1140805 meeting

發佈於 2025-08-05 更新於 2025-08-05 約 674 字預計閱讀 2 分鐘次閱讀

前言

以下使用 dart 模組的 TSMixerModel 進行填補未來 10 天。

實驗

gallery_made_with_nanogallery_exp2

Method	Value
MSPE	4.502955
RMSPE	5.508323
MAPE	4.502955
MSPE%	0.015959
RMSPE%	0.328823
MAPE%	0.015959

結論

此實驗結果都為整數，需再進一步探討。

執行結果參考

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
>>> inference
array([[[276, 271, 269, ..., 288, 287, 288],
        [276, 271, 269, ..., 288, 287, 288],
        [276, 272, 269, ..., 288, 288, 288],
        ...,
        [275, 271, 268, ..., 288, 288, 288],
        [275, 270, 268, ..., 287, 287, 287],
        [275, 270, 268, ..., 287, 287, 287]],

       [[276, 271, 269, ..., 288, 287, 288],
        [276, 271, 269, ..., 288, 287, 288],
        [276, 272, 269, ..., 288, 288, 288],
        ...,
        [275, 271, 268, ..., 288, 288, 288],
        [275, 270, 268, ..., 287, 287, 287],
        [275, 270, 268, ..., 287, 287, 287]],

       [[276, 271, 269, ..., 288, 287, 288],
        [276, 271, 269, ..., 288, 287, 288],
        [276, 272, 269, ..., 288, 288, 288],
        ...,
        [275, 271, 268, ..., 288, 288, 288],
        [275, 270, 268, ..., 287, 287, 287],
        [275, 270, 268, ..., 287, 287, 287]],

       ...,

       [[276, 271, 269, ..., 288, 287, 288],
        [276, 271, 269, ..., 288, 287, 288],
        [276, 272, 269, ..., 288, 288, 288],
        ...,
        [275, 271, 268, ..., 288, 288, 288],
        [275, 270, 268, ..., 287, 287, 287],
        [275, 270, 268, ..., 287, 287, 287]],

       [[276, 271, 269, ..., 288, 287, 288],
        [276, 271, 269, ..., 288, 287, 288],
        [276, 272, 269, ..., 288, 288, 288],
        ...,
        [275, 271, 268, ..., 288, 288, 288],
        [275, 270, 268, ..., 287, 287, 287],
        [275, 270, 268, ..., 287, 287, 287]],

       [[276, 271, 269, ..., 288, 287, 288],
        [276, 271, 269, ..., 288, 287, 288],
        [276, 272, 269, ..., 288, 288, 288],
        ...,
        [275, 271, 268, ..., 288, 288, 288],
        [275, 270, 268, ..., 287, 287, 287],
        [275, 270, 268, ..., 287, 287, 287]]], shape=(24, 10, 1850))


Inference complete! Time taken: 4:59:24.693854

結語

https://raw.githubusercontent.com/Josh-test-lab/website-assets-repository/refs/heads/main/posts/1140805%20meeting/To%20be%20continued.jpg — To be continued!

運行環境

本機作業系統：Windows 11 24H2
- 程式語言：Python 3.12.9
計算平臺：財團法人國家實驗研究院國家高速網路與計算中心臺灣 AI 雲
- 作業系統：Ubuntu
- Miniconda
- GPU：NVIDIA Tesla V100 32GB GPU
- CUDA 12.8 driver
- 程式語言：Python 3.10.16 for Linux

延伸學習

我測試此項目的 Github 儲存庫。

參考資料

Juan Lopez Alcaraz 、 Nils Strodthoff（2022）。Diffusion-based time series imputation and forecasting with structured state space models。Transactions on Machine Learning Research。參考自 https://openreview.net/forum?id=hHiIbk7ApW
SSSD（2022）。GitHub。參考自 https://github.com/AI4HealthUOL/SSSD
SSSD_CP（2024）。GitHub。參考自 https://github.com/egpivo/SSSD_CP
近兩年小時值查詢（無日期）。環境部 - 空氣品質監測網。參考自 https://airtw.moenv.gov.tw/CHT/Query/InsValue.aspx

目錄

1140805 meeting

前言

實驗

結論

結語

運行環境

延伸學習

參考資料

相關文章