TIL
  • Main
  • TIL : ML
    • AI News
      • AI-powered coding, free of charge with Colab
    • Paper
      • Retrieval-augmented generation for knowledge-intensive nlp tasks
        • reference 정리
      • Sequence to Sequence
    • Generative AI
      • LLM
        • κΈ°λ³Έ κ°œλ…
    • Boostcamp
      • Daily Report
        • 2021
          • 09
            • week4
            • week5
          • 10
            • week1
            • week2
            • week3
            • week4
          • 11
            • week1
            • week2
            • week3
            • week4
            • week5
          • 12
            • week1
            • week2
            • week3
      • Final Project
        • ν”„λ‘œμ νŠΈ 기획
          • πŸŒŸμ΅œμ’… ν”„λ‘œμ νŠΈ 기획
          • μ΅œμ’… ν”„λ‘œμ νŠΈ Version
          • 아이디어 μˆ˜μ§‘
          • μš•μ„€, ν˜μ˜€λ°œμ–Έ 감지
          • 라이브 컀머슀 λ ˆν¬νŒ… ν”„λ‘œμ νŠΈ
        • ν”„λ‘œμ νŠΈ 진행
          • week1
          • week2
          • week3
      • Competition
        • 1. [NLP] λ¬Έμž₯ λ‚΄ κ°œμ²΄κ°„ 관계 μΆ”μΆœ
          • Day1 (9.27, μ›”)
          • Day2-3 (9.28~29, ν™”~수)
          • Day4 (9.30, λͺ©)
          • Day5 (10.1, 금)
          • Day6~7 (10.2~3, ν† ~일)
          • Day8 (10.4, μ›”)
          • Day9 (10.5, ν™”)
          • Day10 (10.6, 수)
          • Day 11 (10.7 λͺ©)
        • 2. [NLP] MRC ν”„λ‘œμ νŠΈ
          • Day1 (10.25, μ›”)
          • Day2 (10.26, ν™”)
          • Day3 (10.27, 수)
          • Day4-5 (10.28-29, λͺ©-금)
          • Day6 (11.1, μ›”)
          • Day7 (11.2, ν™”)
          • Day8 (11.3, 수)
          • Day9 (11.4, λͺ©)
        • πŸ”¨3. [NLP] 데이터 μ œμž‘
          • Day1
        • πŸ”¨4. [곡톡] λͺ¨λΈ κ²½λŸ‰ν™”
      • [U Stage] - DL basic
        • (01κ°•) λ”₯λŸ¬λ‹ κΈ°λ³Έ μš©μ–΄ μ„€λͺ… - Historical Review
        • (02κ°•) λ‰΄λŸ΄ λ„€νŠΈμ›Œν¬ - MLP (Multi-Layer Perceptron)
        • (03κ°•) Optimization
        • πŸ”¨(04κ°•) Convolution은 무엇인가?
        • πŸ”¨(05κ°•) Modern CNN - 1x1 convolution의 μ€‘μš”μ„±
        • πŸ”¨(06κ°•) Computer Vision Applications
        • (07κ°•) Sequential Models - RNN
        • (08κ°•) Sequential Models - Transformer
        • Page 2
      • [U Stage] - PyTorch
        • (01κ°•) Introduction to PyTorch
        • (02κ°•) PyTorch Basics
        • (03κ°•) PyTorch ν”„λ‘œμ νŠΈ ꡬ쑰 μ΄ν•΄ν•˜κΈ°
        • (04κ°•) AutoGrad & Optimizer
        • (05κ°•) Dataset & Dataloader
        • (06κ°•) λͺ¨λΈ 뢈러였기
        • (07κ°•) Monitoring tools for PyTorch
        • (08κ°•) Multi-GPU ν•™μŠ΅
        • (09κ°•) Hyperparameter Tuning
        • (10κ°•) PyTorch Troubleshooting
      • [U Stage] - NLP
        • (01κ°•) Introduction to NLP, Bag-of-Words
        • (02κ°•) Word Embedding
        • (03κ°•) Recurrent Neural Network and Language Modeling
        • (04κ°•) LSTM and GRU
        • (05κ°•) Sequence to Sequence with Attention
        • (06κ°•) Beam Search and BLEU score
        • (07-08κ°•) Transformer
        • (09κ°•) Self-supervised Pre-training Models
      • [P Stage] - KLUE
        • (1κ°•) 인곡지λŠ₯κ³Ό μžμ—°μ–΄ 처리
        • (2κ°•) μžμ—°μ–΄μ˜ μ „μ²˜λ¦¬
        • (3κ°•) BERT μ–Έμ–΄λͺ¨λΈ μ†Œκ°œ
        • (4κ°•) ν•œκ΅­μ–΄ BERT μ–Έμ–΄ λͺ¨λΈ ν•™μŠ΅
        • (5κ°•) BERT 기반 단일 λ¬Έμž₯ λΆ„λ₯˜ λͺ¨λΈ ν•™μŠ΅
        • (6κ°•) BERT 기반 두 λ¬Έμž₯ 관계 λΆ„λ₯˜ λͺ¨λΈ ν•™μŠ΅
        • (7κ°•) BERT μ–Έμ–΄λͺ¨λΈ 기반의 λ¬Έμž₯ 토큰 λΆ„λ₯˜
        • μ˜€ν”ΌμŠ€μ•„μ›Œ (9.30, λͺ©)
        • (8κ°•) GPT μ–Έμ–΄ λͺ¨λΈ
        • (9κ°•) GPT μ–Έμ–΄λͺ¨λΈ 기반의 μžμ—°μ–΄ 생성
        • (10κ°•) μ΅œμ‹  μžμ—°μ–΄μ²˜λ¦¬ 연ꡬ
      • [P Stage] - MRC
        • Before Study
        • (1κ°•) MRC Intro & Python Basics
        • (2κ°•) Extraction-based MRC
        • (3κ°•) Generation-based MRC
        • (4κ°•) Passage Retrieval - Sparse Embedding
        • (5κ°•) Passage Retrieval - Dense Embedding
        • μ˜€ν”ΌμŠ€μ•„μ›Œ
        • (6κ°•) Scaling up with FAISS
        • (7κ°•) Linking MRC and Retrieval
        • (8κ°•) Reducing Training Bias
        • (9κ°•) Closed-book QA with T5
        • (10κ°•) QA with Phrase Retrieval
        • λ§ˆμŠ€ν„°ν΄λž˜μŠ€
      • [P Stage] - λ°μ΄ν„°μ œμž‘(NLP)
        • (1κ°•) 데이터 μ œμž‘μ˜ A to Z
        • (2κ°•) μžμ—°μ–΄μ²˜λ¦¬ 데이터 기초
        • (3κ°•) μžμ—°μ–΄μ²˜λ¦¬ 데이터 μ†Œκ°œ 1
        • (4κ°•) μžμ—°μ–΄μ²˜λ¦¬ 데이터 μ†Œκ°œ 2
        • (5κ°•) μ›μ‹œ λ°μ΄ν„°μ˜ μˆ˜μ§‘κ³Ό 가곡
        • μ˜€ν”ΌμŠ€μ•„μ›Œ (11.10, 수)
        • (6κ°•) 데이터 ꡬ좕 μž‘μ—… 섀계
        • (7κ°•) 데이터 ꡬ좕 κ°€μ΄λ“œλΌμΈ μž‘μ„± 기초
        • (8κ°•) 관계 μΆ”μΆœ 과제의 이해
        • (9κ°•) 관계 μΆ”μΆœ κ΄€λ ¨ λ…Όλ¬Έ 읽기
        • (10κ°•) 관계 μΆ”μΆœ 데이터 ꡬ좕 μ‹€μŠ΅
      • [P Stage] - λͺ¨λΈ μ΅œμ ν™”
        • (1κ°•) μ΅œμ ν™” μ†Œκ°œ 및 κ°•μ˜ κ°œμš”
        • (2κ°•) λŒ€νšŒ 및 데이터셋 μ†Œκ°œ
        • (3κ°•) μž‘μ€ λͺ¨λΈ, 쒋은 νŒŒλΌλ―Έν„° μ°ΎκΈ°: AutoML 이둠
        • πŸ”¨(4κ°•) μž‘μ€ λͺ¨λΈ, 쒋은 νŒŒλΌλ―Έν„° μ°ΎκΈ°: AutoML μ‹€μŠ΅
        • (5κ°•) μž‘μ€ λͺ¨λΈ, 쒋은 νŒŒλΌλ―Έν„° μ°ΎκΈ°: Data Augmentation & AutoML κ²°κ³Ό 뢄석
        • πŸ”¨μ˜€ν”ΌμŠ€μ•„μ›Œ -Baseline μ½”λ“œμ— λͺ¨λ“ˆ μž‘μ„±ν•˜κΈ°(μ‹ μ’…μ„  λ©˜ν† λ‹˜)
      • [P Stage] - Product Serving
        • Part 1: Product Serving 개둠
          • 1.1 κ°•μ˜ 진행 방식
          • 1.2 MLOps 개둠
          • 1.3 Model Serving
          • 1.4 λ¨Έμ‹ λŸ¬λ‹ ν”„λ‘œμ νŠΈ 라이프 사이클
        • Part 2: ν”„λ‘œν† νƒ€μž…λΆ€ν„° μ μ§„μ μœΌλ‘œ κ°œμ„ ν•˜κΈ°
          • 2.1 ν”„λ‘œν† νƒ€μ΄ν•‘ - Notebook 베이슀(Voila)
          • 2.2 ν”„λ‘œν† νƒ€μ΄ν•‘ - μ›Ή μ„œλΉ„μŠ€ ν˜•νƒœ(Streamlit)
          • 2.3 Linux & Shell Command
          • 2.4 Cloud
          • 2.5 Github Action을 ν™œμš©ν•œ CI/CD
        • Part 3: 더 μ™„μ„±ν™”λœ μ œν’ˆμœΌλ‘œ
          • 3.1.1 FastAPI
          • 3.1.2 Fast API
          • 3.1.3 Fast API
          • 3.2 Docker
          • 3.3 Logging
          • 3.4 MLFlow
        • Part 4: 심화 μ†Œμž¬
          • 4.1 BentoML
          • 4.2 Airflow
          • 4.3 λ¨Έμ‹ λŸ¬λ‹ λ””μžμΈ νŒ¨ν„΄
          • 4.4 μ•žμœΌλ‘œ 더 κ³΅λΆ€ν•˜λ©΄ 쒋을 λ‚΄μš©
      • νŠΉκ°•
        • (νŠΉκ°•) κΉ€μƒν›ˆ - 캐글 κ·Έλžœλ“œλ§ˆμŠ€ν„°μ˜ λ…Έν•˜μš° λŒ€λ°©μΆœ
        • (νŠΉκ°•) μ΄ν™œμ„ - μ„œλΉ„μŠ€ ν–₯ AI λͺ¨λΈ κ°œλ°œν•˜κΈ°
        • (νŠΉκ°•) κ΅¬μ’…λ§Œ - AI + MLκ³Ό Quant Trading
        • (νŠΉκ°•) λ¬Έμ§€ν˜• - λ‚΄κ°€ λ§Œλ“  AI λͺ¨λΈμ€ ν•©λ²•μΌκΉŒ, λΆˆλ²•μΌκΉŒ
        • (νŠΉκ°•) 이쀀엽 - Full Stack ML Engineer
        • (νŠΉκ°•) 박은정 - AI μ‹œλŒ€μ˜ 컀리어 λΉŒλ”©
        • (νŠΉκ°•) μ˜€ν˜œμ—° - AI Ethics
    • Competition
      • (DACON)ν•œκ΅­μ–΄ λ¬Έμž₯ 관계 λΆ„λ₯˜ κ²½μ§„λŒ€νšŒ
        • Day1(2.14, μ›”)
        • Day2(2.15, ν™”)
        • Day3(2.16, 수)
        • Day4(2.17, λͺ©)
      • 2021 인곡지λŠ₯ 데이터 ν™œμš© κ²½μ§„λŒ€νšŒ
        • μ—­λŸ‰ν‰κ°€
          • Day1 (9.28, ν™”)
          • Day2 (9.29, 수)
          • Day3 (9.30, λͺ©)
        • μ˜ˆμ„ 
          • Data 뢄석
          • NSML
          • What We Have Done?
    • ETC
      • 인터뷰 λŒ€λΉ„
        • Computer Science
        • ML/DL
      • Poetry둜 dependency 관리
        • windowμ—μ„œ μ„€μΉ˜ν•˜κΈ°
      • code block
      • 곡뢀할 것 μž„μ‹œλ³΄κ΄€
      • Transformer to T5
      • Hugging Face Tutorial
        • Ch1. Transformer models
        • Ch2. Using Transformers
        • Ch3. Fine-tuning a model with the Trainer API
      • KLUE
      • Pandas
  • TIL : Ops
    • AWS
      • SageMaker
  • TIL : Computer (CS)
    • Error
      • TextEncodeInput must be Union[TextInputSequence, Tuple[InputSequence, InputSequence]]
    • Algorithm
      • Theory
      • Programmers
        • κΈ°λŠ₯개발
    • ETC
      • Github 컀밋 νžˆμŠ€ν† λ¦¬ μ‚­μ œ
Powered by GitBook
On this page

Was this helpful?

  1. TIL : ML
  2. Boostcamp
  3. Competition

πŸ”¨4. [곡톡] λͺ¨λΈ κ²½λŸ‰ν™”

1. κ°œμš”

  • 일정: 11.30 (μ›”) ~ 12.2 (λͺ©)

  • νŒ€: CLUE (7λͺ…)

PreviousDay1Next[U Stage] - DL basic

Last updated 3 years ago

Was this helpful?