Machine learning in trading: theory, models, practice and algo-trading - page 1959
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
In general, I found a technology that will solve the grail problem. We take GPT-3, write "make a button to mow money" and it spits out working code.
Q: How do you spell morgul?
A: Morgul should be sporked with spork.
Q: How many nuts in a dumpling?
A: There are three nuts in a dumpling.
Q: How many rainbows does it take to jump from Hawaii to seventeen?
A: It takes two rainbows to jump from Hawaii to seventeen.
Q: What colorless green ideas sleep fiercely?
A: Colorless green ideas sleep fiercely.
Q: Do you understand these questions?
A: I understand these questions.
Repeat this instead of the morning prayer.
However strong AI may be, but retardation and lack of understanding 'what to do in general' can only be cured by cramming)
Are you trolling me or what? I was drunk yesterday, I don't know much today, but I won at pool 3:0.)
============================================================
I just started to apply the time filter to my patterns, you react to the pattern only in a certain time range...
I use a time filter on my patterns...
Take 30 pips. Stop 10 pips.3 to 1
+ commission 1p.
If you look at the pattern, you'll see that it is a different one.
cuts :
each line is a separate characteristic of a separate pattern
The patterns are only for a profit, there are about 100 of them...
The picture shows one of the patterns
Every single pattern is traded in a strictly separate timeframe...
Here are the deals of one of the patterns
Equity of one of the patterns
another
the third
And so on to the hundred...
Are you trolling me or what? I was drunk yesterday, I don't know much today, but I won 3:0 in billiards.)
============================================================
I just started to apply the time filter to my patterns, you react to the pattern only in a certain time range...
I use a time filter on my patterns...
take 30 points, stop 10 points.3 to 1
+ commission 1p.
If you look at the pattern, you'll see that it is a different one.
cuts :
each line is a separate characteristic of a separate pattern
The patterns are only for the profit, there are about 100 of them...
Each separate pattern is traded in a strictly separate time interval...
Here are the trades of one of the patterns
+- we're talking about the same thing
i'm not tricking you, i'm just saying the right info is hard to remember if you don't repeat it every day ))
+- We're talking about the same thing.
I'm not joking, I'm just saying that the right information doesn't go straight into your head if you don't repeat it every day).
Yeah, yeah, yeah, yeah.)
Maybe the information does not even fill your head well, the person is not ready for this information morally or informationally, because for years it was circulating but there was no result, and then he matured, began to understand more etc., and now he began to look at the same information differently, from the position of new knowledge and the result is also different, although the data is the same ...
=================================
Pattern with one losing trade
I look at it
I think I may consider it as a bad prediction.)
Slightly different, digital computers work with information as 0 and 1. Whether there is current in the conductor or not forms arrays of higher levels. The analog signal can be converted to digital by descritization. But I think what happens here is not addition and mathematical operations with the fact of the presence of current in the conductor or lack thereof, but with the voltage level in a given conductor, which is the analog signal that they have somehow learned to add up. Well stupidly if you want the number five, it's 5 volts on the processor leg, if you want the number 8.345354346346, it's the same number of volts.
As I said in my video, there are two achilles' heels that are not currently realized with respect to the biological neuron. The first is to be able to cut off and make connections between neurons and the second is to work digitally, where there is a desCRIption in the form of a decimal order of magnitude. In the analog world these dimensions are much larger and I think are individual for each neuron, that is, the analog signal has infinite dimensions where we can refine the number by an infinite number of digits after the decimal point. The world of real numbers is like that :-)
If I understand correctly and the processor will work with analog signals then this is a real breakthrough where one heel of achilles is already annihilated.... Because inside our heads neuronchiks transfer analog signals between each other.
I think in the article it is still analog - the results of the "neuron weights" were written directly into the CPU. (there is no word that they have found a substitute for the transistor), which means that as there was a current in the form of 0 and 1, so it remains. For analog current we have no possibility to change it, no protection from external errors that can change it, no possibility to record and store current with different voltages and to process
I understand that the training plot, but this time I decided the worst result of the basic strategy for the period (before that I took the best) and decided to improve it. Here's how the basic one worked out
And the grid itself. Unfortunately, I have never seen a trained model fail to make a single minus in an optimization period. Correspondingly, it is impossible to calculate a number of parameters.
Besides the prefabricated metric built today that is involved in the training. It would be desirable to have such a tool in R, that we could additionally evaluate the polynomials, not only by their work, but visually too, for example...... There is a value of work of polynomials in real units, and here is how to analyze them in relation to the target?
And the arrows themselves, the way it works on the real.... But man, I had a hundred of them today......
analog grid
I understand that the training plot, but this time I decided the worst result of the basic strategy for the period (before that I took the best) and decided to improve it. Here's how the basic one worked out
And the grid itself. Unfortunately, I have never seen a trained model fail to make a single minus in an optimization period. Correspondingly, it is impossible to calculate a number of parameters.
Besides the prefabricated metric built today that is involved in the training. It would be desirable to have such a tool in P, that we could additionally evaluate the polynomials, not only by their work, but visually too, for example...... There is a value of polynomials in real units, and here is how to analyze them in relation to the target?
that's it, the grail
there will be a signal?