Skip to main content

Making ML more accessible to botting apps. Solving Complex UI Challenges w/ ML.

Project description

Welcome to ML_Botting_Core!

Solving Complex UI Challenges w/ ML
pip install ml-botting-core

TensorFlow Python

Public Eve Online Models

This package will auto download these models from Here at runtime and maintain a copy on your device. (and auto update)
Sample config Here with all models.

Training

Train your own models Here
Sort your training images into folders (images shape must be the same shape) and target the root for training. The classifer will train those image samples to the name of the folder they are in.

  • Image names do not matter.
  • PNGs only.
  • Number of samples per folder do not matter, however you want enough, 80% for training, 20% for validation.

├── training_data
│ ├── char_select
│ │ ├── image_1.png
│ │ ├── image_2.png
│ │ ├── image_3.png
│ ├── connection_lost
│ │ ├── image_1.png
│ │ ├── image_2.png
│ │ ├── image_3.png
│ ├── in_flight
│ │ ├── image_1.png
│ │ ├── image_2.png
│ │ ├── image_3.png
│ ├── in_hanger
│ │ ├── image_1.png
│ │ ├── image_2.png
│ │ ├── image_3.png

Usage:

Check out the samples Here.

ml_botting_core_config.json

{
   "public_models":[
      {
         "game":"eve_online",
         "model_name":"game_state",
         "download_latest":1,
         "download_latest_from":"https://storage.googleapis.com/eve_online_models/",
         "model_root_directory":"O:\\eve_live_models\\game_state",
         "model_log_directory":"O:\\eve_live_logs\\game_state",
         "save_images":0
      }
   ]
}

Implementation

from ml_botting_core import universal_predictor

up_config = json.load(open(r'ml_botting_core_config.json'))
up = universal_predictor(config=up_config)

img = Image.open('some_image.png')
state_result = up.predict(img, 'game_state')

state_result

{
   "epoc_time":"1682138565007.508",
   "argmax_index":2,
   "value_at_argmax":"1.0",
   "class":"jump_though_first",
   "classes":[
      "dock_now",
      "invalid",
      "jump_though_first",
      "jump_through_second",
      "no_action",
      "warp_to_dock_3",
      "warp_to_dock_4"
   ],
   "scores":[
      1.2750345662162804e-15,
      1.4581948495906438e-11,
      1.0,
      5.21881417175057e-17,
      1.4712418422554443e-18,
      1.2777047215389858e-12,
      6.730089694497203e-17
   ],
   "id":"98ad373b-e0a6-11ed-9b27-2cf05d9fe8eb",
   "image_saved":0,
   "model_name":"nav_options"
}

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

ml_botting_core-1.0.7.tar.gz (9.7 kB view hashes)

Uploaded Source

Built Distribution

ml_botting_core-1.0.7-py3-none-any.whl (10.8 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page