# \#yolo

**URL:** https://forum.beginner.center/tag/yolo/191.md

[Latest](https://forum.beginner.center/latest.md) · [Categories](https://forum.beginner.center/categories.md) · [Tags](https://forum.beginner.center/tags.md)

---

## [Commemorating the first furnace of elixir crafted](https://forum.beginner.center/t/topic/485)

<div class="topic-metadata">

**Author:** [@doggie](https://forum.beginner.center/u/doggie)\
**Replies:** 3\
**Last updated:** [November 4, 2023, 9:01am UTC](https://forum.beginner.center/t/topic/485 "2023-11-04T09:01:24Z")

</div>

Used a lazy tutorial from Bilibili, succeeded in training for the first time, made the simplest A/B binary classification, looks pretty good, both training and test set recognition rates are quite high, 95%. But when testing new data, recognition of A is almost entirely correct, while recognition of B is almost entirely wrong (all recognized as A). After discussing with classmates, it might be due to too much background interference, planning to switch to object detection to handle it…
