Machine learning in trading: theory, models, practice and algo-trading - page 2938
![MQL5 - Language of trade strategies built-in the MetaTrader 5 client terminal](https://c.mql5.com/i/registerlandings/logo-2.png)
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Right click on the project name. It should be like this:
There is some kind of authorisation somewhere
Yes, there is such a thing, but it gives an error
There's some sort of authorisation somewhere.
Well, I don't know. Try leaving and deleting the project, then rejoin it, then get an update from the repository.
When you leave a project in the log:
When deleting, nothing is written. When joining a project:
When receiving an update:
Authorisation was only when connecting to the repository - it is impossible to join the project without it.
I can't help you any more. Unless you have an old version - I did it in the latest (beta and release).
Well, I don't know. Try leaving and deleting the project, then rejoin it, then get an update from the repository.
When you leave a project in the log:
When deleting, nothing is written. When joining a project:
When receiving an update:
Authorisation was only when connecting to the repository - without it it is basically impossible to join the project.
I can't help you with anything else. Unless you have an old version - I did it in the latest (beta and release).
Got it, thanks.
Just in case, I'm posting the project in the archive.
What account did you log in to MQL5 Storage?
Did you specify your email or exact case-sensitive login?
Email is not accepted.
on login login, only my login is crooked initially, with dots dr.mr.mom, I have long ago, can you change the login?
Released beta 3589, on which you can test the operation of ONNX models.
Steps:
. Then go to step 2.
The project will be downloaded from MQL5 Storage (Subversion).
In this case, download the archive ONNX Runtime 1.14, signed by Microsoft, expand it to the root directory of the terminal next to terminal64.exe and restart the terminal:
In the project the simplest trained for the sake of example model.onnx is immediately attached so that you can see the demonstration.
To train the model yourself, you need to use the python scripts included in the project:
python.exe -m pip install --upgrade pip pip install --upgrade tensorflow pip install --upgrade pandas pip install --upgrade scikit-learn pip install --upgrade matplotlib pip install --upgrade tqdm pip install --upgrade metatrader5 pip install --upgrade onnx==1.12 pip install --upgrade tf2onnx
The onnx package with explicitly state version 1.12 so as not to conflict with tensorflow requirements.If necessary, specify the path yourself, if the editor can't detect it by itself.
Make sure that integration with Python is enabled in the terminal settings. If it was switched off, switch it on and restart the terminal:
The script will connect to the terminal with the help of MetaTrader5 package (if it is switched off, it will start automatically) and start training on EURUSD, H1:
When finished, the Journal editor will contain:
. The new model.onnx will be saved/updated in the project. It can already be used.
.
This is a specially created example to demonstrate both training and running the models.
You can convert your own TensorFlow models to ONNX format using the python packages onnx and tf2onnx:
ps: use the latest version of the project, as the function prototypes have changed in the old version.
It worked.
Package onnx with an explicit mention of version 1.12 so as not to conflict with tensorflow requirements.
Could not install this package. First it required CMake to be installed. Now it requires Visual Studio to be installed?
Could not install this package. First they demanded to install CMake. And now they require Visual Studio to be installed?
You are still at the beginning of the road of fighting the hell of mutual incompatibilities and creating separate environments for tasks in Python.
Some packages require compilers for the target build.
You're the one who is still at the beginning of the journey of dealing with the hell of mutual incompatibilities and creating separate environments for tasks in Python.
Some packages require compilers for the target build.
Absolutely agree! It's a lot of work. Two questions in this regard:
1. Can models created using GPU be converted to ONNX?
2. We don't need Python to get the model prediction in ONNX in the MQL5 Expert Advisor, do we? The calculations will be performed by the onnxruntime environment?