What to feed to the input of the neural network? Your ideas... - page 46
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
I suggest to all "knowledgeable" to spend your time here, confirming your words with deeds. This will confirm your "professionalism" and save the forum from flooding, mud and insults.
You can organise your own MO championships, spend your time to your advantage.
Topic Transfer...
Don't flood here. Go prove yourself. Or are you only good at revenge?
I won't comment any more on your and your friends "proffesionals" comments. I am waiting for real actions from you in the specified thread, otherwise you are all useless fuflomites.
Don't flood here. Go prove yourself. Or are you only good at revenge?
I won't comment on your and your friends' "proffesionals" anymore. I am waiting for real actions from you in the specified thread, otherwise you are all useless fuflomites.
Moving the topic, admitting defeat, wanting to make up for it in another thread....
Don't flood here. Go prove yourself. Or are you only good at revenge?
I won't comment on your and your friends' "proffesionals" anymore. I am waiting for real actions from you in the specified thread, otherwise you are all useless fuflomites.
I thought the question was simple:
"What to feed to the neural network input? Your ideas..."
I think it was about ideas, not competence).
Initially, the most desirable thing is to feed the original series. But this doesn't always work due to prices going out of the learning range on new data.
Discuss augmentation/differentiation techniques with least loss of information for the second case would be a useful discussion.
Initially, the most desirable thing is to feed the original series. But this does not always work due to prices moving out of the training range on new data.
Discuss augmentation/differentiation techniques with least loss of information for the second case would be a useful discussion.
Always.
" due to prices going out of the learning range on new data. "
That's what normalisation is for.
Always.
" due to prices moving out of the learning range on new data. "
That's what rationing is for.
This is also interesting, e.g. which window to choose for normalisation. After all, if you normalise all the data, the system can get stuck in one of the extreme positions on new data.
That's interesting too, like which window to choose for rationing
Window? Number of NS inputs.
Window? Number of NS inputs.
history window... normalise the entire training history or renormalise it at intervals.
or normalise each time before feeding the NS :)