Install GroundingDINO、GroundingSAM Github: https://github.com/IDEA-Research/Grounded-Segment-Anything#installation change to install like this(every MUST same): cuda:11.6 (this need go to Nvidia web, and install finish, you can at /usr/local/cuda-11.6 to find it ! ) Python: 3.9.16 Pytorch: 1.13.1 torch vision: 0.14.1

暫停更新 2022/06/24 LeetCode-Minimun Size Subarray Sum(C++版)-定洲:影片 | 筆記 2022/05/20 LeetCode-Delete Node-Linked List統整(C++版)-定洲:影片 | 筆記 2022/05/20 LeetCode-Reverse Integer(C++版)-定洲:影片 | 筆記 2022/05/06 LeetCode-Add Two Numbers(C++版)-定洲:影片 | 筆記 2022/05/06 LeetCode-Container With Most Water(C++版)-衡安:影片 | 筆記 Supplement 補充資料

更新時間2022/08/18 LFFD(高速人臉辨識)： LFFD(原作者-MXNet版/NCNN版)-網址 LFFD(Pytorch版)-網址 LFD (一年多前提出的新版本/支援多類別)-網址 高速YOLO：

更新日期:2022/6/42 訓練: 使用darknet cfg檔請參照 https://github.com/LiaoZihZrong/yolo-fastest/blob/main/yolo-fast.cfg 轉模型 轉出我們的檔案 git clone https://github.com/LiaoZihZrong/weight-Yolo-fastest-model-h5

全名You Only Look Once V1: You Only Look Once: Unified, Real-Time Object Detection V2: YOLO9000: Better, Faster, Stronger V3: YOLOv3: An Incremental Improvement

整理:定洲 類型:for loop 題目大要: Given a signed 32-bit integer x, return x with its digits reversed. If reversing x causes the value to go outside the signed 32-bit integer range [-2^31, 2^31 - 1], then return 0. Assume the environment does not allow you to store 64-bit integers (signed or unsigned). 給定一個有號32bit整數x，回傳反轉後的數值。 如果反轉後的x超出有號32bit整數範圍[$-2^{31}$, $2^{31} - 1$]，則回傳 0。

主題:237. Delete Node in a Linked List 整理:定洲 類型:Linked List 題目大要: Write a function to delete a node in a singly-linked list. You will not be given access to the head of the list, instead you will be given access to the node to be deleted directly. It is guaranteed that the node to be deleted is not a tail node in the list.

處理時間:2022/04/23 forked from xialeiliu/Awesome-Incremental-Learning Incremental Learning / Lifelong learning Survey Replay in Deep Learning: Current Approaches and Missing Biological Elements (Neural Computation 2021) [paper] Online Continual Learning in Image Classification: An Empirical Survey (Neurocomputing 2021) [paper] [code] Continual Lifelong Learning in Natural Language Processing: A Survey (COLING 2020) [paper] Class-incremental learning: survey and performance evaluation (arXiv 2020) [paper] [code]

全名You Only Look Once V4 YOLOv4: Optimal Speed and Accuracy of Object Detection PP-YOLO PP-YOLO: An Effective and Efficient Implementation of Object Detector 整理的ppt簡報下載:網址 CBM： Yolov4網路結構中的最小組件，由Conv+Bn+Mish啟動函數三者組成。

整理:定洲 類型:Linked List 題目大要: 給定2個non-empty(不為空)的Linked Lists，裡面的元素皆為正整數，以Reverse形式儲存，每個節點儲存一個個位數，相加之後以Linked List形式回傳結果。 Example 1: Input: l1 = [2,4,3], l2 = [5,6,4] Output: [7,0,8] Explanation: 342 + 465 = 807.

處理時間:2022/03/10 forked from xialeiliu/Awesome-Incremental-Learning 2014 重要 [R-CNN] Rich feature hierarchies for accurate object detection and semantic segmentation | [CVPR' 14] |[pdf] [official code - caffe] 重要 [OverFeat] OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks | [ICLR' 14] |[pdf] [official code - torch] 重要 [SPPnet] Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition | |[pdf]

處理時間:2022/04/23 forked from xialeiliu/Awesome-Incremental-Learning 新增幾篇醫學相關與2022之論文 Incremental Learning / Lifelong learning Survey Replay in Deep Learning: Current Approaches and Missing Biological Elements (Neural Computation 2021) [paper] Online Continual Learning in Image Classification: An Empirical Survey (Neurocomputing 2021) [paper] [code] Continual Lifelong Learning in Natural Language Processing: A Survey (COLING 2020) [paper]

以下是個人的建議以及實際目前業界需求 目錄 [TOC] 1.學習路程 :::success 從進入此領域到畢業後就業到後續精進 :::

1.請先登入學校VPN ( 搜尋中山VPN教學 ) https://lis.nsysu.edu.tw/p/412-1001-7113.php 2.使用圖書館系統搜尋 IEEE ( IP要是 140開頭才是有連上線 ) https://service.lis.nsysu.edu.tw/database/dbs_list?lang=zh&text= 3.登入驗證( ip要是 140開頭 才是有連上學校 VPN ) 就會有一個 ieee xplore lib... 點選後會轉至IEEE

更新日期:2022/4/12 目錄 [toc] 訓練: 使用darknet cfg檔請參照 https://github.com/LiaoZihZrong/yolo-fastest/blob/main/yolo-fast.cfg 轉出我們的檔案

SSD: Single Shot MultiBox Detector 整理的ppt簡報下載:網址 SSD 論文:SSD: Single Shot MultiBox Detector 位置:https://arxiv.org/abs/1512.02325 2015年提出，引用20382次 SSD貢獻總結如下：

目錄 [TOC] 前言: 數學基礎 線性代數、梯度下降跟機率統計 機器學習基礎

目錄 [TOC] 請注意版本 Python 3.6 or higher PyTorch 1.4 CUDA 9.2 or higher Visual Studio 2013-2019 1. Create a conda environment

更改 darknet 照片的框線文字粗細 darknet YOLO test jpg bbox more bold 1.open darknet/src/image.c 2.find line 457 3.==change code== image label = get_label_v3(alphabet, labelstr, (im.h*.021)); ---> (im.h*.021) ---> (im.h*.021) 4.find line 376 5.==change code== int width = im.h * .009;

clang-format 工具和一致的程式撰寫風格 使用一致的 programming style 很重要，我們可透過 clang-format 這個工具來調整作業程式要求的風格，使用方式如下: $ clang-format -i *.[ch] 課程要求的 C 程式撰寫風格簡述: 使用 ==4 個空白字元==來進行縮排，不用 Tab; 為何不比照 Linux 核心都用 tab 呢？