evilsocket / evilsocket/pwnagotchi

[BUG] Counter problem affects on AI training

Open
#892 2 comments 0 reactions 0 assignees View on GitHub
bug
Dominant language
Python
Stars
9.2k
Forks
1.2k
PR merge metrics
No merged PRs in 30d

Description

Some successful "handshakes" do not count.
I created a new pwnagotchi, and I want to train him from scratch. But it **seems** thу learning is not be accurate.
AI does not count all successful handshakes. And this is bad for training.
For example, 7 handshakes were caught during the session (In it, there was 1 epoch), but according to the logs there are only 2. And the counter also shows 2 and recorded only 2 in the epoch.

**Screenshots**
Both screenshots are the first session, and all counters were 0.
![1](https://user-images.githubusercontent.com/25121341/85566467-bdfb2c80-b638-11ea-926d-1d62cae420e3.png)

![2](https://user-images.githubusercontent.com/25121341/85566470-bf2c5980-b638-11ea-9575-5c6c50d734a0.png)
_7 was captured, and AI recorded only 2 to the epoch._

**To Reproduce**
Just create new pwnagotchi with a screen and follow the counter. And also in the log.
Or you can clear the handshake folder so that the total counter is 0. And start it.

**Example**

![gifexample](https://user-images.githubusercontent.com/25121341/85578813-09b2d380-b643-11ea-9b4a-c9e0a5a6f8bc.gif)

**Version with problem**
Pwnagotchi version 1.5.3
Image: pwnagotchi-raspbian-lite-v1.5.3.zip
Raspberry Pi Zero W
The problem is both on the built-in wifi and on the wifi adapter.

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.