ForumSections

Discussing the article: "Neural Networks in Trading: Recurrent Modeling of Market Micromovements (Conclusion)" - page 3

New comment
djgagarin #:
Hmm… I don’t have any such entries in the log

It’s not in the log, but directly on the graph


Vladimir Sanin #:

It’s not in the journal, but directly on the graph


In Study.mq5, the Train method ends with the lines `
` 
   PrintFormat("%s -> %d -> %-15s %10.7f", __FUNCTION__, __LINE__, "STFS", cSTFS.getRecentAverageError());
   PrintFormat("%s -> %d -> %-15s %10.7f", __FUNCTION__, __LINE__, "Actor", cActor.getRecentAverageError());
   PrintFormat("%s -> %d -> %-15s %10.7f", __FUNCTION__, __LINE__, "Critic", cCritic.getRecentAverageError());
   ExpertRemove();
//---
  }
Once all iterations have completed, they output the training result to the log.

I’m still trying to get it to run.)) I launched Study Online in visual mode – and I got a whole load of errors like these

djgagarin #:

I’m still trying to get it to run.)) I launched Study Online in visual mode – and I got a whole load of errors like these

Do you have OpenCL enabled in the tester?
Dmitriy Gizlyk #:
Have you enabled OpenCL in the tester?
All three boxes are ticked

and the ‘Study’ section in the magazine ends like this

djgagarin #:

and the ‘Study’ section in the magazine ends like this

Good afternoon, Study runs on the chart, not in the strategy tester.
123
New comment