# Meeting0720 ## 工作分配 ### 鄭 1. 前端figma [需求描述](https://hackmd.io/dIZxGKwGRhq_6hKrASLAbw#%E6%9E%B6%E6%A7%8B%E7%B8%BD%E8%A6%BD) [figma](https://www.figma.com/design/FgZq5YAy7cj7r8uyYHmiM6/%E9%AB%98%E9%80%9F%E5%85%AC%E8%B7%AF%E4%BA%A4%E9%80%9A%E5%A3%85%E5%A1%9E%E9%A0%90%E6%B8%AC%E6%A8%A1%E5%9E%8B-%E5%89%8D%E7%AB%AF?node-id=0-1&t=x2zbGFiet9JodSvx-1) 2. 找出[論文](https://1drv.ms/b/c/002d06ef873e3a00/ESGwviCJ6l1Gv19nuhWulH4BfbEPR5WS5BvRARVZLpUVeA?e=Zhze7r)預測哪些路段 3. 用程式載`2024/1/1 00:30 的VD靜態資料`,轉成csv。 [資料連結](https://tisvcloud.freeway.gov.tw/history/motc20/VD/) 7/22 ### 傅 1. 修改畫圖(跟表格)程式,以更符合之前說的要求。[新資料](https://github.com/project-team-2025/LSTM/blob/main/data/merged_37.csv) 2. 手動比對檢查m06 1月(檢查3個row,各自截圖官網跟資料,全部應該共6張) 3. 幫lstm的code裡面寫一個function,傳入資料,return 預測結果,傳入的資料只傳入必要的部分。用於之後部署時預測即時資料。https://github.com/project-team-2025/LSTM/blob/main/src/model_pytorch.py ### 王 1. 使用`mae+ratio*custom_mae`當作loss進行backforward。(LSTM) ### 森 1. 讓pytorch在訓練跟驗證過程中把x、actual_y、predict_y都輸出到一個txt,用於驗證程式是否正確。[LSTM](https://github.com/project-team-2025/LSTM/blob/main/src/model_pytorch.py) [參考](https://github.com/project-team-2025/LSTM/blob/bc16a0a9e9d46949b4d78293954e5891f8ac213a/src/cnn_check.py) ### 毅
×
Sign in
Email
Password
Forgot password
or
By clicking below, you agree to our
terms of service
.
Sign in via Facebook
Sign in via Twitter
Sign in via GitHub
Sign in via Dropbox
Sign in with Wallet
Wallet (
)
Connect another wallet
New to HackMD?
Sign up