Stockfish Evaluation Guide

rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1

FEN

URL

RESET

	a	b	c	d	e	f	g	h
8									8
7									7
6									6
5									5
4									4
3									3
2									2
1									1
	a	b	c	d	e	f	g	h

Code

This shows a javascript implementation of this evaluation term .

This function calls following evaluation subterms: Middle game evaluation, End game evaluation, Phase, Scale factor, Tempo
It is used in no parent evaluation terms.

Per square values

	a	b	c	d	e	f	g	h
8									8
7									7
6									6
5									5
4									4
3									3
2									2
1									1
	a	b	c	d	e	f	g	h

	a	b	c	d	e	f	g	h
8									8
7									7
6									6
5									5
4									4
3									3
2									2
1									1
	a	b	c	d	e	f	g	h

Position representation and global functions used in evaluation terms.

Code

Position is represented with javascript object like this.

Global functions board(), colorflip() and sum().

Show position graph | Show game history graph

Stockfish NNUE is a port of a shogi neural network named NNUE (efficiently updateable neural network backwards) to Stockfish. In this tool only HalfKP 256x2-32-32-1 (rotate and flip) and HalfKAv2 512x2-16-32-1 with multiple PSQT outputs and multiple subnetworks (new) is currently supported (more info here).

Neural network

To use NNUE evaluation load supported neural network here:
You can download some networks here: https://tests.stockfishchess.org/nns

Current neural network: -
Description: -

PSQT WHITEPSQT BLACK

Per square value estimation

This table shows how much NNUE evaluation in pawns changes after piece is removed from chessboard.

	a	b	c	d	e	f	g	h
8									8
7									7
6									6
5									5
4									4
3									3
2									2
1									1
	a	b	c	d	e	f	g	h

Neuron activation

SHOW ACTIVE FEATURES

Code

Per square values

See also

External links

Code

Neural network

Per square value estimation

Neuron activation