terminal

PM AI LAB.ML in Practice

search
/

folder_open _categories

sell _tags

16์ง„๋ฒ•(1) 2์ง„๋ฒ•(1) 3V(1) 4V(1) ADsP(7) AI(9) AIBasic(13) AITech(45) Absmax Quantization(1) Algorithm(1) Arcface(1) Association Analysis(2) AudioLM(4) AutoEncoder(1) AutoRec(1) Black-box KD(1) CAR(1) CDAE(1) CF(1) CLIP(1) CRISP-DM ๋ถ„์„ ๋ฐฉ๋ฒ•๋ก (1) Clustering(4) Collaborative Filtering(1) Content-based Recommendation(1) Cross_validation(1) DBMS(3) DBSCAN(2) DIKWํ”ผ๋ผ๋ฏธ๋“œ(1) DNN(2) Database(3) Deep Learning(1) Deep_Learning(1) EDA(2) EM(1) ER(1) Ensemble(1) Exploding(1) FM(1) FP(1) Factorization_machine(1) Feature-based KD(1) Feature_engineering(2) Few-shot(1) Filter Pruning(1) Fourier_transform(1) GMM(3) GNN(1) GRU(1) GRU4Rec(1) GenAI(1) GenerativeAI(1) Helium(1) Hyperparameter(1) IBCF(1) IMP(1) INT(1) Imitation Learning(1) Item-based CF(1) Iterative Magnitude Pruning(1) K-Means(2) K-Nearest Neighbors CF(1) KD(2) KDD ๋ถ„์„ ๋ฐฉ๋ฒ•๋ก (1) Knowledge Distillation(2) LAN(2) LLM(3) LM(1) LSTM(1) Language_Modeling(1) LightGCN(1) Lightweight(2) Likelihood(1) Logit-based KD(1) Lottery Ticket Hypothesis(1) MFVI(1) ML(14) MLE(2) MLP(2) MNIST(1) Matrix Sparsity(1) Mimi(1) Moshi(1) Multi-teacher(1) NBCF(1) NCF(1) NDCG@K(1) NGCF(1) Neighborhood-based CF(1) Network(3) NoSQL(1) OS(3) On/Offline Test(1) One-shot(1) Orca(1) Overfitting(1) PRML(1) Paper_review(5) Popularity(1) Precison@K(1) Pretrained-model(1) Prototype(1) Pruning(2) Pruning in BERT(1) Pruning in CNN(1) PyTorch(8) Quantization(1) RDBMS(3) RNN(4) RecSys(9) Recall@K(1) Retrieval(1) SALMONN(1) SQL(1) Sensitivity Analysis(1) Streamlit(1) TF-IDF(1) UBCF(1) User-based CF(1) VAE(1) VI(1) Vanishing(1) Variational_Inference(1) Variety(1) Velocity(1) Veracity(1) Vicuna(1) Volume(1) WAN(2) Whisper(1) WizardLM(1) Word2Vec(1) YouTube_Recommendation(1) Zero-point Quantization(1) Zero-shot(1) amplitude(1) array(1) audio(4) bitdepth(1) codec(2) computer_science(18) context-aware(1) cs(18) dacon(1) embedding(2) frequency(2) frequency_domain(1) greedy(3) hash_table(1) kaggle(1) lightweighting(5) linked_list(1) neural(1) node(1) optuna(1) python(34) read(1) readline(1) recsys(17) sampling(1) solved(34) stack(1) stdin(1) sys(1) time_domain(1) tree(1) wandb(1) waveform(1) ๊ฐ€์ƒ๋ฉ”๋ชจ๋ฆฌ(1) ๊ฐ€์น˜ ํŒจ๋Ÿฌ๋‹ค์ž„(1) ๊ฐ€์น˜๊ธฐ๋ฐ˜ ๋ถ„์„(1) ๊ฐ์ •๋ถ„์„(1) ๊ฐ•ํ™”ํ•™์Šต(1) ๊ณต๊ฐ„๋ณต์žก๋„(1) ๊ณผ์ ํ•ฉ(1) ๊ต์ฐจ๊ฒ€์ฆ(1) ๊ตฌํ˜„(2) ๊ทธ๋ฆฌ๋””(3) ๊ธฐ์ˆ (1) ๊ธฐ์šธ๊ธฐ์†Œ์‹ค(1) ๊ธฐ์šธ๊ธฐํญ๋ฐœ(1) ๋„คํŠธ์›Œํฌ(3) ๋…ธ๋“œ(1) ๋”๊ทธ ๋ž˜๋‹ˆ(1) ๋ฐ์ดํ„ฐ(2) ๋ฐ์ดํ„ฐ ์‚ฌ์ด์–ธ์Šค(1) ๋ฐ์ดํ„ฐ ์‚ฌ์ด์–ธ์Šค์™€ ์ธ๋ฌธํ•™(1) ๋ฐ์ดํ„ฐ ์‚ฌ์ด์–ธ์Šค์˜ ๊ตฌ์„ฑ์š”์†Œ(1) ๋ฐ์ดํ„ฐ ์‚ฌ์ด์–ธํ‹ฐ์ŠคํŠธ(1) ๋ฐ์ดํ„ฐ๋ฒ ์ด์Šค(1) ๋ฐ์ดํ„ฐ๋ฒ ์ด์Šค ์ •์˜(1) ๋ฐ์ดํ„ฐ๋ฒ ์ด์Šค ์ข…๋ฅ˜(1) ๋ฐ์ดํ„ฐ๋ฒ ์ด์Šค ํŠน์ง•(1) ๋ฐ์ดํ„ฐ๋ฒ ์ด์Šค ํ™œ์šฉ(1) ๋ฐ์ดํ„ฐ์™€ ์ •๋ณด(1) ๋ฐ์ดํ„ฐ์˜ ์ดํ•ด(1) ๋ฐ์ดํ„ฐ์˜ ์ •์˜(1) ๋””์ž์ธ ์”ฝํ‚น(1) ๋จธ์‹ ๋Ÿฌ๋‹(1) ๋ฐฐ์—ด(1) ๋ฐฑ์ค€(34) ๋ณ€๋ถ„์ถ”๋ก (1) ๋ถ€์ŠคํŠธ์บ ํ”„(45) ๋ถ„๋ฅ˜(1) ๋ถ„์„ ๊ณผ์ œ ํƒ์ƒ‰(1) ๋ถ„์„ ๊ธฐํš(1) ๋ถ„์„ ๋ฐฉ๋ฒ•๋ก (1) ๋น„์ง€๋„ํ•™์Šต(1) ๋น…๋ฐ์ดํ„ฐ(3) ๋น…๋ฐ์ดํ„ฐ ๋ถ„์„ ๋ฐฉ๋ฒ•๋ก (1) ๋น…๋ฐ์ดํ„ฐ ํ™œ์šฉ 3์š”์†Œ(1) ๋น…๋ฐ์ดํ„ฐ ํ™œ์šฉ ํ…Œํฌ๋‹‰ 7๊ฐ€์ง€(1) ๋น…๋ฐ์ดํ„ฐ๊ฐ€ ์ƒ์„ฑํ•˜๋Š” ๋ณ€ํ™”(1) ๋น…๋ฐ์ดํ„ฐ์™€ ์ธ์‚ฌ์ดํŠธ(1) ์ƒํ–ฅ์‹ ์ ‘๊ทผ๋ฒ•(1) ์†Œ์…œ ๋„คํŠธ์›Œํฌ ๋ถ„์„(1) ์†๊ธ€์”จ ์ธ์‹(1) ์Šค๋ ˆ๋“œ(2) ์Šคํƒ(1) ์‹œ๊ฐ„๋ณต์žก๋„(1) ์‹œ๋ฎฌ๋ ˆ์ด์…˜(2) ์‹œ์Šคํ…œ์ฝœ(1) ์•„ํ‚คํ…์ฒ˜(3) ์•Œ๊ณ ๋ฆฌ์ฆ˜(1) ์•™์ƒ๋ธ”(1) ์—ญ์ •๊ทœํ™”(1) ์—ฐ๊ฒฐ๋ฆฌ์ŠคํŠธ(1) ์—ฐ๊ด€ ๊ทœ์น™ ํ•™์Šต(1) ์˜ค๋””์˜ค์••์ถ•(1) ์™„์ „ํƒ์ƒ‰(2) ์šด์˜์ฒด์ œ(6) ์œ ์ „ ์•Œ๊ณ ๋ฆฌ์ฆ˜(1) ์œ ํ˜• ๋ถ„์„(1) ์ด์ง„๋ฒ•(1) ์ธ๊ณต์ง€๋Šฅ(1) ์ธ๋ ฅ(1) ์ผ์ฐจ์›์  ๋ถ„์„(1) ์ž„๋ฒ ๋”ฉ(1) ์ž๋ฃŒ๊ตฌ์กฐ(6) ์ž์—ฐ์–ด์ฒ˜๋ฆฌ(4) ์ž์›ํ• ๋‹น(1) ์ „์ฒ˜๋ฆฌ(1) ์ •๊ทœํ˜•(1) ์ •๊ทœํ™”(1) ์ฃผํŒŒ์ˆ˜์˜์—ญ(1) ์ค‘์‹ฌ๊ทนํ•œ์ •๋ฆฌ(1) ์ง€๋„ํ•™์Šต(1) ์ฐจ์›๊ฐ์†Œ(1) ์ถ”๋ก ๊ธฐ๋ฐ˜๊ธฐ๋ฒ•(1) ์ปดํ“จํ„ฐ(3) ์ปดํ“จํ„ฐ๊ตฌ์กฐ(8) ํด๋ผ์šฐ๋“œ ์ปดํ“จํŒ…(1) ํด๋Ÿฌ์Šคํ„ฐ๋ง(5) ํ…์ŠคํŠธ ๋งˆ์ด๋‹(1) ํ†ต๊ณ„(4) ํŠœ๋‹(1) ํŠธ๋ฆฌ(1) ํŠน์ง• ์ถ”์ถœ(1) ํŒจํ‚ท(2) ํŒจํ„ด์ธ์‹(1) ํ‘ธ๋ฆฌ์—๋ณ€ํ™˜(1) ํ”„๋กœ์„ธ์Šค(2) ํ”„๋กœํ† ์ฝœ(2) ํ•˜์ดํผํŒŒ๋ผ๋ฏธํ„ฐ(1) ํ•˜ํ–ฅ์‹ ์ ‘๊ทผ๋ฒ•(1) ํ•ด์‹œํ…Œ์ด๋ธ”(1) ํ•ด์‹œํ•จ์ˆ˜(1) ํ™•๋ฅ ๋ถ„ํฌ(1) ํšŒ๊ท€๋ถ„์„(1)
POSTED ON 2024-09-22 2 min read

[ML/์‹คํ—˜ ๊ด€๋ฆฌ] ๋จธ์‹ ๋Ÿฌ๋‹ ๋ชจ๋ธ ์‹คํ—˜ ๊ด€๋ฆฌ

1. Experiment Management ์‹คํ—˜ ๊ด€๋ฆฌ ๋จธ์‹ ๋Ÿฌ๋‹ ์‹คํ—˜ ๊ด€๋ฆฌ ML ๋ชจ๋ธ ๊ฐœ๋ฐœ ๋ฐ ์ตœ์ ํ™” ๊ณผ์ •์—์„œ, ์‹คํ—˜์— ์‚ฌ์šฉ๋œ ๋‹ค์–‘ํ•œ ์š”์†Œ๋“ค์„ ์ฒด๊ณ„์ ์œผ๋กœ ์ถ”์  ๋ฐ ๊ด€๋ฆฌํ•˜๋Š” ๊ฒƒ ML ์‹คํ—˜์—์„œ ์ค‘์š”ํ•œ ๋ฉ”ํƒ€๋ฐ์ดํ„ฐ ...

#ml_in_practice #๋ถ€์ŠคํŠธ์บ ํ”„ #AITech
POSTED ON 2024-09-20 5 min read

[ML/๋ชจ๋ธ ํŠœ๋‹ ๊ธฐ๋ฒ•] ์•™์ƒ๋ธ”

1. ๋ชจ๋ธ ์•™์ƒ๋ธ” ์•™์ƒ๋ธ”(Ensemble) ๋ฐ์ดํ„ฐ ์˜ˆ์ธก ์ˆ˜ํ–‰ ์‹œ, ๋‹จ์ผ ๋ชจ๋ธ๋งŒ์„ ์‚ฌ์šฉํ•˜๊ธฐ ๋ณด๋‹ค ์—ฌ๋Ÿฌ ๋ชจ๋ธ์„ ๊ฐ™์ด ์ด์šฉํ•˜๋Š” ๊ฒƒ์ด ๋” ์ข‹์€ ์˜ˆ์ธก ์„ฑ๋Šฅ์„ ๋ณด์ผ ์ˆ˜ ์žˆ์Œ โ†’ ๋‹จ์ผ ์•Œ๊ณ ๋ฆฌ์ฆ˜์œผ๋กœ ์ ๋‹นํ•œ ์•Œ๊ณ ๋ฆฌ์ฆ˜์„ ์—ฌ๋Ÿฌ ๊ฐœ ์กฐํ•ฉํ•˜์—ฌ, ๋‹จ์ผ ์•Œ๊ณ ๋ฆฌ์ฆ˜์— ๋น„ํ•œ ์„ฑ๋Šฅ ํ–ฅ์ƒ์„ ๊ธฐ...

#ml_in_practice #๋ถ€์ŠคํŠธ์บ ํ”„ #AITech
POSTED ON 2024-09-19 5 min read

[ML / ๋ชจ๋ธ ํ•™์Šต ํŒŒ์ดํ”„๋ผ์ธ] ๋ชจ๋ธ ๊ณผ์ ํ•ฉ / ๋ชจ๋ธ ๊ต์ฐจ ๊ฒ€์ฆ

1. ๋ชจ๋ธ ๊ณผ์ ํ•ฉ(Over-fitting) ๊ณผ์ ํ•ฉ ์ •์˜ ํ›ˆ๋ จํ•œ ๋ฐ์ดํ„ฐ์— ๋Œ€ํ•ด์„œ๋Š” ์˜ˆ์ธก์„ ์ž˜ํ•˜๋Š” ๋ฐ˜๋ฉด, ๊ฒ€์ฆ ๋ฐ์ดํ„ฐ๋‚˜ ์ƒˆ๋กœ์šด ๋ฐ์ดํ„ฐ์— ๋Œ€ํ•ด์„œ๋Š” ์„ฑ๋Šฅ์ด ๋–จ์–ด์ง€๋Š” ํ˜„์ƒ ๊ณผ์ ํ•ฉ vs ๊ณผ์†Œ์ ํ•ฉ ๊ณผ์†Œ์ ํ•ฉ ...

#ml_in_practice #๋ถ€์ŠคํŠธ์บ ํ”„ #AITech
POSTED ON 2024-09-14 4 min read

[ML / ML ๊ฐœ์š”] EDA / Feature Engineering

1. EDA ๊ฐœ์š” EDA๋ž€? ๋ฐ์ดํ„ฐ๋ฅผ ์š”์•ฝํ•˜๊ณ , ์ฃผ์š” ํŠน์„ฑ์„ ์‹œ๊ฐ์ ์œผ๋กœ ํƒ์ƒ‰ํ•˜๋Š” ๊ณผ์ • EDA ์ฃผ์š” ๋ชฉ์  ๋ฐ์ดํ„ฐ ๊ตฌ์กฐ ์ดํ•ด ์ด์ƒ์น˜ / ๋ฐ์ดํ„ฐ ํŒจํ„ด ๋ฐœ๊ฒฌ ๊ฐ€์„ค ์ˆ˜๋ฆฝ์„ ํ†ตํ•œ ํ†ต๊ณ„ ๋ชจ๋ธ๋ง...

#ml_in_practice #๋ถ€์ŠคํŠธ์บ ํ”„ #AITech
POSTED ON 2024-09-14 4 min read

[ML / ML ๊ฐœ์š”] ML ๊ฒฝ์ง„๋Œ€ํšŒ๋ฅผ ์ž„ํ•˜๋Š” ๋ฐฉ๋ฒ•

์š”์ฆ˜ ์˜จ๋ผ์ธ์˜ ์ˆ˜๋งŽ์€ ํ”Œ๋žซํผ์—์„œ ๋งŽ์€ ML ๊ฒฝ์ง„๋Œ€ํšŒ๊ฐ€ ์ง„ํ–‰๋˜๊ณ  ์žˆ๋‹ค. ๊ทธ ์ค‘ ๊ฐœ์ธ์ ์œผ๋กœ ์ฒ˜์Œ ML ๋Œ€ํšŒ๋ฅผ ์ง„ํ–‰ํ•ด๋ณด๋Š” ๋งŒํผ, ์ด์— ์•ž์„œ ์œ ๋…ํ•˜๊ณ  ๊ฐ€๋ฉด ์ข‹์„ ๋ถ€๋ถ„์— ๋Œ€ํ•ด ์ •๋ฆฌํ•ด ๋ณด๊ณ ์ž ํ•œ๋‹ค.

#ml_in_practice #๋ถ€์ŠคํŠธ์บ ํ”„ #AITech