At this moment don't see a mistake. But I agree results are "too good to be true". Will implement trading on a demo account then will see.. Here tested with SL10 and TP10. Included stats regarding Asia session performance:

no. AI made with python build in AI modules

Exciting results.. turns out self learning AI is possible and the longer AI trades/learns the better results it provide. how I tested this: 1. Made 45000 H1 candles training data 2. Split data by 100 H1 candles so AI trained first iterations with ...

thank you! you make good points Here is what OpenAI answered to me after several discussion iterations where I confronted OpenAI with your posts: I'm using the Decision Trees algorithm for my AI model and right now, I'm using the default ...

Yes, the divergence of a model when using it on future data is a common issue in machine learning and is known as overfitting. The model is only able to accurately predict values based on the data it was trained on, so as new data becomes available, ...

Spent about 3 hours with OpenAI guidance and learned how to make simple AIs with python. Here is what I did: 1. Made another MQL4 script to export CSV training data for H1 history 2. With OpenAI help (explanations and generated python code) made ...

all done. fine-tuning ended.. but.. results are weird.. when I send my special prompt.. AI did not reply with one of 3 answers expected (as in the training data provided) but returned this mess instead: ...

seems this helped.. training 9000 H1 data validation 1000 H1 data most cheapest model "ada". now in training progress.. 25% so far. cost of this fine tune: 6.13 usd processed 100% but detected that my training data was wrong... Edited: now prepared ...

Here how would look my prompt message: M11 D3 H18 RSI=(38 39 59 50 46 31 35 36 45 33 44 49 54 50 52 54 52 48 50 47 45 47 50 50 55 45 36 28 32 25 28 31 21 31 31 36 39 32 41 49 53 53 53 52 54 54 53 47) MA=(1.54100 1.54169 1.54202 1.54214 1.54254 ...

Tried to train (fine tune) my OpenAI instance to trade EURAUD.. Here steps what i did: 1. Developed MQL4 script to export training data in jsonl file. Prompts and correct answers 2. Created 2 files - one for training AI (90000 H1 candles data) and ...

openAI have data only till 2021. But.. it is possible to use their API to fine-tune with fresh learning data for AI. This requires coding skills.

Openai have data on 2021. And no updating with fresh up to date info for now.

Openai is not 100% reliable. I had several issues with it.. for example ai could not say correct that trading 0.01 lots on 100 USD account is same risk as 1 lot on 10000 usd account. Had to ask additional questions till finally admits that risk % ...

Typically, there should not be liquidity issues when trading 1 lot of a major currency pair. These pairs have deep liquidity, which means that there is a large amount of buying and selling activity taking place, and it should be relatively easy to ...

Very rare making 10 pips trades. Mostly SL/TP from 20 to 100 pips. In backtests noticed that mostly after year 2017 patterns started to behave different. But they are still usable.. As example you can see analysis and live trades of notable numbers ...

Yes. My signal webpage alerted and it was possible to make entries as I posted. One more entry for AUDUSD recently made TP 60 pips.

Humans have more data input channels compared to AI now.. ok clear that all physical senses will be available to AI too. But there is 1 "sense" which is questionalbe will it be possible for AI - it is "soul" "intuition" or name it as you wish but ...

When AI will start to develop a new AI or upgrade itself... this will be tipping point. Exponential recursive re-development and development in any field - pharma, new materials, new energy sources, new technologies, new inventions, robots etc all ...

If AI could make money it would do it for companies that own AI infrastructure. AI will replace quite a lot of jobs related to content services (authors, lawyers, coders, etc). It is not that those authors/lawyers etc will get money.. AI owners will ...

After additional 2 hours spent.. finally, get working EA for MT4. I did not make fixes by using my coding skills.. this was a pure AI coding result. I just instructed AI to fix various issues till we finally get it working. // Input parameters input ...