01
Fondation IA
Xam luy IA, ni sistem yi di jàngee, fan lañuy lajj, ak ni ñuy àttee kàddu yi te duñu yëngu.
Bibliothèque IA gratuite
Xarala tegtalMoom ba fàww.
177 guide yu leer ci Àngle, yooni jàng yuñ yamale, ak bibliotek bu ubbeeku — bi 501(c)(3) mbootaay bu moom boppam tabax ngir ku nekk mëna xam IA bu bees bi.
177Gid yu amul fayda
1topic topic
~2 minguide bu nekk
~6hJamonoy jàng
Tambalil fii
Juroom njàngale yu lalu ci njariñ
Njang moomu dafay àndaale ak ay njariñ yu leer, ay mënin yuñ mëna def ci kàrt, ay liggéey yuñ ciy jëfandikoo, ak benn capstone buñ jëfandikoo.
02
Jëfandikukat IA bu am responsabilite
Jëfandikool IA ci anam wu am njariñ boole ci aar sa bopp, saytu li ñuy génne, ba noppi baña bàyyi xel ci nit ñi.
03
IA ci liggéey
Saytu anam yi ñuy jëfandikoo ci barabu liggéey bi, def ay pilote yu wóor, natt valeur yi, te joxe xibaar ci coppite yi ci anam wu jaar yoon.
04
Politigu IA ak Askan
Jàngat sistem IA ci yelleef, yamale, nguur, kaaraange ak njariñu njariñu ñépp.
05
Tabax ak sistem IA
Xam xeeti làkk yi, seetlu yi, ndawu liggéey yi, jàngat yi, njëg yi, ak kaaraange jëfandikoo gi jaaraleko ci jëmmal sistem bi.
topic topic
Saytu ci piste
Tëmb ci barab bi nga bëgg. Bépp track amna guide yu bari ci Àngle bu leer.
Fondamentaal yiBaatu-jàngat yi: ni IA di doxee ci dayo bu kawe.84 tegtal IA làkkLLMs, PNL, ChatGPT, laaj, ak xam mbind.155 tegtal IA buy waneDefar nataal, xool ordinatër, wideo ak jëmmal.133 tegtal Audio IABaat, music, podcast, ak IA bu njëkk ci audio.117 tegtal Sosiete yiProfil yu OpenAI, Anthropic, Google, Meta, ak yeneen.121 tegtal Aplikaasioŋ yiNaka IA di feeñee ci kodage, seetlu, ndawu liggéey ak dundu bis bu nekk.111 tegtal Liggéeyukaay yiIA ci wàllu faju, xaalis, njàng, kuuraŋ ak yeneen mbir.91 tegtal XaralaPuce yi, ajustement bu baax, RAG, vecteur ak jumtukaayi IA.177 tegtal Askan wiTeg sa bopp, nëbbëtu, yelleefu moomeel, kaaraange, ak ëlëgu IA.30 tegtal
Bibliothèque bu mat
guide yépp
177 de 1019 tegtal yi ñu wane. Seggal ci track wala seetlu ci kaw.
Xarala
Streaming buy xalaat ak wax luy am ci token yu bari
Streaming speculatif ak wax luy waaja am ci token yu bari dañuy gaawal defar modelu làkk ci xayma ay token yu bari ci ëlëg benn yoon ba noppi xool leen ci benn yoon...
XaralaBERTScore ak jàngat ci luy tekki
BERTScore dafay natt ni mbind mi masin defaree méngoo ak royuwaay bi, ci méngale li muy tekki, te baña méngale kàddu yi gëna jub.
XaralaLineer ak Kernel yiy def
Foofu ligneer mooy wecci fowokaayu softmax quadratic ci Transformers ak benn kaf math buy eskale ligneer ak guddaayu toppalante bi.
XaralaYaRN ak guddaayu ëmbiit li
YaRN (Beneen RoPE extension) xarala bu am solo la ngir tàllal palanteer bi ñuy jëfandikoo ci model bi, dem fu sori ci li ñu ko tàggat.
Xaralainterpolaasioŋ posisioŋ ngir muy lu gudd
Interpolation positionnelle (PI) pexem yomb la, am doole buy yokk palanteer bi ci Transformateur ci di tas indices position yu bees ci biir...
Xaralanjaxasu kàngam yi ci LoRA
Njaxasu LoRA Experts (MoLE) dafa boole adaptatër yu ndaw yu bari, yu yomb te ñu tàggat ak routeur buñ jàng, suko defee benn modelu base mën na nekk ci anam wu yomb...
XaralaCoppite yu am solo ngir modeli kode
Coppite yu xalaat yi dañuy tax soppi kodu IA di yëg ci saasi ndax dafay wax luy waaja am ci fichier bi du soppeeku, te dafay xool pàcc yu ndaw yi wuute.
XaralaIngénieur de Direction ak Representation
Direksioŋ biy tàmbali dafay taxawal doxalinu model bi ci yokk wala dindi ay vecteur ci biir aktivaasioŋ yu nëbbu yi ci runtime, soxlawul ñu tàggataat ko.
XaralaAutoencoder yu néew ngir mëna tekki
Sparse autoencoders (SAEs) jumtukaay la buy dindi mbir yu jaxasoo yi ci biir reso neuronal ci benn xeetu set bu gëna mag...
XaralaSupposisioŋ ak Polisemantik ci IA buy Tekki
Jàngaleel superposisioŋ ak polysemanticity ci tekki IA: lu tax model yi dañuy boole màndarga yi ci yoon yuñ bokk, lu tax neuron yi nekk polysemantic...
XaralaLentille Logit et Lentille Tune
Lens logit ak lens tuned ay pexem tekki lañu yuy xool ci nekkin yu nëbbu yi ci transformateur bi layer par layer ngir xam li model bi di 'xalaat'...
XaralaGaawaay bu xóot ak Megatron
DeepSpeed (Microsoft) and Megatron-LM (NVIDIA) are the software stacks that make training models with billions of parameters across thousands of GPUs…
Jeexal nga jàng? Firndeel ko.
Check what you learned with a topic quiz, then explore our structured courses or work toward a certificate. Every guide stays free to read.