TradingView
auroagwei
2019年11月12日午後10時29分

NAND Perceptron 

OANDA:EURUSD
EUR/USDOANDA

詳細

⋅
Experimental NAND Perceptron based upon Python template that aims to predict NAND Gate Outputs. A Perceptron is one of the foundational building blocks of nearly all advanced Neural Network layers and models for Algo trading and Machine Learning.

The goal behind this script was threefold:
  • To prove and demonstrate that an ACTUAL working neural net can be implemented in Pine, even if incomplete.
  • To pave the way for other traders and coders to iterate on this script and push the boundaries of Tradingview strategies and indicators.
  • To see if a self-contained neural network component for parameter optimization within Pinescript was hypothetically possible.


NOTE: This is a highly experimental proof of concept - this is NOT a ready-made template to include or integrate into existing strategies and indicators, yet (emphasis YET - neural networks have a lot of potential utility and potential when utilized and implemented properly).


Hardcoded NAND Gate outputs with Bias column (X0):
// NAND Gate + X0 Bias and Y-true
// X0 // X1 // X2 // Y
// 1 // 0 // 0 // 1
// 1 // 0 // 1 // 1
// 1 // 1 // 0 // 1
// 1 // 1 // 1 // 0


  • Column X0 is bias feature/input
  • Column X1 and X2 are the NAND Gate
  • Column Y is the y-true values for the NAND gate
  • yhat is the prediction at that timestep
  • F0,F1,F2,F3 are the Dot products of the Weights (W0,W1,W2) and the input features (X0,X1,X2)
  • Learning rate and activation function threshold are enabled by default as input parameters

    Uncomment sections for more training iterations/epochs:
  • Loop optimizations would be amazing to have for a selectable length for training iterations/epochs but I'm not sure if it's possible in Pine with how this script is structured.


  • Error metrics and loss have not been implemented due to difficulty with script length and iterations vs epochs - I haven't been able to configure the input parameters to successfully predict the right values for all four y-true values for the NAND gate (only been able to get 3/4; If you're able to get all four predictions to be correct, let me know, please).


// //---- REFERENCE for final output
// A3 := 1, y0 true
// B3 := 1, y1 true
// C3 := 1, y2 true
// D3 := 0, y3 true

PLEASE READ: Source article/template and main code reference:
* towardsdatascience.com/6-steps-to-write-any-machine-learning-algorithm-from-scratch-perceptron-case-study-335f638a70f3
* towardsdatascience.com/what-the-hell-is-perceptron-626217814f53
* towardsdatascience.com/how-to-build-your-own-neural-network-from-scratch-in-python-68998a08e4f6

リリースノート

⋅
//v5.6c - activation function error fix (was F > 0.25; corrected 1), line 99

リリースノート

⋅
//v5.6d - correction to activation function variable z not being keyed in + W0/W1/W2 not being factored in for initial iterations

リリースノート

⋅
// v6.4 - Dot product operation error for F0-F3 and W0-F3 fixed. Test for loop iterator for training.
// v6.5d -
// Loop Iteration for epoch training implemented
// Sum of Squared Error (SSE) implemented
// Y-pred vs Y-true color coded output option function (green/red)
// Custom input options for all arrays, including W0-W2
// Allows for custom of input features, weights, and bias - Default is NAND gate.
// Placeholder "========" for input options seperator for settings panel
// 3x Infopanel component for display output + match color (green/orange/red.)
// v6.6
// Gate detection including XOR/NOR (despite not being able to converge/solve with SLP Neurons - MLP + nonlinear activations required for XOR/NOR training and detection)

リリースノート

⋅
// v6.6b
// Missing XOR/XNOR MLP + nonlinear activation warning/message in yellow upon detection - fixed.
Centered OscillatorsLinear Regressionnnneuralnetperceptronannexperimentalmachinelearningdeeplearningtemplate
コメント
jackxryan
⋅
how to use it? Please describe it in layman language. Thanks.
OutsourcE
⋅
@jackxryan, greenish output means a long bias/redish output means a short bias, me thinks
allanster
⋅
Very nice. Thank you for pushing the boundaries of Pine and also for sharing!
auroagwei
⋅
@allanster, ye real neural network components/applications is some exciting stuff - I'm looking forward to seeing how this script is potentially used by other traders/programmers
dpanday
⋅
WOW!!
DavisNR
⋅
This effort should be front and center for Day Trading since the AI has taking on a whole new level of confidence. Your thoughts? I am ready to code it up. Need a team. Let me know .........
Meshidorin
⋅
Truly amazing work with this script. Are you still working on it now that we finally have arrays? :)
mrgr888n
⋅
and the oscar for hardcoding goes toooo .... =))

can it be used on xbt/usd pair?
auroagwei
⋅
@mrgr888n, see the description:

"NOTE: This is a highly experimental proof of concept - this is NOT a ready-made template to include or integrate into existing strategies and indicators, yet (emphasis YET - neural networks have a lot of potential utility and potential when utilized and implemented properly).

The goal behind this script was threefold:
To prove and demonstrate that an ACTUAL working neural net can be implemented in Pine, even if incomplete.
To pave the way for other traders and coders to iterate on this script and push the boundaries of Tradingview strategies and indicators.
To see if a self-contained neural network component for parameter optimization within Pinescript was hypothetically possible."
RafaelZioni
⋅
@auroagwei, look very interesting idea, great work :)
詳細