# 01/16/25 Meeting Notes #18
# This Week's Progress 這週進度
# Classification Model 1 (garment classification)
* Tried classifying VISUELLE dataset using model1.
* Problems: Most of the clothing items were classified wrongly. Still need to improve the model.

# Classification model 2 (style classification)
Tried to use the dataset to train classification model 2.

# Dataset
* Model1 dataset
* 3761 files in 01Dress
* 4017 files in 02Shirt
* 4377 files in 03Tshirt
* ***1492*** files in 04Hoodie -> 3043
* ***2053*** files in 05Sweater -> 3081
* **1396** files in 06PoloShirt -> 2049
* 4899 files in 07Pants
* 3849 files in 08Shorts
* 3022 files in 09Skirt
* 3871 files in 10Flats
* ***265*** files in 11Heels
* ***1703*** files in 12Shoes
* 3816 files in 13Sneakers

* Model2 dataset
* 220 Bohemian
* 1588 Casual
* 419 beach
* 1169 casual
* 1528 Formal -> 1527 Formal
* 1510 Semi-Formal -> 1703 Semi-Formal
* 308 buisness casual -> 349 buisness casual
* 917 classic -> 978 classic
* 285 preppy -> 376 preppy
* 349 Sporty
* 380 Streetwear

# Generative Model
Lost all data, laptop drive crashed.

# NSTC Proposal
Finished writing the NSTC Proposal.
https://docs.google.com/document/d/1BVFdVj92j1GGW5aQRl5k04Tw_gCovDSvi79kTu5ItPA/edit?usp=sharing

# To Do 需做
* Dataset for model1: 06PoloShirt, 11Heels, 12Shoes
* Add more variety to model1 dataset so that the model can be more adaptive
* try testing model 1 with a different style of clothes
* model 2 accuracy is low, need at least 80%. Dataset count for boho, sport and streetwear is also low, need to increase
# Next Weeks's Meeting 下週會議

---
Previous:
Next:
Full Content List [here](https://hackmd.io/@emps-113up/full-list)