Ƙirƙiri bututun shigar bayanai | B4N1

Ƙirƙiri bututun shigar bayanai

3 months ago · b4n1 Read time: 6 min

** Gudanar da DevOps na AI-Powered tare da DSPy da LangChain: Tsarin Gine-gine mai ƙima don Binciken Laifi Mai sarrafa kansa ***

A cikin yanayin yanayin dijital mai sauri na yau, ƙungiyoyin DevOps suna fuskantar ƙalubale mai rikitarwa: ganowa da warware abubuwan da suka faru da kyau da inganci. Yayin da girma da saurin fitowar software ke ƙaruwa, yuwuwar kurakurai da keta tsaro suna ƙaruwa. Don ci gaba da gaba, dole ne ƙungiyoyi su ɗauki sabbin dabaru waɗanda ke yin amfani da ikon Artificial Intelligence (AI) don sarrafa sarrafa bincike da ƙuduri.

A cikin wannan labarin, za mu shiga cikin cikakkun bayanan aiwatar da fasaha na amfani da DSPy da LangChain don ƙirƙirar gine-gine mai ƙima don binciken abin da ya faru ta atomatik. Ta hanyar haɗa ƙarfin waɗannan fasahohin yanke-yanke guda biyu, za mu nuna yadda ake gina ƙaƙƙarfan tsari mai inganci wanda ke haɓaka inganci da inganci na DevOps.

** Fahimtar DSPy da LangChain ***

Kafin mu nutse cikin aiwatar da fasaha, bari mu ɗan gabatar da DSPy da LangChain.

DSPy

DSPy ɗakin karatu ne na Python wanda ke ba da babbar hanyar sadarwa don ginawa da tura aikace-aikacen AI masu ƙarfi. Yana ba da fasali da yawa, gami da:

  • ** Samfuran Sabis ***: DSPy yana ba ku damar bautar ƙirar AI azaman APIs, yana sauƙaƙa haɗa su cikin aikace-aikacen da ke akwai. ** Gudanar da Samfurin ***: DSPy yana ba da kayan aiki don sarrafa samfuran AI, gami da ƙaddamar da ƙira, saka idanu, da sigar.
  • **Haɗin bayanai ***: DSPy yana ba da damar haɗin kai mara kyau tare da hanyoyin bayanai daban-daban, gami da bayanan bayanai, APIs, da tsarin fayil.

LangChain

LangChain ɗakin karatu ne na Python wanda ke ba da tsari don gina aikace-aikacen AI na tattaunawa. Yana ba da fasali da yawa, gami da:

  • ** Hanyoyin tattaunawa ***: LangChain yana ba da damar ƙirƙirar mu'amalar tattaunawa, kamar su taɗi da mataimakan murya.
  • ** Gudanar da Harshen Halitta (NLP): LangChain yana ba da kayan aiki don ayyukan NLP, gami da rarrabuwar rubutu, nazarin jin daɗi, da sanin mahalli.
  • ** Gudanar da jadawali na ilimi ***: LangChain yana ba da damar ƙirƙira da sarrafa jadawalin ilimin, waɗanda ke da mahimmanci don aikace-aikacen AI na tattaunawa.

Gina Tsarin Gine-gine Mai Ma'ana

Don gina gine-gine mai ƙima don binciken abin da ya faru ta atomatik, za mu haɗu da ƙarfin DSPy da LangChain. Gine-ginenmu zai ƙunshi abubuwa masu zuwa:

  1. ** Data Ingestion ***: Za mu yi amfani da DSPy don shigar da bayanai daga daban-daban kafofin, ciki har da rajistan ayyukan, awo, da APIs.
  2. ** Gudanar da Bayanai ***: Za mu yi amfani da DSPy don sarrafa bayanan da aka ci, amfani da samfurin AI don gano abubuwan da suka faru.
  3. ** Gano Ganewa ***: Za mu yi amfani da LangChain don ƙirƙirar haɗin tattaunawa don gano abin da ya faru, ƙyale masu amfani su yi hulɗa tare da tsarin da kuma samar da mahallin binciken abin da ya faru.
  4. ** Gudanar da Zane na Ilimi ***: Za mu yi amfani da LangChain don ƙirƙira da sarrafa jadawalin ilimi, wanda zai adana bayanai game da abubuwan da suka faru, gami da abubuwan da suka haifar, tasirin su, da ƙuduri.
  5. ** Resolution ***: Za mu yi amfani da DSPy don amfani da samfurin AI don warware abubuwan da suka faru, yin amfani da jadawalin ilimin don sanar da yanke shawara.

Aikin Fasaha

Bari mu nutse cikin cikakkun bayanan aiwatar da fasaha na gine-ginen mu.

Ciwon Data

Za mu yi amfani da DSPy don shigar da bayanai daga tushe daban-daban, gami da rajistan ayyukan, awo, da APIs. Za mu ƙirƙiri bututun shigar da bayanai ta amfani da ajin 'DataPipeline' na DSPy, wanda zai kula da sarrafa bayanai da adanawa.

`` Python shigo da dspy a matsayin dsp

bututu = dsp.DataPipeline()

Ƙara tushen bayanai zuwa bututun

pipeline.add_source ("logs", dsp.LogSource()) pipeline.add_source ("ma'auni", dsp.MetricSource()) pipeline.add_source ("api", dsp.APISource())

Fara bututun

pipeline.fara()


### Gudanar da Data

Za mu yi amfani da DSPy don aiwatar da bayanan da aka shigar, yin amfani da ƙirar AI don gano abubuwan da suka faru. Za mu ƙirƙiri bututun sarrafa bayanai ta amfani da ajin 'DataPipeline' na DSPy, wanda zai kula da sarrafa bayanai da adanawa.

`` Python
shigo da dspy a matsayin dsp

# Ƙirƙiri bututun sarrafa bayanai
bututu = dsp.DataPipeline()

# Ƙara samfuran AI zuwa bututun
pipeline.add_model ("ganewar abin da ya faru", dsp.IncidentDetectionModel())
pipeline.add_model("ganewar_anomaly", dsp.AnomalyDetectionModel())

# Fara bututun
pipeline.fara()

Gano Haƙiƙa

Za mu yi amfani da LangChain don ƙirƙirar haɗin tattaunawa don gano abin da ya faru, ba da damar masu amfani su yi hulɗa tare da tsarin da samar da mahallin binciken abin da ya faru.

`` Python shigo da langchain kamar yadda lc

Ƙirƙiri hanyar tattaunawa

dubawa = lc.ConversationalInterface()

Ƙara chatbot zuwa ke dubawa

interface.add_chatbot("ganewar lamarin", lc.Chatbot())

Fara fara dubawa

interface.start()


### Gudanar da Hotunan Ilimi

Za mu yi amfani da LangChain don ƙirƙira da sarrafa jadawali na ilimi, wanda zai adana bayanai game da abubuwan da suka faru, gami da abubuwan da suka faru, tasirin su, da ƙuduri.

`` Python
shigo da langchain kamar yadda lc

# Ƙirƙiri jadawalin ilimi
jadawali = lc.KnowledgeGraph()

# Ƙara nodes zuwa jadawali
graph.add_node ("wani abu1", lc.Node())
graph.add_node ("wani abu2", lc.Node())

# Ƙara gefuna zuwa jadawali
graph.add_edge ("wasu faru1", "watsawa2", lc.Edge())

Ƙaddamarwa

Za mu yi amfani da DSPy don amfani da samfuran AI don warware abubuwan da suka faru, muna yin amfani da jadawalin ilimin don sanar da yanke shawara.

`` Python shigo da dspy a matsayin dsp

Ƙirƙiri bututun ƙuduri

bututu = dsp.DataPipeline()

Ƙara samfuran AI zuwa bututun

pipeline.add_model("hasuwar_resolution", dsp.IncidentResolutionModel())

Fara bututun

pipeline.fara()


** 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

* `dsp.DataPipeline()`: Yana ƙirƙira bututun bayanai don ci da sarrafa bayanai.
* `dsp.add_source()': Yana ƙara tushen bayanai zuwa bututun.
* `dsp.add_model()`: Yana ƙara ƙirar AI zuwa bututun.
* `dsp.start()`: Fara bututun.
* `dsp.stop()`: Yana tsayar da bututun.

### LangChain

* `lc.ConversationalInterface()`: Yana ƙirƙira hanyar tattaunawa don gano abin da ya faru.
* `lc.add_chatbot()`: Yana ƙara ma'amalar taɗi zuwa mahalli.
* `lc.start()`: Yana fara dubawa.
* `lc.KnowledgeGraph()`: Yana ƙirƙira jadawalin ilimi don sarrafa abin da ya faru.
* `lc.add_node()`: Yana ƙara kumburi zuwa jadawali.
* `lc.add_edge()`: Yana ƙara gefe zuwa jadawali.

**Kallon gani**

Anan ga hangen nesa na gine-ginen mu ta amfani da Chart.js:

Json
{
  "type": "pieChart",
  "data": {
    "labels": ["Data Ingestion", "Tsarin Bayanan", "Ganewar Lamarin", "Gudanar da Zane-zane", "Ƙaddamarwa"],
    "datasets": [
      {
        "label": "Kayan aikin gine-gine",
        "bayanai": [30, 20, 20, 15, 15],
        "BackgroundLauni": [
          "rgba (255, 99, 132, 0.2)",
          "rgba (54, 162, 235, 0.2)",
          "rgba (255, 206, 86, 0.2)",
          "rgba (75, 192, 192, 0.2)",
          "rgba (153, 102, 255, 0.2)"
        ],
        "Launi Border": [
          "rgba (255, 99, 132, 1)",
          "rgba (54, 162, 235, 1)",
          "rgba (255, 206, 86, 1)",
          "rgba (75, 192, 192, 1)",
          "rgba (153, 102, 255, 1)"
        ],
        "BorderWidth": 1
      }
    ]
  },
  "zaɓi": {
    "take": {
      "nuna": gaskiya,
      "rubutu": "Kayan aikin gine-gine"
    },
    "tatsuniya": {
      "nuni": gaskiya
    }
  }
}

Wannan hangen nesa yana nuna sassa biyar na gine-ginen mu, tare da shigar da bayanai da abubuwan sarrafawa suna ɗaukar kaso mafi girma.

Kammalawa

A cikin wannan labarin, mun nuna yadda ake gina gine-gine mai ƙima don binciken abin da ya faru ta atomatik ta amfani da DSPy da LangChain. Ta hanyar haɗa ƙarfin waɗannan fasahohin yanke-yanke guda biyu, mun ƙirƙiri ƙaƙƙarfan tsari mai inganci wanda ke haɓaka inganci da inganci na DevOps. Mun kuma bayar da misalin aiwatarwa mai amfani, gami da snippets code da hangen nesa na gine-gine. Ta hanyar ɗaukar wannan gine-ginen, ƙungiyoyi za su iya inganta lokutan amsa abubuwan da suka faru, rage raguwa, da haɓaka amincin tsarin gaba ɗaya.

Have Questions?

Let's Talk

Contact me on LinkedIn