목록분류 전체보기 (65)
i like reading python library documentation. Today is sunday ,so i started finding which documentation i'm gonna read in this week,doing object detection project, i saw some pil functions that handled image in code. In point of handling image, i think there're these libraries ;PIL, torchvision(in torch),OpenCv In this post , I'm gonna introduce PIL ! (it may have prejudice opinions) In near futu..
I'm in the situation I have to implement paper that is about object detection. That makes me so interesting ...but the problem is ..time is short So i started researching about this area. Summary of what i learned Object detection = Localization + classification In doing this , There're two kind of methods 1. 1-stage method (simultaneously) 2. 2-stage method (classification after getting resiona..
www.youtube.com/watch?v=JvQMoLtoFg8 이 교수님을 개인적으로 모르는 사람이다만 영상을 보고 그 철학관에 대하여 굉장히 존경스러웠다. 찾아보게된 계기는 fashionMNIST DATASET 에 대하여 state of the art 를 찍고있는 (≓가장 좋은 performance 를 내는) 것이 무엇인지 보다가 Fine-tuning DARTS 가 1위를 찍고 있었다(2021.4.26 기준) paper를 보려고하는데 저자에 kaist 가 있어서 오 국내발이 구나하고 나라에 대한 자긍심을 3초간 느끼고 3분중 한국인 사람의 이력을 검색하다보니 위와같은 영상이 뜨게 되어 보게되었다. 그리고 이와 같은 철학을 유지하며 그 꾸준함을 인생에서 지속하고 자하는 생각이 들었다. fundam..
adam is usually used in many cases. as a default and general optimizer. But it has a few shortcomings (convergenced fastly in image =>make overfitting table of content 1. 2. each method
0. Goal of this lecture 1. Overall scenario of data analysis (ml/dl) 1.1 what is Machine Learning & Deep learning 1.2 kinds of ml supervised learning / unsupervised / reinforcement 1.3 problems in ml (point of Data) not enough data doesn't have enough Representativeness . bad quality data (error ,noisy , outlier) not related features (Feature engineering can solve this problem) (point of Mod..
data science process Project Scoping(Define Problem) Data Collection EDA Data Preprocessing Feature Engineering Modeling Evaluation Project Delivery / Insights Feature Engineering Feature selection Filtered (using statistical skills) Wrapper (set certain features ,repeat evaluation ,choose better combination) forward backward etc Embedded (depending on model ) Feature Extracction (PCA..) Feature..
2-1 project planning , 2-2 project planning (2)(organization , risk, project plan) things to do in planning phase determine scope (what to do and not to do) small tasks and sequence organization (2-2) cost schedule Quality maintenance risk lists pm(project manager) have to do Analyze validity of project value technical economical about organization determine strategy SWOT Analyze (Strength Weakn..