SunOne

sunxds_0.7.1 (YOLOv11)

Added more classes of heads, players of dead bodies, and fixed incorrectly marked dataset classes.

Export model in Helper with image size 640.

sunxds_0.7.0_up (YOLOv11)

Approximately 3,000 player classes have been added, and the Grand Theft Auto: San Andreas and Grand Theft Auto: 5 datasets have also been updated.

Please note that the model was trained with an image size of 320 for speed, as well as with a small number of layers.

Export model in Helper with image size 320.

sunxds_0.6.9 (YOLOv11)

Added 5000 classes (players, heads).

sunxds_0.6.8 (YOLOv11)

Updated Deadlock, The Finals and other games.

sunxds_0.6.7_up (YOLOv11) (ultra-potato)

The model was trained with a small number of layers and with a small image size of 320, so it's stupid and fast.

Export model in Helper with image size 320.

sunxds_0.6.6 (YOLOv11)

The model was trained with the old dataset, but already in the YOLOv11 version. To use the Python version, make sure that you have ultralytics 8.30.0 installed (Run helper to auto update or type `pip install -U ultralytics`).

Support for the CPP version will be released later in version 1.4.

Export model in Helper with image size 640.  

sunxds_0.6.5 (YOLOv10)

Updating the dataset as well as a large number of fixes. The model is trained with a large number of layers compared to the previous one in order to more accurately identify objects.

Export model in Helper with image size 640.

sunxds_0.6.4 (YOLOv10)

Updating the FPS dataset. This is a test version of a new model training method. so if there are any glitches, then write in private messages.

Export model in Helper with image size 640.  

sunxds_0.6.3 (YOLOv10)

Updating the dataset of third-person games, as well as games that have not yet been released (This is a big secret).

I also took the basis of the configuration of the previous AI model.

Export model in Helper with image size 640.