Caution

I have decided to stop developing this tool and move on to my other projects. However all scripts and codes are in this repository, so if anyone is interested, this tool can be updated and used normally.

As of March 21, 2025 E7 Vault, a site I get character portrait for image recognition is closed. This means that the tool is no longer able to detect new skins/characters. You may upload the images yourself in the folder to keep using the tool. The tool will also need a new training data (following the correct format), and trained which can be done with the get rec model script in workflow scripts folder. See the github actions for more details such as the python env and requirements...

Important folders/files if you wish to self update

• dataset: this is where all of the character images are stored for image recognition.
• data: rec models and bunch of csvs needed for the software/training is stored here
• match_histories: all match history csvs are stored here. These can be later merged with your updated match history csv if you wish with the training notebook
• workflow_script: pretty much all of the scripts that you will need to self update the tool is here.
  • full_training_notebook.ipynb: notebook that has codes that are needed for training rec model.
  • get_matches.py: write top player's matches into csv. However, this script will need an update if you wish to use it as the official e7 match history website updated their code recently.
  • other necessary updating scripts (other than get_matches.py and rec model training) can be ran with .github/workflows/update_data.yml github action or following similar steps with different tool

E7 RTA DRAFT HELPER

This tool uses image recognition to detect the game screen and employs a pre-trained model to recommend the best characters for the current RTA (Real Time Arena) draft
Quick Start · Report Bug · Request Feature

Table of Contents

1. About The Project
   • Built With
2. Features
3. Getting Started
   • Prerequisites
   • Installation
4. Initial Setup
5. Troubleshooting
6. Frequently Asked Questions
7. Roadmap
8. Contributing
9. License
10. Contact
11. Acknowledgments

About The Project

I created this tool because I am really bad at this game even though this game is competitive, there are very limited tools/guides compared to other competitive games such as league of legends, Valorant, Dota2, etc. This tool uses SIFT algorithm to detect characters on your game screen and then uses pretrained model to predict which character would fit the best in the current draft scenario. You can also get information on characters quickly by clicking on their portraits.

Note

The program relies solely on publicly available information, such as the RTA game screen and Epic 7 match history, to make recommendations. It does not have information on who your opponent is, who the user is or any other game or user specific data.

(back to top)

Built With

• 
• 
• 
• 
• 
• 
• 

Data from:

• E7 Match History
• Cecilia Bot

(back to top)

Features

• Display user statistics

  • You can set your account on Set User Data in the settings
  • Your stats will now display on detection screen

• 

• Detect your game screen and display data

  • Recommend top 10 characters that fits into your draft

• 

  • Predict your win rate with your current drafts

• 

  • Quickly get information about your enemy's pick

• 

  • Pause detection by pressing esc

• Get character information by clicking on character portrait

  • Information is available thanks to Ceciliabot

(back to top)

Getting Started

Follow these steps to get started

Prerequisites

Supported OS: Windows and Mac OS M series

Caution

Mac OS Version is currently not supported due to weird crashes and security issues. However there are plans in the future for it

Installation

1. Download Git from here
2. Make sure to have Git in path (Installer should have done this if you downloaded it with recommended settings and with the Setup and not Portable)
3. Download the latest release from releases

Note

.7z is much faster but you may need to install 7zip separately

4. Extract the files

5. run init_windows.bat

Note

The process may hang when installing python packages. This is normal behaviour. Please wait until it finishes.

(back to top)

Initial Setup

You will need to play a game of RTA to set up the program

1. Launch E7RTAHelper.exe

Important

If you are using BlueStacks or other emulators, you may need to launch E7RTAHelper as admin otherwise the program will not be able to detect your game

On Mac, you may need to give permissions such as recording your screen on the security & privacy settings

2. Click init/settings
3. Click Set Detection Screen Size
4. Open Epic 7
5. Select your game window title from the Select Window dropdown
6. Then go into a RTA match
7. After ban phase, click screenshot
8. First Adjust Crop top, right, bottom and left settings then press set so it captures like below

8. Then set the crop center settings so it removes the center

Your window should look like this after

9. Lastly make sure you update recommender to the latest version: Click Check for Updates in the settings. Make sure to Restart the tool after updating!

Now you are ready! Press Start Detection on the front page to get started and make sure to set First Pick from the drop down menu

Tip

Press Esc after drafting to switch the tool to idle mode and reduce lag

(back to top)

Troubleshooting

• It uses too much cpu/resources
  • The image recognition algorithm is set to use all the available resources by default. Consider lowering OpenCV Number of Threads to 1 or lower number instead of -1 in the performance settings
    
• It is stuck at Server Connected (1/2)
  • Make sure to have Git installed and on PATH (It should have been installed on PATH if you just downloaded git with default settings)
    

(back to top)

Frequently Asked Questions

• How do I update?

  • You can update the recommender model by clicking Check for Updates in the settings
    • Make sure to Restart your tool after the update!
  • For the update on the Software itself, you would need to download the new release on the github releases
    • You can keep your current settings by moving your config.cfg file to the newly downloaded folder
      • You may need to update the recommender model by clicking Check for Updates in the settings
        

• How frequently does the model and hero descriptions update?

  • Currently the model and the hero descriptions update weekly
    

(back to top)

Roadmap

• Display information for character's EE
• Setting for limiting characters in the recommender
• Make a settings for detection boundary so the user can lower or increase if character is not being detected properly
• Meta tracker
• Profile review page
• Auto update user stats
• Update algorithm for Synergy/Counter characters
• Preban and Postban

See the open issues for a full list of proposed features (and known issues).

(back to top)

Contributing

Contributions are what make the open source community such an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated.

If you have a suggestion that would make this better, please fork the repo and create a pull request. You can also simply open an issue with the tag "enhancement". Don't forget to give the project a star! Thanks again!

1. Download Godot version 4.2.2
2. Fork the Project
3. Clone and open the project on Godot
4. Make changes
5. Create your Feature Branch (git checkout -b feature/AmazingFeature)
6. Commit your Changes (git commit -m 'Add some AmazingFeature')
7. Push to the Branch (git push origin feature/AmazingFeature)
8. Open a Pull Request

(back to top)

License

Distributed under the Apache-2.0 License. See LICENSE.txt for more information.

(back to top)

Contact

For now, please open up an issue if you are having problem

(back to top)

Acknowledgments

• Ceciliabot for letting me use their amazing website for character descriptions
• This tool is a fan made tool and we are not affiliated with SmileGate or EpicSeven in any way.

(back to top)