Intangiriro

Impinduramatwara ya DevOps yahinduye uburyo software itezwa imbere, ikoreshwa, kandi icungwa. Ariko, uko bigoye sisitemu igezweho ya software igenda ikura, niko hakenerwa automatike, kubahiriza, hamwe nubunini. Aha niho haza DevOps ikoreshwa na AI - uburyo bwo guhindura umukino bukoresha ubwenge bwubukorikori (AI) no kwiga imashini (ML) kugirango byorohereze imiyoboro ya DevOps. Muri iki kiganiro, tuzasesengura uburyo bwo gutunganya ibikoresho bya AI bikoreshwa na AI hamwe na LangChain na DSPy, ibice bibiri bigezweho bifasha abakozi kwikora.

LangChain ni iki?

LangChain nurufunguzo rufunguye rwemerera abitezimbere kubaka ibiganiro bya AI ukoresheje modular, nini yububiko. Itanga urutonde rwibikoresho byateguwe mbere nibikoresho byo gukora AI ishobora guhuza namakuru atandukanye, gukora imirimo, no gufata ibyemezo. LangChain yubatswe hejuru yururimi rwa porogaramu ya Rust kandi irashobora gukoreshwa nizindi ndimi nka Python na Go.

DSPy ni iki?

DSPy ni isomero rya Python ritanga ibintu byoroshye, Pythonic interface kumurongo wa LangChain. Iyemerera abitezimbere kubaka abakozi ba AI bakoresheje syntax isanzwe imenyerewe hamwe nakazi, bigatuma byoroha guhuza LangChain numuyoboro wa Python ushingiye kuri DevOps.

Guhuza LangChain na DSPy hamwe numuyoboro wa DevOps

Kugirango duhuze LangChain na DSPy numuyoboro wa DevOps, dukeneye gukora umukozi wa AI ushobora gukora imirimo, gufata ibyemezo, no gukorana namakuru atandukanye. Dore urwego rwo hejuru murwego rwo hejuru rwibikorwa:

1. ** Sobanura umukozi wa AI **: Kora umukozi wa LangChain ushobora gukora imirimo, nko kohereza code, gukora ibizamini, no gukurikirana imikorere.
2. ** Kwinjiza hamwe nibikoresho bya DevOps **: Huza umukozi wa AI nibikoresho bya DevOps, nka Jenkins, GitLab CI / CD, cyangwa Docker.
3. ** Hindura abakozi ba AI **: Hindura umukozi wa AI gukorana namakuru atandukanye, nka data base, APIs, cyangwa sisitemu ya dosiye. 4 ..

Urugero Rufatika: Kohereza Kode hamwe na LangChain na DSPy

Dore urugero rworoshye rwuburyo bwo kohereza code ukoresheje LangChain na DSPy: `` python gutumiza mu mahanga Kuva kumurongo.ibikoresho bitumiza abakozi Kuva kumurongo.llms itumiza LLM

Kora umukozi wa LangChain

umukozi = Intumwa ( llm = LLM ( icyitegererezo_izina = "davinci", max_tokens = 2048, ubushyuhe = 0.5, ), urunigi = [ Urunigi. Urunigi ( izina = "kohereza_kode", intambwe = [ Ururimi. Intambwe ( izina = "git_clone", imikorere = lambda: git_clone ("https://github.com/user/repo.git"), ), Ururimi. Intambwe ( izina = "kubaka_code", imikorere = lambda: kubaka_code (), ), Ururimi. Intambwe ( izina = "kohereza_kode", imikorere = lambda: kohereza_kode (), ), ], ), ], )

Kohereza code ukoresheje umukozi wa AI

umukozi.deploy_code () `` ` Mururugero, dushiraho umukozi wa LangChain ukora intambwe eshatu:

1. Kora ububiko bwa kode ukoresheje Git.
2. Yubaka code ukoresheje igikoresho cyo kubaka.
3. Kohereza kode ukoresheje igikoresho cyo kohereza.

Umukozi wa AI akoresha icyitegererezo cyururimi (LLM) kugirango afate ibyemezo kandi akore imirimo.

Hejuru 5-10 Amabwiriza Yakoreshejwe Byinshi / Uduce

Hano hari amwe mumabwiriza akoreshwa cyane hamwe nibisobanuro bya LangChain na DSPy:

1. ** Kora umukozi wa LangChain **: agent = Intumwa (llm = LLM (model_name =" davinci ", max_tokens = 2048, ubushyuhe = 0.5), urunigi = git_clone ("https://github.com/user/repo.git"))])))
2. ** Kohereza kode ukoresheje umukozi wa AI **: agent.deploy_code ()
3. ** Hugura umukozi wa AI **: `agent.train (data =
4. ** Kwinjiza hamwe nibikoresho bya DevOps **: agent.integrate_with_jenkins ()
5. ** Hindura umukozi wa AI **: agent.configure (llm = LLM (model_name =" davinci ", max_tokens = 2048, ubushyuhe = 0.5))
6. ** Fata imiterere ya agent ya AI **: agent.status ()
7. ** Shakisha ibiti bya agent wa AI **: agent.logs ()
8. ** Kuvugurura iboneza ryabakozi ba AI **: agent.update_config (llm = LLM (model_name =" davinci ", max_tokens = 2048, ubushyuhe = 0.5))
9. ** Siba umukozi wa AI **: agent.delete ()
10. ** Fata ibipimo bya agent wa AI **: agent.metrics ()

Kugaragara

Dore amashusho yububiko bwa LangChain na DSPy ukoresheje Chart.js: json { "ubwoko": "akabari", "amakuru": { "ibirango": ["LangChain", "DSPy"], "imibare": [ { "label": "Ibigize", "amakuru": [10, 5], "backgroundColor": ["# 4CAF50", "# FF9800"], "umupaka w'amabara": ["# 3e8e41", "# ff9900"], "Ubugari bwumupaka": 1 } ] }, "amahitamo": { "umutwe": { "kwerekana": ukuri, "inyandiko": "LangChain na DSPy Ubwubatsi" }, "umugani": { "kwerekana": ukuri }, "umunzani": { "yAxes": [ { "amatiku": { "gutangiraAtZero": ukuri } } ] } } } ` Iyi visualisation yerekana umubare wibigize muri LangChain na DSPy.

Umwanzuro

Muri iki kiganiro, twasuzumye uburyo bwo gutunganya ibikoresho bya AI bikoreshwa na AI hamwe na LangChain na DSPy, ibice bibiri bigezweho bifasha abakozi kwikora. Twaganiriye ku nyungu za DevOps ikoreshwa na AI, harimo kuzamura automatike, kubahiriza, no gupima. Twatanze kandi urugero rufatika rwo kohereza code dukoresheje LangChain na DSPy, kimwe nurutonde rwamabwiriza akoreshwa cyane. Hanyuma, twashushanyije ubwubatsi bwa LangChain na DSPy dukoresheje Chart.js. Hamwe na LangChain na DSPy, abitezimbere barashobora kubaka imiyoboro ikora neza, igaragara, kandi ikora ya DevOps ikoresha imbaraga za AI no kwiga imashini.