# 10/31/24 Meeting Notes #11
# This Week's Progress 這週進度
## Classification Model #1
- No progress `無進度`
## Classification Model #2
- Tested model accuracy using *FashionStyle14* dataset. This dataset has already been categorized into 13 categories
`目前正在使用 FashionStyle14 數據集測試模型準確度。此數據集已分為 13 個類別。`
- FashionStyle14 categories

-- Conservative
-- Dressy
-- Ethnic 
-- Fairy 
-- Feminine
-- Gal 
-- Casual 
-- Lolita 
-- Mode
-- Natural 
-- Retro 
-- Rock
-- Street

- After training the model with 50 epochs, the results are `模型訓練50回合結果`:
- Training accuracy `訓練準確率`: 91.91%
- Test accuracy`測試準確率`: 53.93%


## Generative Model
- Finished installing psbody in docker
`已在Docker完成下載psbody`
- Missing PyOpenGL source for MeshViewer
`缺少 MeshViewer 的 PyOpenGL 原始碼`

- Learning how to properly use the container in docker
`正在學習如何正確使用 Docker 中的 container`

## Datasets
- Will use *Fashion550k* dataset for classification model #2 training instead of dataset from *Dress Code: High-Resolution Multi-Category Virtual Try-On (ECCV 2022)*
`分類模型#2的訓練改換使用 Fashion550k 數據集， 不再用 Dress Code: High-Resolution Multi-Category Virtual Try-On (ECCV 2022) 的數據集`
- Currently still dividing *Fashion550k* dataset into its respective categories (Casual, Business, Semi-Formal, Sporty, and Vintage) (Total 405,500 photos)
`目前正在將 Fashion550k 數據集分類到其各自的類別中（休閒、正式、半正式、運動風和復古），共計 405,500 張照片。`
# To Do 需做
- Divide as many *Fashion550k* dataset as possible.
`把 *Fashion550k* 數據集分完`
- If finish dividing *Fashion550k* dataset, will do VISELLE dataset, which will be used for classification model #2 testing. This dataset will need to be divided into its categories (shirt, t-shirt, pants, etc.) then merge with model.
`若 *Fashion550k* 數據集分完，接下來會分 *VISELLE* 數據集。這個數據會用在分類模型 #2 的測試。此數據集需按類別（如襯衫、T 恤、褲子等）分類，並與模型合併。`
# Next Weeks's Meeting 下週會議

---
Previous: [10/24/24 Meeting Notes #10](https://hackmd.io/@emps-113up/meeting10)
Next: [11/07/24 Meeting Notes #12](https://hackmd.io/@emps-113up/meeting12)
Full Content List [here](https://hackmd.io/@emps-113up/full-list)