Hordhac
Kakanaanta joogtada ah ee nidaamyada software-ka casriga ah ayaa horseeday korodhka soo noqnoqda iyo darnaanta shilalka, samaynta baaritaanka shilalka firfircoon iyo xallinta arrin muhiim ah oo DevOps ah. Sirdoonka macmalka ah (AI) iyo barashada mashiinka (ML) waxay u soo baxeen sidii qalab awood leh si kor loogu qaado dhuumaha DevOps, awood u siinaya AI-ku-hawlgalayaasha otomaatiga iyo abaabulka si loo ogaado oo loo xalliyo shilalka ka hor intaysan saameyn ganacsiga. Maqaalkan, waxaan ku baari doonaa sida loo hirgeliyo tuubooyinka DevOps ee AI ku shaqeeya iyadoo la adeegsanayo DSPy iyo LangChain, iyadoo diiradda la saarayo baaritaanka shilalka firfircoon iyo xallinta iyadoo la adeegsanayo AIOps iyo moodooyinka RAG, iyo isku dhafka AWS DevOps Agent ee baaritaanka shilalka otomaatiga ah.
Dulmarka DSPy iyo LangChain
DSPy waa maktabad Python oo bixisa hab fudud oo hufan oo lagu dhiso laguna hawlgeliyo dhuumaha DevOps ee ku shaqeeya AI. Waxay u ogolaataa horumariyayaashu inay ku dhex daraan noocyo kala duwan oo AI iyo ML ah, oo ay ku jiraan habaynta luqadda dabiiciga ah (NLP), aragtida kombuyuutarka, iyo falanqaynta saadaalinta, ee tubooyinka DevOps. LangChain, dhanka kale, waa maktabad Python ah oo u sahlaysa horumariyayaashu inay dhisaan oo geeyaan moodooyinka AI ee baaxadda leh iyagoo isticmaalaya qaab-dhismeedka qaab-dhismeedka iyo qaab-dhismeedka cabbirka leh.
Dhaqangelinta AI-Powered DevOps dhuumaha DSPy iyo LangChain
Si loo hirgeliyo tuubooyinka DevOps ee ku shaqeeya AI ee leh DSPy iyo LangChain, waxaan u baahanahay inaan raacno tillaabooyinkan:
- Aruurinta Xogta: Ka soo ururi xogta laxidhiidha ilo kala duwan, sida diiwaanka, cabirka, iyo galalka habaynta.
- Preprocessing Data: Horay u socodsii xogta la ururiyey si loogu diyaariyo falanqaynta iyo qaabaynta.
- **Tababarka Model ***: Tababar AI iyo moodooyinka ML adoo isticmaalaya xogta hore loo sii diyaariyay si loo ogaado loona saadaaliyo dhacdooyinka.
- ** Soo Dejinta Qaabka ***: Geli moodooyinka tababaran ee dhuumaha DevOps adoo isticmaalaya DSPy iyo LangChain.
- Baaritaanka Dhacdada: Adeegso moodooyinka la diray si aad u baadho shilalka una aqoonsato sababaha asaasiga ah.
- Xallinta Dhacdada: Isticmaal fikradaha laga helay baadhista dhacdada si aad shilalka ugu xalliso si firfircoon.
Tusaalaha Koodhka: Hirgelinta Qaabka RAG ee Fudud ee DSPy iyo LangChain
Waa kuwan tusaale kood fudud oo muujinaya sida loo hirgeliyo qaabka RAG (Falanqaynta Sababta xididka) iyadoo la adeegsanayo DSPy iyo LangChain: `` Python soo dejinta dspy Laga soo dejiyo langchain LLMChain
Soo rar xogta la sii diyaariyay
xogta = pd.read_csv("data.csv")
Qeex qaabka RAG
rag_model = dspy.RAGModel( input_features=["log_level", "error_message"], Output_features=["root_cause"] )
Tababar qaabka RAG
rag_model.tareen(xogta)
Geli qaabka RAG
llm_chain = LLMChain ( model="davinci:002", max_length=2048 ) rag_model.deploy(llm_chain)
Isticmaal qaabka RAG ee la diray si aad u baarto shilalka
Dhacdada_data = pd.DataFrame ({ "log_level": ["ERROR"], "error_message": ["Waxbaa khaldamay"] }) root_cause = rag_model.predict(xogta_dhacdooyinka) daabac (xidid_sabab) `` Tusaalaha koodka ayaa muujinaya sida loo hirgeliyo qaabka RAG fudud iyadoo la adeegsanayo DSPy iyo LangChain. Qaabka RAG wuxuu qaataa heerka log iyo fariinta qaladka sida astaamaha gelinta wuxuuna saadaaliyaa sababta asalka u ah dhacdada.
5-10ka amar ee ugu badan ee la isticmaalo
Waa kuwan 5-10-ka amarrada ugu badan ee la isticmaalo ee DSPy iyo LangChain:
- `dspy.RAGModel
dspy.RAGModel.train(xogta): Tababar qaabka RAG adigoo isticmaalaya xogta la bixiyay.dspy.RAGModel.deploy(llm_chain): Geli nooc RAG tababaran adoo isticmaalaya LangChain LLMChain.llm_chain = LLMChain (qaabka, max_dhererka): Samee LangChain LLMChain leh moodal cayiman iyo dhererka ugu sarreeya.rag_model.predict(data): Adeegso qaabka RAG ee la diray si aad u saadaaliso asalka sababta shilka.dspy.DataPreprocessor(input_data): Qeex dhuunta xogta ka horraysa adoo isticmaalaya DSPy.- `dspy.ModelTrainer(model, data): Tababar qaabka AI ama ML adigoo isticmaalaya DSPy.
dspy.ModelDeployer(model, llm_chain): Gey nooc tababaran adoo isticmaalaya DSPy iyo LangChain.langchain.LLMChain.predict(data_input): Adeegso LangChain LLMChain si aad u saadaaliso soosaarka moodeel.dspy.RAGModel.valuate(xogta): Qiimee waxqabadka qaabka RAG adoo isticmaalaya DSPy.
Muuqaalka
Waa kuwan tusaale sawir-qaadid iyadoo la isticmaalayo Chart.js:
json { "type": "bar", "xogta": { "labels": ["Dhacdada Nooca A", "Dhacdada Nooca B", "Dhacdada Nooca C"], "datasets": [ { "label": "Sababta Xididada A", "xogta": [10, 15, 7], "backgroundColor": "rgba(255, 99, 132, 0.2)", "BorderColor": "rgba(255, 99, 132, 1)", "BorderWidth": 1 }, { "label": "Sababada B", "xogta": [8, 12, 9], "backgroundColor": "rgba(54, 162, 235, 0.2)", "BorderColor": "rgba(54, 162, 235, 1)", "BorderWidth": 1 }, { "label": "Sababada C", "xogta": [14, 18, 11], "backgroundColor": "rgba(255, 206, 86, 0.2)", "BorderColor": "rgba(255, 206, 86, 1)", "BorderWidth": 1 } ] }, "ikhtiyaar": { "ciwaan": { "muujin": run, "text": "Falanqaynta Sababta xididka" }, "Miisaanka": { "yAxes": [ { "ticks": { "beginAtZero": run } } ] } } }
Aragtidani waxay muujinaysaa qaybinta sababaha asalka u ah noocyada dhacdooyinka kala duwan.
Gabagabo
Maqaalkan, waxaan ku sahamiyay sida loo hirgeliyo tuubooyinka DevOps ee ku shaqeeya AI iyadoo la adeegsanayo DSPy iyo LangChain, iyadoo diiradda la saarayo baaritaanka shilalka firfircoon iyo xallinta iyadoo la adeegsanayo AIOps iyo moodooyinka RAG, iyo isku dhafka AWS DevOps Agent ee baaritaanka shilalka otomaatiga ah. Waxaan soo bandhignay tusaale kood fudud oo fuliya qaabka RAG iyadoo la adeegsanayo DSPy iyo LangChain, waxaanan bixinnay liiska 5-10 ee ugu sarreeya amarada/qaybaha loo isticmaalo DSPy iyo LangChain. Waxa kale oo aanu ku darnay tusaale sawir-qaadis anagoo adeegsanayna Chart.js. Markaad raacdo tillaabooyinka lagu qeexay maqaalkan, horumariyayaashu waxay dhisi karaan oo geyn karaan tuubooyinka DevOps ee AI ku shaqeeya kuwaas oo ogaanaya oo xallinaya shilalka si firfircoon, hagaajinta waxtarka guud iyo isku halaynta nidaamyada softiweerkooda.