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
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.
Data from:
-
Display user statistics
- You can set your account on
Set User Data
in the settings - Your stats will now display on detection screen
- You can set your account on
-
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
- Pause detection by pressing
-
Get character information by clicking on character portrait
- Information is available thanks to Ceciliabot
Follow these steps to get started
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
- Download Git from here
- Make sure to have Git in path (Installer should have done this if you downloaded it with recommended settings and with the
Setup
and notPortable
) - Download the latest release from releases
Note
.7z is much faster but you may need to install 7zip separately
-
Extract the files
-
run
init_windows.bat
Note
The process may hang when installing python packages. This is normal behaviour. Please wait until it finishes.
You will need to play a game of RTA to set up the program
- 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
- Click
init/settings
- Click
Set Detection Screen Size
- Open Epic 7
- Select your game window title from the
Select Window
dropdown - Then go into a RTA match
- After ban phase, click
screenshot
- First Adjust Crop top, right, bottom and left settings then press set so it captures like below
- Then set the crop center settings so it removes the center
Your window should look like this after
- 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
- 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
to1
or lower number instead of-1
in theperformance settings
- The image recognition algorithm is set to use all the available resources by default. Consider lowering
- 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)
- Make sure to have Git installed and on PATH (It should have been installed on PATH if you just downloaded git with default settings)
-
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
- You may need to update the recommender model by clicking
- You can keep your current settings by moving your
- You can update the recommender model by clicking
-
How frequently does the model and hero descriptions update?
- Currently the model and the hero descriptions update weekly
- Currently the model and the hero descriptions update weekly
- 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).
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!
- Download Godot version 4.2.2
- Fork the Project
- Clone and open the project on Godot
- Make changes
- Create your Feature Branch (
git checkout -b feature/AmazingFeature
) - Commit your Changes (
git commit -m 'Add some AmazingFeature'
) - Push to the Branch (
git push origin feature/AmazingFeature
) - Open a Pull Request
Distributed under the Apache-2.0 License. See LICENSE.txt
for more information.
For now, please open up an issue if you are having problem
- 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.