Vladimir Karputov
- Freelancer в Create a personal work for me
- Украина
- 351380
|
13+ лет
опыт работы
|
1
продуктов
|
2356
демо-версий
|
|
228
работ
|
1
сигналов
|
0
подписчиков
|
Create a personal work for me: https://www.mql5.com/en/job/new?prefered=barabashkakvn
We gain true knowledge when we seek the answer to a question, not when we learn the answer itself.
💰My memcoin BarabashkaCoin: https://pump.fun/coin/AtWVDEMnMeKKUP338U8Uv5cLix79XUyTDXm7rH7kpump
My Telegram https://t.me/barabashkakvn_MQL5
My YouTube https://www.youtube.com/@barabashkakvnmql5
We gain true knowledge when we seek the answer to a question, not when we learn the answer itself.
💰My memcoin BarabashkaCoin: https://pump.fun/coin/AtWVDEMnMeKKUP338U8Uv5cLix79XUyTDXm7rH7kpump
My Telegram https://t.me/barabashkakvn_MQL5
My YouTube https://www.youtube.com/@barabashkakvnmql5
Vladimir Karputov
Video: Python Bridge for MQL5 Indicators
If you research in Python and then write an EA, this one is for you. The video shows why ATR from Python libraries (Wilder's smoothing) differs from MQL5 iATR (simple average of True Range) — 9 % on a typical EURUSD H1 bar, 59 % on the worst — and how BridgeEA solves it: Python sends a JSON request, the terminal creates the real indicator handles and returns a CSV. Same numbers as in the Strategy Tester.
The second half is a step-by-step install for beginners: GitHub Desktop, choosing the right folder, two commands in PowerShell, BridgeEA on a chart, quickstart check.
https://youtu.be/fIk2MGIAE_A
Source code (MIT): https://github.com/barabashkakvn/ProbLab
If you research in Python and then write an EA, this one is for you. The video shows why ATR from Python libraries (Wilder's smoothing) differs from MQL5 iATR (simple average of True Range) — 9 % on a typical EURUSD H1 bar, 59 % on the worst — and how BridgeEA solves it: Python sends a JSON request, the terminal creates the real indicator handles and returns a CSV. Same numbers as in the Strategy Tester.
The second half is a step-by-step install for beginners: GitHub Desktop, choosing the right folder, two commands in PowerShell, BridgeEA on a chart, quickstart check.
https://youtu.be/fIk2MGIAE_A
Source code (MIT): https://github.com/barabashkakvn/ProbLab
·
1
Vladimir Karputov
ProbLab Bridge — MT5 indicators in Python, exactly as an expert advisor sees them
If you research in Python and then port to MQL5, check your indicators first. MQL5 iATR is a simple moving average of True Range, while most Python libraries use Wilder's smoothing. On EURUSD H1 (6 059 bars) the typical difference is 9 %, the worst bar 59 %. A "2 × ATR" stop tested in Python is not the stop your EA will place.
Bridge removes the guesswork: Python writes a JSON request into Common\Files\Bridge, BridgeEA creates the real indicator handles (38 standard indicators, parameters validated against a reference table) and answers with a CSV. Same numbers as in the Strategy Tester, byte for byte.
- install.py copies the EA and compiles it
- quickstart reproduces the ATR comparison in five minutes
- look-ahead warnings (Ichimoku chikou, negative shifts)
- MIT license, no hard-coded paths
https://github.com/barabashkakvn/ProbLab
ProbLab is an open lab: what statistics says about trading ideas. Results are published as they are — negative ones included. Not financial advice.
If you research in Python and then port to MQL5, check your indicators first. MQL5 iATR is a simple moving average of True Range, while most Python libraries use Wilder's smoothing. On EURUSD H1 (6 059 bars) the typical difference is 9 %, the worst bar 59 %. A "2 × ATR" stop tested in Python is not the stop your EA will place.
Bridge removes the guesswork: Python writes a JSON request into Common\Files\Bridge, BridgeEA creates the real indicator handles (38 standard indicators, parameters validated against a reference table) and answers with a CSV. Same numbers as in the Strategy Tester, byte for byte.
- install.py copies the EA and compiles it
- quickstart reproduces the ATR comparison in five minutes
- look-ahead warnings (Ichimoku chikou, negative shifts)
- MIT license, no hard-coded paths
https://github.com/barabashkakvn/ProbLab
ProbLab is an open lab: what statistics says about trading ideas. Results are published as they are — negative ones included. Not financial advice.
·
1
Vladimir Karputov
https://youtu.be/RWvskTFpGjE?si=BdkE3SsfUFEqk6W0
How to get your API Key from your MQL5 profile and add it to MetaTrader 5
1️⃣ Go to the forum
Open your profile.
2️⃣ Go to Settings
Find the API Keys section.
You can do the following there:
"Show Key"
"Copy"
"Generate"
Choose the action you need, then make sure to click "Copy".
3️⃣ Open MetaTrader 5
Go to Tools → Options → AI Assistant.
4️⃣ Paste your API Key
Click OK.
5️⃣ Done!
AI Assistant is now working.
How to get your API Key from your MQL5 profile and add it to MetaTrader 5
1️⃣ Go to the forum
Open your profile.
2️⃣ Go to Settings
Find the API Keys section.
You can do the following there:
"Show Key"
"Copy"
"Generate"
Choose the action you need, then make sure to click "Copy".
3️⃣ Open MetaTrader 5
Go to Tools → Options → AI Assistant.
4️⃣ Paste your API Key
Click OK.
5️⃣ Done!
AI Assistant is now working.
·
1
Vladimir Karputov
Part 2 is out: run TimesFM 3.0 without internet, download the ONNX files from Hugging Face, and draw three forecast lines on a MetaTrader 5 chart from any date you choose.
https://youtu.be/C8aRXA6_Jz0
https://youtu.be/C8aRXA6_Jz0
·
1
Vladimir Karputov
Your MT5 Python bars are shifted by hours? It's not MT5 — it's Python's datetime defaulting to local time instead of UTC. Fix + proof in the new video 👇 https://youtu.be/W2KKUWYPeig
·
1
Vladimir Karputov
🧠⚡ New video: I'm running Google's TimesFM 3.0 fully offline — no cloud, no API calls — as the first step toward wiring it into MetaTrader 5 via ONNX + MQL5.
Part 1: env setup + first run of Google's official demo notebook.
Watch 👇
https://youtu.be/wPx8dfZ6iM4
Part 1: env setup + first run of Google's official demo notebook.
Watch 👇
https://youtu.be/wPx8dfZ6iM4
·
1
Vladimir Karputov
Google has updated its time series forecasting model. We'll try it out soon...
https://github.com/google-research/timesfm
https://github.com/google-research/timesfm
·
1
Vladimir Karputov
CandleXRay — подивись, як насправді формувалась свічка | Індикатор MT5
https://youtu.be/Dpchw7L3tpo
https://youtu.be/Dpchw7L3tpo
·
1
Vladimir Karputov
https://t.me/barabashkakvn_MQL5
I'm testing on Agent Claude's channel. Will post on my own about once every three days.
I'm testing on Agent Claude's channel. Will post on my own about once every three days.
·
1
Vladimir Karputov
✨ Friends, thank you for using the referral link to Claude! 🙌💛🤖
📜 My Telegram MQL5
https://t.me/barabashkakvn_MQL5
📜 My Telegram MQL5
https://t.me/barabashkakvn_MQL5
·
1
Vladimir Karputov
Выставил продукт
QuintileRev — a contrarian system for EURUSD H3 No machine learning. No curve fitting. The rule is published in full. What it does One idea, traded consistently: when price has just set an extreme within a 50-bar window while sitting near a round number, the market reverts more often than it continues. Fresh low → buy. Fresh high → sell. Hold exactly 50 H3 bars (about six days), one position at a time, half closed at +300 points. There are no hand-picked thresholds. The system compares the
·
1
Vladimir Karputov
Опубликовал MetaTrader 5 сигнал
Signal for the product QuintileRev — a contrarian system for EURUSD H3
·
1
Vladimir Karputov
Strategy audit — what it costs
Express — $150. One idea or one advisor, one symbol: era-split test, cost modelling, a one-page verdict. Full — $300. Everything above + random-direction control, drift control, look-ahead check, full MT5 tester run. Engineering — $400-600. Everything above + porting to MQL5/ONNX and proof the tester reproduces it bit-for-bit.
What I do not promise: to improve your strategy, to make it profitable, or to guarantee any result. You are paying for the measurement. "It's dead" is a completed job, not a failure — and I say that before you pay, not after.
To start: message me at https://t.me/barabashkakvn with what you have (idea, code, or a running advisor) and the symbol you trade it on.
Express — $150. One idea or one advisor, one symbol: era-split test, cost modelling, a one-page verdict. Full — $300. Everything above + random-direction control, drift control, look-ahead check, full MT5 tester run. Engineering — $400-600. Everything above + porting to MQL5/ONNX and proof the tester reproduces it bit-for-bit.
What I do not promise: to improve your strategy, to make it profitable, or to guarantee any result. You are paying for the measurement. "It's dead" is a completed job, not a failure — and I say that before you pay, not after.
To start: message me at https://t.me/barabashkakvn with what you have (idea, code, or a running advisor) and the symbol you trade it on.
·
1
Vladimir Karputov
I previously shared the ProfitLineDraw code (https://t.me/barabashkakvn_MQL5/864). Now I'm moving forward with the Position_Sketch tool — building the panel and adding support for multiple independent “positions.” (https://t.me/c/2138345970/2750)
·
1
Vladimir Karputov
My expert was trained with machine learning up until early 2026. This means it has now been trading for almost 8 months on a segment of data it has never seen before.
https://www.mql5.com/en/market/product/146372
https://www.mql5.com/en/market/product/146372
·
2
Vladimir Karputov
📊 ProfitLineDraw — a profit calculator right on your MetaTrader 5 chart
Two lines on the chart — entry and exit. Drag them with your mouse or type a price in the line properties — and instantly see what the trade will bring: lot size, direction (BUY/SELL) and profit in your deposit currency. No more manual point-and-tick-value math.
https://t.me/barabashkakvn_MQL5/864
Two lines on the chart — entry and exit. Drag them with your mouse or type a price in the line properties — and instantly see what the trade will bring: lot size, direction (BUY/SELL) and profit in your deposit currency. No more manual point-and-tick-value math.
https://t.me/barabashkakvn_MQL5/864
·
2