Daesoo Lee's Blog

Data/Machine learning

• Export the environment specification with Anaconda (environment.yml)

  Data/Machine learning 2022. 8. 11. 18:21

  Similar to "requirements.txt", Anaconda provides an exporting option for its environment (i.e., installed libraries) as "environment.yml". You can simply run conda env create -f environment.yml Then, you'd get something like: Reference: https://www.anaconda.com/blog/moving-conda-environments

  Read More
• Informer

  Data/Machine learning 2022. 5. 19. 01:40

  Input Target Formatting Reference: https://github.com/zhouhaoyi/Informer2020/blob/main/data/data_loader.py (refer to __getitem__(..))

  Read More
• UNet

  Data/Machine learning 2022. 1. 10. 23:26

  Architecture Explained: https://towardsdatascience.com/unet-line-by-line-explanation-9b191c76baf5 Loss function Explained: https://towardsdatascience.com/u-net-b229b32b4a71

  Read More
• Recall and Precision

  Data/Machine learning 2022. 1. 10. 18:16

  Example: https://developers.google.com/machine-learning/crash-course/classification/precision-and-recall Classification: Precision and Recall | Machine Learning Crash Course | Google Developers Google is committed to advancing racial equity for Black communities. See how. Send feedback Classification: Precision and Recall Estimated Time: 9 minutes Precision Precision attempts to answer the follo..

  Read More
• Vision Transformer (ViT) Study Material

  Data/Machine learning 2021. 12. 5. 18:48

  1. https://youtu.be/j6kuz_NqkG0 2. https://youtu.be/TrdevFK_am4 3. What is the Class Token? One of the interesting things about the Vision Transformer is that the architecture uses Class Tokens. These Class Tokens are randomly initialized tokens that are prepended to the beginning of your input sequence. What is the reason for this Class Token and what does it do? Note that the Class Token is ra..

  Read More
• Transformer Study Materials

  Data/Machine learning 2021. 12. 4. 16:24

  1. https://youtu.be/z1xs9jdZnuY 2. https://youtu.be/4Bdc55j80l8 About Positional Encoding https://kazemnejad.com/blog/transformer_architecture_positional_encoding/ Transformer Architecture: The Positional Encoding - Amirhossein Kazemnejad's Blog Transformer architecture was introduced as a novel pure attention-only sequence-to-sequence architecture by Vaswani et al. Its ability for parallelizabl..

  Read More
• Machine learning - Introduction to Gaussian processes

  Data/Machine learning 2021. 9. 15. 14:44

  https://www.youtube.com/watch?v=4vGiHC35j9s&ab_channel=NandodeFreitas

  Read More
• Evidence Lower Bound (ELBO)

  Data/Machine learning 2021. 8. 24. 18:21

  In the variational auto encoder (VAE) (my blog posting link), a probability of observing $z$ given $x$ can be presented as follows using Bayes' theorm: $$p(z | x) = \frac{ p(x|z) p(z) }{ p(x) }$$ where $p(z|x)$ is a posterior. However, this posterior is difficult to compute due to its marginal likelihood $p(x)$. $p(x)$ can be expanded as below using the law of total probability: $$ p(x) = \int..

  Read More