'auni": {
      "yAxes": [
        {
          "zama": {
            "beginAtZero": gaskiya
          }
        }
      ]
    }
  }
}

Wannan hangen nesa yana nuna fasalulluka na tsaro na amintattun bayanan buɗaɗɗen tushen bayanai na Percona, amintaccen tsari na Chainguard, da LangChain's LLM.

Kammalawa

Aiwatar da amintaccen tsarin gine-gine don aikace-aikacen gidan yanar gizo masu ƙarfin AI yana da mahimmanci a cikin yanayin barazanar yau. Ta hanyar haɗa amintattun bayanan buɗe tushen tushen bayanai na Percona tare da amintaccen hanyar Chainguard da haɗawa tare da LangChain, za mu iya ƙirƙirar ƙaƙƙarfan tsarin gine-gine wanda zai iya jure yuwuwar hare-haren AI. Wannan labarin ya ba da jagora mai amfani kan yadda ake aiwatar da wannan gine-gine, gami da misalan lambobi da abubuwan gani.

Ƙirƙiri misalin LangChain LLM