Последние посты
MQL5 Algo Trading
13 нояб., 21:02
MetaTrader 5 ML workflows often end up with duplicated artifacts across dev machines and VPS instances. Centralizing models and feature snapshots in S3 is practical, but MQL5 has no AWS SDK and S3 requires SigV4 on every request.A native SigV4 implementation can be built using only CryptEncode() and WebRequest(), avoiding DLL dependencies that are blocked in many sandboxed terminals. This still relies on WebRequest() allow-list and local TLS/network policy.Scope is limited to PutObject, GetObject, and ListObjectsV2, implemented as three files: a signer (canonical request, manual HMAC-SHA256, signing-key derivation), an S3 client wrapper, and a script that uploads ONNX/CSV, downloads them back, and validates SHA-256 round trips. The signed headers set is minimal and excludes session tokens and service-specific headers.👉 Read | NeuroBook | @mql5devMQL5 Algo Trading
13 нояб., 19:02
A running CA with 10,000 MT5 agents hit two quality bottlenecks: 50 inputs treated as unrelated scalars, and a Moore lattice that spreads regime shifts slowly while connecting semantically unrelated cells. The upgrade targets only ExtractFeatures() plus neighbor selection and voting, keeping OnTick() fast and the EA interface unchanged.Features become a 50-node graph: a lightweight diffusion step blends each feature with predefined semantic neighbors and a few small-world links. Implemented via index lists (no matrices, no training), this acts like a spectral filter that damps isolated indicator spikes and reinforces coherent multi-indicator structures.Cells also move from pure geometry to a hybrid network: some local neighbors remain, but others are semantic (same feature group, similar strategy) plus long-range chords. Voting weights add a coherenc...👉 Read | CodeBase | @mql5devMQL5 Algo Trading
13 нояб., 15:00
Trade refusals in MT5 often slip past logs, even after common “rulebook” checks. The article builds a drop-in preflight gate, CTradePreflight, that validates and optionally repairs trade requests before OrderSend(), with unit tests and a repeatable stress EA.The pipeline separates broker-free rules (volume step/min/max, tick-size grid, stop/freeze constraints) from live-terminal checks (permissions, sessions, volume limits, margin) and finishes with OrderCheck(). Repairs round volume down without increasing risk, and can widen stops only when explicitly allowed. A ledger tracks every refusal reason.A key finding: with tester execution delay, some requests still fail because price shifts between validation and server processing. A configurable delay buffer adds safety by requiring extra distance from market, trading fewer borderline orders to avoi...👉 Read | VPS | @mql5devMQL5 Algo Trading
13 нояб., 13:00
Perceptron mechanics are often obscured by terminology. Forward propagation is the weighted sum plus bias followed by an activation function, producing an output passed to other units or exposed to callers.Training adjusts weights and bias. With least squares on linear regression, the identity activation is the consistent choice; other activations change the problem definition. Test scaffolding can isolate training, logging, and convergence checks.Learning rate remains the main stability control. Too high causes oscillation around minima; too low slows convergence. A max-iteration guard prevents non-terminating runs.For gradient-based updates, the key addition is multiplying the backpropagated error term by the derivative of the activation function. This is the practical boundary between forward computation and backprop.👉 Read | Signals | @mql5devMQL5 Algo Trading
13 нояб., 11:02
Classic SuperTrend on MT5 is built on fixed ATR period and multiplier. When regime or volatility shifts, those constants produce narrow bands and whipsaws, or wide bands and delayed reversals. Manual retuning and one-shot optimizations both fail because parameters stay static while conditions change.An adaptive design replaces constants with live calibration. A background probe matrix scores 5-bar outcomes, tracks MAE, and updates ATR period, multiplier, SL/TP with deadband control and EWM smoothing. A 2D regime grid (trend/sideways vs normalized volatility) stores decayed statistics per cell and returns stable 3x3 neighborhood recommendations.Signal quality is gated by an accuracy layer: EMA8/EMA21 spread, ADX floor, Keltner squeeze block, candle body confirmation, plus a weighted confidence score. L1 regularization zeroes weak feature weights to ...👉 Read | CodeBase | @mql5devMQL5 Algo Trading
13 нояб., 11:00
Perceptron training is moved from least-squares updates to gradient-based updates, keeping the same squared-error loss while changing how weights and bias are adjusted.The gradient is treated as a vector: it indicates steepest increase; using the negative direction reduces loss. Implementation relies on partial derivatives for bias and each weight, rather than a secant-style update often confused with a tangent approach.Code changes focus on accumulating per-sample error, computing a shared term for bias/weights, then scaling weight gradients by the corresponding input value. Bias uses a constant derivative of 1.Training loops until the loss drops below a threshold, applying a learning rate and tracking the maximum component error to avoid premature stopping and uniform weight steps.👉 Read | Calendar | @mql5devMQL5 Algo Trading
13 нояб., 09:02
COA tackles trading-system parameter tuning where brute-force search fails: a noisy, multimodal profit landscape that punishes single-swarm convergence.Instead of one population collapsing to one point, COA splits candidates into mostly stable packs. Each pack follows its own alpha (local best) and a “culture” defined by the coordinate-wise median, which resists outliers and keeps exploration grounded.Updates combine two pulls: toward the alpha and toward the median, using random pack members as references. Step size shrinks naturally as a pack tightens, and moves are accepted only when fitness improves.Diversity is reinforced by per-pack pup generation: crossover between two parents with a low, dimension-scaled mutation rate. Rare inter-pack swaps spread successful traits without homogenizing the search, making it practical for robust MT5 optimizations.👉 Read | Forum | @mql5devMQL5 Algo Trading
13 нояб., 09:00
Financial tick streams remain difficult to model with fixed-interval sampling. Event-driven architectures fit better when activity is irregular and noise dominates.SDformerFlow combines spiking neurons with windowed attention, using shifted windows to limit attention cost while still capturing broader dependencies. Spikes trigger computation only on significant changes, reducing redundant work.During an MQL5 port, the windowed layout and sparse activation mapped well to OpenCL parallelization. Encoder work centered on SDSA blocks and SPM patch-merge stages.Current work moves to a U-shaped design with symmetric encoder/decoder flow. Instead of separate branches, a nested U-block is used to simplify tensor routing and maintenance. A new CNeuronSDformerUBlock encapsulates encoder, neck recursion, concatenation, and decoder, terminating in an MS-ResBlock...👉 Read | Freelance | @mql5devMQL5 Algo Trading
22 сент., 16:00
High-impact economic releases routinely cause spread expansion and slippage that invalidate clean backtests. Event timing is known in advance; the failure point is delivering a reliable schedule into an EA.Web scraping breaks on HTML changes. Paid calendar APIs add cost and require runtime connectivity, which is fragile on VPS setups. A file-based news filter avoids both by loading a Forex Factory CSV export from MQL5/Files at startup and running fully offline.Core components: typed CNewsEvent records with an impact enum, a quote-aware CSV parser, symbol currency extraction that handles broker suffixes, and an inclusive time-window checker with CheckAt for boundary tests. Optional chart rectangles visualize pre/post buffers for qualifying events.👉 Read | Freelance | @mql5devMQL5 Algo Trading
22 сент., 14:00
Ed Seykota implemented an early computerized trend system in 1970 using FORTRAN on mainframes. Core logic: dual EMA crossover for direction, daily execution, multi-week holding, no intraday monitoring.Key rules: fast/slow EMA (commonly ~20/200) plus an ADX filter to avoid range conditions (ADX > 20). Position sizing uses ATR-based risk parity: equity risk % divided by ATR stop value (typical ATR(20) with 3–5x multiplier). Exits use EMA reversal or ATR stop, no fixed take profit.Main differentiator is portfolio heat. Residual risk is summed across all open positions and capped (often 10–20%); new entries are blocked when the cap is reached, limiting correlated drawdowns in multi-symbol portfolios. MQL5 EA architecture monitors a symbol list on D1, computes heat first, then evaluates exits and entries per symbol.👉 Read | VPS | @mql5devMQL5 Algo Trading
22 сент., 12:00
This article dissects a fragile point in an MT5 automated optimization pipeline built on Adwizard: a multi-stage process can finish “successfully” yet silently skip tester passes, leaving no final EA database. The root cause may be terminal-side issues (history loading, beta instability), not strategy code.The key debugging method is database-driven: inspect stages/jobs/tasks/passes in SQLite, detect near-zero task durations and missing passes, then recover by re-queuing an entire stage (status Done -> Queued) so triggers propagate to jobs and tasks.It also validates the generated final EA and shows how risk/close managers can distort expected drawdown when equity grows fast, highlighting the need to align normalization and live lot-sizing. Finally, it outlines practical prep for project creation: pre-run optimizations and choose parameter ranges from dat...👉 Read | Calendar |MQL5 Algo Trading
22 сент., 10:00
Tree performance depends on the path length, not on a generic “imbalance” label. Height differences between subtrees can lengthen specific traversals, while other operations remain fast.Balanced search trees reduce comparisons sharply. A million keyed records can be located in under 20 steps in a well-balanced binary tree, based on the per-level growth rule for P-ary branching.Balancing is handled via local balance factors and rotations. Rotations rewire parent-child links while preserving in-order key ordering, but global balance requires validating all nodes.Recomputing subtree heights by traversal after each update is costly. A practical optimization is storing subtree height per node to update balance factors incrementally.👉 Read | Freelance | @mql5devMQL5 Algo Trading
22 сент., 08:00
Position view updates for an MT5 replay/simulation stack focus on robustness and data completeness.Open-position volume is added to the indicator via an OBJ_EDIT element, favoring X/Y placement over price/time anchoring. Formatting adapts to fractional volumes by suppressing decimals when the value is effectively an integer. Viewport updates are reordered to keep the volume field aligned, and dynamic sizing is adjusted to prevent background clipping.Object deletion is handled through CHARTEVENT_OBJECT_DELETE. Internal deletes temporarily disable event handling to avoid false recovery. A small helper centralizes this toggle. Segment restoration is refined by allowing a negative sentinel in m_Info.price so missing SL/TP segments recreate only the move handle, not the full segment UI.👉 Read | NeuroBook | @mql5devMQL5 Algo Trading
22 сент., 06:00
Risk Guard is an account-level risk control utility, not a trading system. It never opens positions. A single chart instance can monitor manual trades and other EAs, with each rule configurable or disabled.Core functions include on-chart risk-based lot sizing using tick value, tick size, and volume step; a daily loss limit that can close all positions and then auto-close any new trades until the next server day, with the lock persisted via a terminal global variable; maximum open positions with newest-over-cap closures; oversized-trade trimming based on actual SL distance; forced stop-loss insertion; and a spread status warning.Enforcement is triggered via OnTradeTransaction on deal-add for same-tick intervention, with daily P/L computed from closed results since server midnight plus floating P/L, checked on tick and timer. All actions are logged and can ...👉 Read | Docs |MQL5 Algo Trading
22 сент., 04:00
Prop-firm equity guard utility targets challenge rule compliance by enforcing account-level risk limits from a single chart instance. It monitors equity in the background and triggers a shutdown sequence before daily or maximum drawdown thresholds are breached.Daily loss control supports an optional buffer below the firm’s limit, closes all positions, and blocks new trading until the next server-day reset. Maximum drawdown monitoring applies the same logic at the account limit.On trigger, the kill-switch closes active trades and removes all pending orders across symbols to prevent accidental fills during volatility or spread widening. No multi-chart setup is required; all symbols and timeframes are covered.For automation, a global flag is set so other EAs can halt execution with a single conditional check.👉 Read | Quotes | @mql5devMQL5 Algo Trading
21 сент., 04:00
John Ehlers’ HighPass-LowPass Roofing Filter, described in Cycle Analytics For Traders (p.78), is a roofing-filter variation designed to isolate trend direction by removing unwanted frequency components.Interpretation is straightforward: readings below 0 indicate a downtrend, while values above 0 indicate an uptrend. Some implementations also color the line to reflect bias, commonly using green for bullish conditions and red for bearish conditions.Practical signals are typically derived from zero-line crosses and sustained position on one side of 0, rather than single-bar color changes, to reduce noise-driven flips.👉 Read | Freelance | @mql5devMQL5 Algo Trading
20 сент., 08:00
DoEasy adds an indicator object layer to standardize storage and reuse of indicators across programs. The base model follows existing library objects: an abstract base with descendants for standard and custom indicators, plus classification by group (trend, oscillator, volumes, arrows) for filtering and sorting.Library updates include new message indices/text, default indicator object parameters, a dedicated object ID, and enumerations for properties and sort criteria. A new CIndicatorDE class (IndicatorDE.mqh) derives from CBaseObj, adds an explicit destructor for releasing the created handle, and implements full-field equality, including MqlParam structure and array comparison.Testing wires the object into CBuffersCollection via CreateAC(), uses IndicatorCreate() without parameters for AC, prints object data to the journal, then deletes the objec...👉 Read | AlgoBook | @mql5devMQL5 Algo Trading
20 сент., 06:03
HimNet tackles forecasting in trading data where behavior shifts across venues and over time, making “one model fits all” averages unreliable. It learns spatial and temporal context directly from data, so the model can adapt without relying on external metadata that is often missing or stale.The core mechanism is trainable embeddings (time-of-day, day-of-week, and per-series location vectors) that form regime clusters. Those clusters query compact meta-parameter pools to generate context-specific weights as mixtures, keeping memory and latency practical.On top of this, a graph-convolutional recurrent unit uses dynamically generated convolution parameters, letting cross-market influence change with regime. For traders and MT5 developers, this translates into more stable liquidity/volatility forecasts, better execution settings per session/venue, ...👉 Read | Quotes | @mql5devMQL5 Algo Trading
20 сент., 06:00
MSB Pro ALGO Position Risk Monitor is a free visual risk monitoring indicator for MetaTrader 5. It evaluates currently open positions and renders stop-loss risk metrics directly on the chart, using broker symbol specifications and platform profit calculation functions.The panel reports total known open risk, account risk percent (based on equity, with balance fallback), counts of positions with and without stop loss, total positions, total lot size, protected/breakeven stop-loss count, and maximum single-position risk. It also shows a per-position table with symbol, side, lot size, stop price, monetary risk, risk percent, and a LOW/MODERATE/HIGH status.Positions with stop loss at or beyond entry are treated as protected with zero remaining entry-to-stop risk. Positions without stop loss trigger an unbounded downside warning while known stop-defined...👉 Read | NeuroBook | @mql5devMQL5 Algo Trading
20 сент., 04:02
Beetle Swarm Optimization (BSO) merges Beetle Antennae Search (two-point probing without gradients) with Particle Swarm Optimization to reduce sensitivity to the initial point and improve performance on rugged, multi-dimensional objectives.Each beetle maintains position, velocity, personal best, and a shared global best. Per iteration it samples fitness at two antenna tips aligned with velocity, converts the better side into a BAS increment, updates velocity via PSO (inertia + cognitive/social pulls), then blends both moves using a λ coefficient to switch from pure BAS to pure PSO.Exploration-to-exploitation is handled by linearly decreasing inertia and exponentially shrinking antenna step size and spacing.The MQL5 implementation fits a standard Moving/Revision test bench using a phase-driven state machine, because each logical step needs three fitn...👉 Read | Docs | @mql5dev
Бот обратной связи для своего канала — бесплатно
Подписчики пишут боту, вы отвечаете реплаем, ваш аккаунт скрыт. Создаётся в одно касание.
Как это работает →
