KobeYourself2324
  • NEW!
    NEW!  Connect Ideas Across Notes
    Save time and share insights. With Paragraph Citation, you can quote others’ work with source info built in. If someone cites your note, you’ll see a card showing where it’s used—bringing notes closer together.
    Got it
      • Create new note
      • Create a note from template
        • Sharing URL Link copied
        • /edit
        • View mode
          • Edit mode
          • View mode
          • Book mode
          • Slide mode
          Edit mode View mode Book mode Slide mode
        • Customize slides
        • Note Permission
        • Read
          • Only me
          • Signed-in users
          • Everyone
          Only me Signed-in users Everyone
        • Write
          • Only me
          • Signed-in users
          • Everyone
          Only me Signed-in users Everyone
        • Engagement control Commenting, Suggest edit, Emoji Reply
      • Invite by email
        Invitee

        This note has no invitees

      • Publish Note

        Share your work with the world Congratulations! 🎉 Your note is out in the world Publish Note No publishing access yet

        Your note will be visible on your profile and discoverable by anyone.
        Your note is now live.
        This note is visible on your profile and discoverable online.
        Everyone on the web can find and read all notes of this public team.

        Your account was recently created. Publishing will be available soon, allowing you to share notes on your public page and in search results.

        Your team account was recently created. Publishing will be available soon, allowing you to share notes on your public page and in search results.

        Explore these features while you wait
        Complete general settings
        Bookmark and like published notes
        Write a few more notes
        Complete general settings
        Write a few more notes
        See published notes
        Unpublish note
        Please check the box to agree to the Community Guidelines.
        View profile
      • Commenting
        Permission
        Disabled Forbidden Owners Signed-in users Everyone
      • Enable
      • Permission
        • Forbidden
        • Owners
        • Signed-in users
        • Everyone
      • Suggest edit
        Permission
        Disabled Forbidden Owners Signed-in users Everyone
      • Enable
      • Permission
        • Forbidden
        • Owners
        • Signed-in users
      • Emoji Reply
      • Enable
      • Versions and GitHub Sync
      • Note settings
      • Note Insights New
      • Engagement control
      • Make a copy
      • Transfer ownership
      • Delete this note
      • Save as template
      • Insert from template
      • Import from
        • Dropbox
        • Google Drive
        • Gist
        • Clipboard
      • Export to
        • Dropbox
        • Google Drive
        • Gist
      • Download
        • Markdown
        • HTML
        • Raw HTML
    Menu Note settings Note Insights Versions and GitHub Sync Sharing URL Create Help
    Create Create new note Create a note from template
    Menu
    Options
    Engagement control Make a copy Transfer ownership Delete this note
    Import from
    Dropbox Google Drive Gist Clipboard
    Export to
    Dropbox Google Drive Gist
    Download
    Markdown HTML Raw HTML
    Back
    Sharing URL Link copied
    /edit
    View mode
    • Edit mode
    • View mode
    • Book mode
    • Slide mode
    Edit mode View mode Book mode Slide mode
    Customize slides
    Note Permission
    Read
    Only me
    • Only me
    • Signed-in users
    • Everyone
    Only me Signed-in users Everyone
    Write
    Only me
    • Only me
    • Signed-in users
    • Everyone
    Only me Signed-in users Everyone
    Engagement control Commenting, Suggest edit, Emoji Reply
  • Invite by email
    Invitee

    This note has no invitees

  • Publish Note

    Share your work with the world Congratulations! 🎉 Your note is out in the world Publish Note No publishing access yet

    Your note will be visible on your profile and discoverable by anyone.
    Your note is now live.
    This note is visible on your profile and discoverable online.
    Everyone on the web can find and read all notes of this public team.

    Your account was recently created. Publishing will be available soon, allowing you to share notes on your public page and in search results.

    Your team account was recently created. Publishing will be available soon, allowing you to share notes on your public page and in search results.

    Explore these features while you wait
    Complete general settings
    Bookmark and like published notes
    Write a few more notes
    Complete general settings
    Write a few more notes
    See published notes
    Unpublish note
    Please check the box to agree to the Community Guidelines.
    View profile
    Engagement control
    Commenting
    Permission
    Disabled Forbidden Owners Signed-in users Everyone
    Enable
    Permission
    • Forbidden
    • Owners
    • Signed-in users
    • Everyone
    Suggest edit
    Permission
    Disabled Forbidden Owners Signed-in users Everyone
    Enable
    Permission
    • Forbidden
    • Owners
    • Signed-in users
    Emoji Reply
    Enable
    Import from Dropbox Google Drive Gist Clipboard
       Owned this note    Owned this note      
    Published Linked with GitHub
    • Any changes
      Be notified of any changes
    • Mention me
      Be notified of mention me
    • Unsubscribe
    --- tags: 實習日記 --- # 實習日記10/24~10/28 [TOC] ## 10/24 1. 開會 2. 柏林建議可以找這本書來看,有更多 ai 的內容可以玩 ![](https://i.imgur.com/d9LO67z.png) - 工研院的轉正和 ocr 模型就是用 pytorch 寫的 3. 研究工學院車牌辨識報告 ``` Yolo 的優點 ``` - 免去RegionProposal步驟,直接用CNN進行Regression,速度很快 ``` Yolo 的缺點 ``` - 預測物件的總數受限於Grid的總數 - 能夠辨識物件的種類很少 - 面積占比小的物件因為IOU小,準確率很低 - 之前的專題曾經想使用yolo來抓高爾夫球,正是因為小物件準確度很低,所以無法成功抓取到。 - 群體小物件,例如鳥群,只能檢測其中一個,準確率很低 - 影像辨識的種類 - 標記麻煩程度:實例分割>語意分割>物件偵測>物件分類 - 資訊量:實例分割>語意分割>物件偵測>物件分類 4. 標籤什麼照片要等公司處理好影像,我才有辦法做標籤 5. 看第3點的書 ## 10/25 1. 研究利用via標籤 ![](https://i.imgur.com/dywsa92.png) ### 練習成果 ``` 第一張 ``` ![](https://i.imgur.com/CkxU1aZ.jpg) - 用正向長方型標記,並加上車牌號碼的訊息。 ![](https://i.imgur.com/YI1vcnV.jpg) - 左上顯示車牌號碼資訊 ![](https://i.imgur.com/IGca8Lm.jpg) - 輸出json檔案 ``` 第二張 ``` ![](https://i.imgur.com/u31FByz.jpg) - 使用任意四邊形標記 ![](https://i.imgur.com/d1P9Nfs.png) - 車牌標記成功 ![](https://i.imgur.com/Hky2NaH.png) - 輸出json檔案 ![](https://i.imgur.com/wsSY3UT.jpg) ![](https://i.imgur.com/U0lc1xD.jpg) 2. 看pytorch的書和其他pytorch相關資料 3. 參考資料 - https://www.youtube.com/watch?v=-3WVSxNLk_k - https://medium.com/pyladies-taiwan/%E6%B7%B1%E5%BA%A6%E5%AD%B8%E7%BF%92%E6%96%B0%E6%89%8B%E6%9D%91-pytorch%E5%85%A5%E9%96%80-511df3c1c025 ## 10/26 1. 繼續了解pytorch a. Deep learning framework b. 以Python優先的深度學習框架(對Python的支援性高) c. 人工智慧(Artificial Intelligence):電腦與機器模仿人類思考進而模擬人類的能力與行為(較大的集合) d. 機器學習(Machine Learning):從資料中學習分析獲得規律,並利用獲得的規律對未知的資料進行預測 (為人工智慧的子集合) e. 深度學習(Deep Learning):以神經網路為架構,對資料進行特徵學習的演算法(為機器學習的子集合) ![](https://i.imgur.com/e1CIrsS.png) 2. 看programing pytorch for deep learning 3. 練習基本pytorh的程式語法 4. 參考資料: - https://www.readfog.com/a/1633738521353228288 - https://medium.com/ching-i/pytorch-%E5%9F%BA%E6%9C%AC%E4%BB%8B%E7%B4%B9%E8%88%87%E6%95%99%E5%AD%B8-ac0e1ebfd7ec - https://ithelp.ithome.com.tw/articles/10244106 - https://www.youtube.com/results?search_query=pythorch ## 10/27 1. 讀工研院course02 ### 簡單重點整理 - 機器學習與傳統規則模型十分不同 - 傳統規則模型: 人工尋找資料規律並訂定規則,規則須明確闡明,使得機器足以根據規則下判斷 - 機器學習模型: 資料驅動規則訂定,準備大量資料並使用數學模型歸納規則,規則通常難以明確闡明 - 機器學習模型永遠使用最簡單的方式完成我們準備的考卷 - 張量 tendsor,ndarry - Scalar:0-dim tensor - Vector:1-dim tensor - Matrix:2-dim tensor - Tensor:3-dim tensor - 從立體空間到平面影像 - 物體被拍攝後會從三維影像變成二維影像,有很多轉換形式可以解決,有些複雜 - 透視投影 - homographic matrix - 平移 - 旋轉 - 等比縮放 - 非等比縮放 - 仿射轉換 - 透視轉換 - SWAG vs. Yolo - Yolo也可以框車號,但每個數字英文都要標,並註明他的類別,十分麻煩 2. 盡力了解course2裡面的數學和公式 3. 看programing pytorch for deep learning ## 10/28 1. 研究工研院course3 2. 看programing pytorch for deep learning ``` 重點整理 ``` ### Overfitting: - 訓練資料當中學習到錯誤的規則,這些規則在訓練資 料內或許成立,但實際考慮訓練資料以外的情況則否 - 模型的 能力越來越強時,訓練資料上的準確率越來越高,但是在測試資料的 準確率卻越來越低 ![](https://i.imgur.com/f3xDuDy.png) ### Solution of overfitting 1. 正規化: 在 cost function 的後面增加一個懲罰項 ![](https://i.imgur.com/Nw8VsOB.png) 2. Early Stop: 在每一個 epoch 結束時計算驗證集(validation data)的準確率,當準確率不再提高就停止訓練 ![](https://i.imgur.com/iZEoUrD.png) 3. Augmentation:考慮洽當的資料增量,可以以少量資料模擬多量資料的效果,當資料 越來越多時,縱使模型能力很強也不容易過擬合 ![](https://i.imgur.com/HKvGKUy.png) 4. Cutout:強迫模型多學幾種額外的特徵 ![](https://i.imgur.com/eMy9Gdz.png) 使用類似high availablity的觀念,強迫模型多雪幾種特徵 5. Random Padding:YOLO有時候會框歪物件,導致後面的模型辨識效果不佳,不如我們一開始就假設YOLO會框歪,將此效果直接加入YOLO後面 模型的資料增量 6. Random Perspective:轉動車牌的角度,增加不同的特徵 ![](https://i.imgur.com/fh9ad0B.png) ### Hyper-parameter Tuning Tips 1. 調整增量程度的參數專有名詞為超參數(hyper-parameters),為了避 免與模型內的參數(parameters)混淆,模型以外的參數都稱超參數 2. 資料增量手法眾多(上面所提到的6個方法),會留下許多的超參數,這些超參數因為不可微分 ,無法使用先前的梯度下降(gradient descent method,之前訓練模 型內參數的方式)進行最佳化 3. 超參數可分為重要與不重要,代表其數值對於模型準確率影響的程度 ,不重要的參數不管怎麼設對準確率都沒什麼影響,重要的參數要設 得正確才能達到高準確率 ### Solution of Hyper-parameter 1. 網格搜尋 (grid search):使用網格搜尋每次只搜尋一個維度,剩下N-1個維度皆為一樣的數值, 若搜尋的維度為不重要的,那就浪費了計算資源只是得到類似效果的 模型 2. 網格搜尋 (grid search):使用隨機搜尋可以在每個維度都嘗試不一樣的數值,探索空間的程度 更高 3. 貝式最佳化 (Bayesian optimization):貝式最佳化基於統計學的貝氏理論,特 色為一邊蒐集資訊一邊做決策。在現有 最佳超參數附近,加上考量不確定性來 決定下一輪該嘗試什麼![](https://i.imgur.com/M9g53ek.png) 5. 增強式學習 (reinforcement learning):增強式學習直接將最佳化當作機器學習 問題處理,特色為可以直接處理不可微 分的問題,以模型準確率為獎勵,尋找 最佳化超參數以獲取最大獎勵 ### Object Detection 1. 準確率為衡量模型預測與正確答案之間的差異 2. 物件偵測任務對應的標記為每個畫面中有數個方框,每個模型預測的方 框還包含信心值,而正確答案則無 3. 定義物件有沒有被模型偵測到,需大於信心門檻值TYOLO 4. 定義物件有沒有被正確偵測到,通常使用IoU進行衡量,IoU大於門檻 值TIoU才能視為正確偵測 ![](https://i.imgur.com/dAFWFpY.png) ### Recall and Precision 1. ![](https://i.imgur.com/KF1Gmbs.png) 2. 在不同信心門檻值下,模型預測結果也會跟著改變 - 信心門檻值越高,物件容易被門檻值濾掉,模型偵測到的物件越少 Precision很高不會多判,recall很低容易漏判 - 信心門檻值越低,物件信心值容易過門檻,模型偵測到的物件越多 Recall很高不會漏判,precision很低容易多判 ### 同時考量各種信心值且也同時考量recall及precision,該怎麼做? - Mean Average Precision 1. Consider rank position of each relevant doc(-K1,K2,....KR) 2. Compute Precision@K for each K1, K2, … KR 3. Average precision = average of P@K, for each K ![](https://i.imgur.com/wwVo5qJ.png) ![](https://i.imgur.com/uu3VIcj.png) 4. AP也可解釋成precision-recall curve中曲線下的面積 ![](https://i.imgur.com/gI1P35D.png) ![](https://i.imgur.com/EHAfuWq.png) - 參考資料 1. https://ithelp.ithome.com.tw/articles/10203371 2. https://www.youtube.com/watch?v=pM6DJ0ZZee0 3. https://medium.com/lifes-a-struggle/mean-average-precision-map-%E8%A9%95%E4%BC%B0%E7%89%A9%E9%AB%94%E5%81%B5%E6%B8%AC%E6%A8%A1%E5%9E%8B%E5%A5%BD%E5%A3%9E%E7%9A%84%E6%8C%87%E6%A8%99-70a2d2872eb0

    Import from clipboard

    Paste your markdown or webpage here...

    Advanced permission required

    Your current role can only read. Ask the system administrator to acquire write and comment permission.

    This team is disabled

    Sorry, this team is disabled. You can't edit this note.

    This note is locked

    Sorry, only owner can edit this note.

    Reach the limit

    Sorry, you've reached the max length this note can be.
    Please reduce the content or divide it to more notes, thank you!

    Import from Gist

    Import from Snippet

    or

    Export to Snippet

    Are you sure?

    Do you really want to delete this note?
    All users will lose their connection.

    Create a note from template

    Create a note from template

    Oops...
    This template has been removed or transferred.
    Upgrade
    All
    • All
    • Team
    No template.

    Create a template

    Upgrade

    Delete template

    Do you really want to delete this template?
    Turn this template into a regular note and keep its content, versions, and comments.

    This page need refresh

    You have an incompatible client version.
    Refresh to update.
    New version available!
    See releases notes here
    Refresh to enjoy new features.
    Your user state has changed.
    Refresh to load new user state.

    Sign in

    Forgot password
    or
    Sign in via Google Sign in via Facebook Sign in via X(Twitter) Sign in via GitHub Sign in via Dropbox Sign in with Wallet
    Wallet ( )
    Connect another wallet

    New to HackMD? Sign up

    By signing in, you agree to our terms of service.

    Help

    • English
    • 中文
    • Français
    • Deutsch
    • 日本語
    • Español
    • Català
    • Ελληνικά
    • Português
    • italiano
    • Türkçe
    • Русский
    • Nederlands
    • hrvatski jezik
    • język polski
    • Українська
    • हिन्दी
    • svenska
    • Esperanto
    • dansk

    Documents

    Help & Tutorial

    How to use Book mode

    Slide Example

    API Docs

    Edit in VSCode

    Install browser extension

    Contacts

    Feedback

    Discord

    Send us email

    Resources

    Releases

    Pricing

    Blog

    Policy

    Terms

    Privacy

    Cheatsheet

    Syntax Example Reference
    # Header Header 基本排版
    - Unordered List
    • Unordered List
    1. Ordered List
    1. Ordered List
    - [ ] Todo List
    • Todo List
    > Blockquote
    Blockquote
    **Bold font** Bold font
    *Italics font* Italics font
    ~~Strikethrough~~ Strikethrough
    19^th^ 19th
    H~2~O H2O
    ++Inserted text++ Inserted text
    ==Marked text== Marked text
    [link text](https:// "title") Link
    ![image alt](https:// "title") Image
    `Code` Code 在筆記中貼入程式碼
    ```javascript
    var i = 0;
    ```
    var i = 0;
    :smile: :smile: Emoji list
    {%youtube youtube_id %} Externals
    $L^aT_eX$ LaTeX
    :::info
    This is a alert area.
    :::

    This is a alert area.

    Versions and GitHub Sync
    Get Full History Access

    • Edit version name
    • Delete

    revision author avatar     named on  

    More Less

    Note content is identical to the latest version.
    Compare
      Choose a version
      No search result
      Version not found
    Sign in to link this note to GitHub
    Learn more
    This note is not linked with GitHub
     

    Feedback

    Submission failed, please try again

    Thanks for your support.

    On a scale of 0-10, how likely is it that you would recommend HackMD to your friends, family or business associates?

    Please give us some advice and help us improve HackMD.

     

    Thanks for your feedback

    Remove version name

    Do you want to remove this version name and description?

    Transfer ownership

    Transfer to
      Warning: is a public team. If you transfer note to this team, everyone on the web can find and read this note.

        Link with GitHub

        Please authorize HackMD on GitHub
        • Please sign in to GitHub and install the HackMD app on your GitHub repo.
        • HackMD links with GitHub through a GitHub App. You can choose which repo to install our App.
        Learn more  Sign in to GitHub

        Push the note to GitHub Push to GitHub Pull a file from GitHub

          Authorize again
         

        Choose which file to push to

        Select repo
        Refresh Authorize more repos
        Select branch
        Select file
        Select branch
        Choose version(s) to push
        • Save a new version and push
        • Choose from existing versions
        Include title and tags
        Available push count

        Pull from GitHub

         
        File from GitHub
        File from HackMD

        GitHub Link Settings

        File linked

        Linked by
        File path
        Last synced branch
        Available push count

        Danger Zone

        Unlink
        You will no longer receive notification when GitHub file changes after unlink.

        Syncing

        Push failed

        Push successfully