Skip to content

Commit

Permalink
map
Browse files Browse the repository at this point in the history
  • Loading branch information
FelixFu520 committed Sep 4, 2021
1 parent 578ef36 commit cb25d33
Show file tree
Hide file tree
Showing 3 changed files with 14 additions and 6 deletions.
17 changes: 11 additions & 6 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -517,15 +517,18 @@
## 1⃣️ 分类/Classification

- #### 🐾 [classification](https://github.com/FelixFu520/classification)
- 🐾 [classification](https://github.com/FelixFu520/classification)
- 🐾 [classification](https://github.com/FelixFu520/LCD-luster)

## 2⃣️ 语义分割/Segmentation

- #### 🐾 [segmentation](https://github.com/FelixFu520/segmentation)
- 🐾 [segmentation](https://github.com/FelixFu520/segmentation)

## 3⃣️ 目标检测/Detection

- #### 🐾 [YoloX](https://github.com/FelixFu520/YOLOX)
- 🐾 [Yolo v1](https://github.com/TD-4/yolov1) | [notes](train/detection/yolov1.md)
- 🐾 [Yolo v4](https://github.com/FelixFu520/YOLOX)
- 🐾 [YoloX](https://github.com/FelixFu520/YOLOX)

## 4⃣️ 小样本/Few Shot Learning

Expand All @@ -537,9 +540,11 @@

## 5⃣️ 异常检测/Anomaly Detection

- [PaDiM Anomaly Detection](https://github.com/FelixFu520/PaDiM-Anomaly-Detection-Localization-master) | [notes](train/anomalydetection/padim.md)
- [SPADE](https://github.com/TD-4/SPADE-pytorch)
- [PaDim & PatchCore ](https://github.com/TD-4/anodet)
- 🐾 [PaDiM Anomaly Detection](https://github.com/FelixFu520/PaDiM-Anomaly-Detection-Localization-master) | [notes](train/anomalydetection/padim.md)
- 🐾 [SPADE](https://github.com/TD-4/SPADE-pytorch)
- 🐾 [PaDim & PatchCore ](https://github.com/TD-4/anodet)

## 5⃣️ 图像生成/Image Generative

## 5⃣️ 部署

Expand Down
Binary file added train/assessment/imgs/123.jpg
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
3 changes: 3 additions & 0 deletions train/assessment/map.md
Original file line number Diff line number Diff line change
Expand Up @@ -10,6 +10,8 @@

---

![](imgs/123.jpg)

准确率、精确率(查准率)、召回率(查全率)、F1值、ROC曲线的AUC值、mAP都可以作为评价一个机器学习模型好坏的指标(evaluation metrics),而这些评价指标直接或间接都与混淆矩阵有关,前四者可以从混淆矩阵中直接计算得到,ROC、mAP和混淆矩阵联系密切,所以在了解这些评价指标之前,先知道什么是混淆矩阵很有必要,也方便记忆。

## 0. 混淆矩阵
Expand Down Expand Up @@ -384,3 +386,4 @@ AP衡量的是学出来的模型在给定类别上的好坏,而mAP衡量的是
> - TP: IoU>0.5的检测框数量(同一Ground Truth只计算一次)
> - FP: IoU<=0.5的检测框,或者是检测到同一个GT的多余检测框的数量
> - FN: 没有检测到的GT的数量

0 comments on commit cb25d33

Please sign in to comment.