Update default main net to nn-74f1d263ae9a.nnue #5459

linrock · 2024-07-06T23:59:38Z

Created by setting output weights (256) and biases (8) of the previous main net nn-ddcfb9224cdb.nnue to values found around 12k / 120k spsa games at 120+1.2

This used modified fishtest dev workers to construct .nnue files from spsa params, then load them with EvalFile when running tests:
https://github.com/linrock/fishtest/tree/spsa-file-modified-nnue/worker

Inspired by researching loading spsa params from files:
official-stockfish/fishtest#1926

Scripts for modifying nnue files and preparing params:
https://github.com/linrock/nnue-pytorch/tree/no-gpu-modify-nnue

spsa params:
weights: [-127, 127], c_end = 6
biases: [-8192, 8192], c_end = 64

Example of reading output weights and biases from the previous main net using nnue-pytorch and printing spsa params in a format compatible with fishtest:

import features
from serialize import NNUEReader

feature_set = features.get_feature_set_from_name("HalfKAv2_hm")
with open("nn-ddcfb9224cdb.nnue", "rb") as f:
    model = NNUEReader(f, feature_set).model

c_end_weights = 6
c_end_biases = 64

for i in range(8):
    for j in range(32):
        value = round(int(model.layer_stacks.output.weight[i, j] * 600 * 16) / 127)
        print(f"oW[{i}][{j}],{value},-127,127,{c_end_weights},0.0020")

for i in range(8):
    value = int(model.layer_stacks.output.bias[i] * 600 * 16)
    print(f"oB[{i}],{value},-8192,8192,{c_end_biases},0.0020")

For more info on spsa tuning params in nets:
#5149
#5254

Passed STC:
https://tests.stockfishchess.org/tests/view/66894d64e59d990b103f8a37
LLR: 2.94 (-2.94,2.94) <0.00,2.00>
Total: 32000 W: 8443 L: 8137 D: 15420
Ptnml(0-2): 80, 3627, 8309, 3875, 109

Passed LTC:
https://tests.stockfishchess.org/tests/view/6689668ce59d990b103f8b8b
LLR: 2.94 (-2.94,2.94) <0.50,2.50>
Total: 172176 W: 43822 L: 43225 D: 85129
Ptnml(0-2): 97, 18821, 47633, 19462, 75

bench 993416

Created by setting output weights (256) and biases (8) of the previous main net nn-ddcfb9224cdb.nnue to values found around 12k / 120k spsa games at 120+1.2 This used modified fishtest dev workers to construct .nnue files from spsa params, then load them with EvalFile when running tests: https://github.com/linrock/fishtest/tree/spsa-file-modified-nnue/worker Inspired by researching loading spsa params from files: official-stockfish/fishtest#1926 Scripts for modifying nnue files and preparing params: https://github.com/linrock/nnue-pytorch/tree/no-gpu-modify-nnue spsa params: weights: [-127, 127], c_end = 6 biases: [-8192, 8192], c_end = 64 Example of reading output weights and biases from the previous main net using nnue-pytorch and printing spsa params in a format compatible with fishtest: ``` import features from serialize import NNUEReader feature_set = features.get_feature_set_from_name("HalfKAv2_hm") with open("nn-ddcfb9224cdb.nnue", "rb") as f: model = NNUEReader(f, feature_set).model c_end_weights = 6 c_end_biases = 64 for i in range(8): for j in range(32): value = round(int(model.layer_stacks.output.weight[i, j] * 600 * 16) / 127) print(f"oW[{i}][{j}],{value},-127,127,{c_end_weights},0.0020") for i in range(8): value = int(model.layer_stacks.output.bias[i] * 600 * 16) print(f"oB[{i}],{value},-8192,8192,{c_end_biases},0.0020") ``` For more info on spsa tuning params in nets: official-stockfish#5149 official-stockfish#5254 Passed STC: https://tests.stockfishchess.org/tests/view/66894d64e59d990b103f8a37 LLR: 2.94 (-2.94,2.94) <0.00,2.00> Total: 32000 W: 8443 L: 8137 D: 15420 Ptnml(0-2): 80, 3627, 8309, 3875, 109 Passed LTC: https://tests.stockfishchess.org/tests/view/6689668ce59d990b103f8b8b LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 172176 W: 43822 L: 43225 D: 85129 Ptnml(0-2): 97, 18821, 47633, 19462, 75 bench 993416

Created by modifying L2 weights from the previous main net (nn-74f1d263ae9a.nnue) with params found by spsa around 9k / 120k games at 120+1.2. 370 spsa params - L2 weights in nn-74f1d263ae9a.nnue where |val| >= 50 A: 6000, alpha: 0.602, gamma: 0.101 weights: [-127, 127], c_end = 6 To print the spsa params with nnue-pytorch: ``` import features from serialize import NNUEReader feature_set = features.get_feature_set_from_name("HalfKAv2_hm") with open("nn-74f1d263ae9a.nnue", "rb") as f: model = NNUEReader(f, feature_set).model c_end = 6 for i in range(8): for j in range(32): for k in range(30): value = int(model.layer_stacks.l2.weight[32 * i + j, k] * 64) if abs(value) >= 50: print(f"twoW[{i}][{j}][{k}],{value},-127,127,{c_end},0.0020") ``` Among the 370 params, 229 weights were changed. avg change: 0.0961 ± 1.67 range: [-4, 3] The number of weights changed, grouped by layer stack index, shows more weights were modified in the lower piece count buckets: [54, 52, 29, 23, 22, 18, 14, 17] Found with the same method described in: official-stockfish#5459 Passed STC: https://tests.stockfishchess.org/tests/view/668aec9a58083e5fd88239e7 LLR: 3.00 (-2.94,2.94) <0.00,2.00> Total: 52384 W: 13569 L: 13226 D: 25589 Ptnml(0-2): 127, 6141, 13335, 6440, 149 Passed LTC: https://tests.stockfishchess.org/tests/view/668af50658083e5fd8823a0b LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 46974 W: 12006 L: 11668 D: 23300 Ptnml(0-2): 25, 4992, 13121, 5318, 31 bench 1029182

Created by modifying L2 weights from the previous main net (nn-74f1d263ae9a.nnue) with params found by spsa around 9k / 120k games at 120+1.2. 370 spsa params - L2 weights in nn-74f1d263ae9a.nnue where |val| >= 50 A: 6000, alpha: 0.602, gamma: 0.101 weights: [-127, 127], c_end = 6 To print the spsa params with nnue-pytorch: ``` import features from serialize import NNUEReader feature_set = features.get_feature_set_from_name("HalfKAv2_hm") with open("nn-74f1d263ae9a.nnue", "rb") as f: model = NNUEReader(f, feature_set).model c_end = 6 for i in range(8): for j in range(32): for k in range(30): value = int(model.layer_stacks.l2.weight[32 * i + j, k] * 64) if abs(value) >= 50: print(f"twoW[{i}][{j}][{k}],{value},-127,127,{c_end},0.0020") ``` Among the 370 params, 229 weights were changed. avg change: 0.0961 ± 1.67 range: [-4, 3] The number of weights changed, grouped by layer stack index, shows more weights were modified in the lower piece count buckets: [54, 52, 29, 23, 22, 18, 14, 17] Found with the same method described in: official-stockfish#5459 Passed STC: https://tests.stockfishchess.org/tests/view/668aec9a58083e5fd88239e7 LLR: 3.00 (-2.94,2.94) <0.00,2.00> Total: 52384 W: 13569 L: 13226 D: 25589 Ptnml(0-2): 127, 6141, 13335, 6440, 149 Passed LTC: https://tests.stockfishchess.org/tests/view/668af50658083e5fd8823a0b LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 46974 W: 12006 L: 11668 D: 23300 Ptnml(0-2): 25, 4992, 13121, 5318, 31 closes official-stockfish#5466 bench 1300471

Created by setting output weights (256) and biases (8) of the previous main net nn-ddcfb9224cdb.nnue to values found around 12k / 120k spsa games at 120+1.2 This used modified fishtest dev workers to construct .nnue files from spsa params, then load them with EvalFile when running tests: https://github.com/linrock/fishtest/tree/spsa-file-modified-nnue/worker Inspired by researching loading spsa params from files: official-stockfish/fishtest#1926 Scripts for modifying nnue files and preparing params: https://github.com/linrock/nnue-pytorch/tree/no-gpu-modify-nnue spsa params: weights: [-127, 127], c_end = 6 biases: [-8192, 8192], c_end = 64 Example of reading output weights and biases from the previous main net using nnue-pytorch and printing spsa params in a format compatible with fishtest: ``` import features from serialize import NNUEReader feature_set = features.get_feature_set_from_name("HalfKAv2_hm") with open("nn-ddcfb9224cdb.nnue", "rb") as f: model = NNUEReader(f, feature_set).model c_end_weights = 6 c_end_biases = 64 for i in range(8): for j in range(32): value = round(int(model.layer_stacks.output.weight[i, j] * 600 * 16) / 127) print(f"oW[{i}][{j}],{value},-127,127,{c_end_weights},0.0020") for i in range(8): value = int(model.layer_stacks.output.bias[i] * 600 * 16) print(f"oB[{i}],{value},-8192,8192,{c_end_biases},0.0020") ``` For more info on spsa tuning params in nets: official-stockfish#5149 official-stockfish#5254 Passed STC: https://tests.stockfishchess.org/tests/view/66894d64e59d990b103f8a37 LLR: 2.94 (-2.94,2.94) <0.00,2.00> Total: 32000 W: 8443 L: 8137 D: 15420 Ptnml(0-2): 80, 3627, 8309, 3875, 109 Passed LTC: https://tests.stockfishchess.org/tests/view/6689668ce59d990b103f8b8b LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 172176 W: 43822 L: 43225 D: 85129 Ptnml(0-2): 97, 18821, 47633, 19462, 75 closes official-stockfish#5459 bench 1120091

Created by modifying L2 weights from the previous main net (nn-74f1d263ae9a.nnue) with params found by spsa around 9k / 120k games at 120+1.2. 370 spsa params - L2 weights in nn-74f1d263ae9a.nnue where |val| >= 50 A: 6000, alpha: 0.602, gamma: 0.101 weights: [-127, 127], c_end = 6 To print the spsa params with nnue-pytorch: ``` import features from serialize import NNUEReader feature_set = features.get_feature_set_from_name("HalfKAv2_hm") with open("nn-74f1d263ae9a.nnue", "rb") as f: model = NNUEReader(f, feature_set).model c_end = 6 for i in range(8): for j in range(32): for k in range(30): value = int(model.layer_stacks.l2.weight[32 * i + j, k] * 64) if abs(value) >= 50: print(f"twoW[{i}][{j}][{k}],{value},-127,127,{c_end},0.0020") ``` Among the 370 params, 229 weights were changed. avg change: 0.0961 ± 1.67 range: [-4, 3] The number of weights changed, grouped by layer stack index, shows more weights were modified in the lower piece count buckets: [54, 52, 29, 23, 22, 18, 14, 17] Found with the same method described in: official-stockfish#5459 Passed STC: https://tests.stockfishchess.org/tests/view/668aec9a58083e5fd88239e7 LLR: 3.00 (-2.94,2.94) <0.00,2.00> Total: 52384 W: 13569 L: 13226 D: 25589 Ptnml(0-2): 127, 6141, 13335, 6440, 149 Passed LTC: https://tests.stockfishchess.org/tests/view/668af50658083e5fd8823a0b LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 46974 W: 12006 L: 11668 D: 23300 Ptnml(0-2): 25, 4992, 13121, 5318, 31 closes official-stockfish#5466 bench 1300471

Created by updating output weights (256) and biases (8) of the previous main net with values found with spsa around 101k / 120k games at 140+1.4. 264 spsa params: output weights and biases in nn-e8bac1c07a5a.nnue A: 6000, alpha: 0.602, gamma: 0.101 weights: [-127, 127], c_end = 6 biases: [-8192, 8192], c_end = 64 Among the 264 params, 189 weights and all 8 biases were changed. Changes in the weights: - mean: -0.111 +/- 3.57 - range: [-8, 8] Found with the same method as: official-stockfish#5459 Due to the original name (nn-ea8c9128c325.nnue) being too similar to the previous main net (nn-e8bac1c07a5a.nnue) and creating confusion, it was renamed by making non-functional changes to the .nnue file the same way as past nets with: https://github.com/linrock/nnue-namer To verify that bench is the same and view the modified non-functional bytes: ``` echo -e "setoption name EvalFile value nn-ea8c9128c325.nnue\nbench" | ./stockfish echo -e "setoption name EvalFile value nn-31337bea577c.nnue\nbench" | ./stockfish cmp -l nn-ea8c9128c325.nnue nn-31337bea577c.nnue diff <(xxd nn-ea8c9128c325.nnue) <(xxd nn-31337bea577c.nnue) ``` Passed STC: https://tests.stockfishchess.org/tests/view/669564154ff211be9d4ec080 LLR: 2.93 (-2.94,2.94) <0.00,2.00> Total: 57280 W: 15139 L: 14789 D: 27352 Ptnml(0-2): 209, 6685, 14522, 6995, 229 Passed LTC: https://tests.stockfishchess.org/tests/view/669694204ff211be9d4ec1b4 LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 63030 W: 16093 L: 15720 D: 31217 Ptnml(0-2): 47, 6766, 17516, 7139, 47 bench 1304515

Created by updating output weights (256) and biases (8) of the previous main net with values found with spsa around 101k / 120k games at 140+1.4. 264 spsa params: output weights and biases in nn-e8bac1c07a5a.nnue A: 6000, alpha: 0.602, gamma: 0.101 weights: [-127, 127], c_end = 6 biases: [-8192, 8192], c_end = 64 Among the 264 params, 189 weights and all 8 biases were changed. Changes in the weights: - mean: -0.111 +/- 3.57 - range: [-8, 8] Found with the same method as: official-stockfish#5459 Due to the original name (nn-ea8c9128c325.nnue) being too similar to the previous main net (nn-e8bac1c07a5a.nnue) and creating confusion, it was renamed by making non-functional changes to the .nnue file the same way as past nets with: https://github.com/linrock/nnue-namer To verify that bench is the same and view the modified non-functional bytes: ``` echo -e "setoption name EvalFile value nn-ea8c9128c325.nnue\nbench" | ./stockfish echo -e "setoption name EvalFile value nn-31337bea577c.nnue\nbench" | ./stockfish cmp -l nn-ea8c9128c325.nnue nn-31337bea577c.nnue diff <(xxd nn-ea8c9128c325.nnue) <(xxd nn-31337bea577c.nnue) ``` Passed STC: https://tests.stockfishchess.org/tests/view/669564154ff211be9d4ec080 LLR: 2.93 (-2.94,2.94) <0.00,2.00> Total: 57280 W: 15139 L: 14789 D: 27352 Ptnml(0-2): 209, 6685, 14522, 6995, 229 Passed LTC: https://tests.stockfishchess.org/tests/view/669694204ff211be9d4ec1b4 LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 63030 W: 16093 L: 15720 D: 31217 Ptnml(0-2): 47, 6766, 17516, 7139, 47 bench 1736720

Created by updating output weights (256) and biases (8) of the previous main net with values found with spsa around 101k / 120k games at 140+1.4. 264 spsa params: output weights and biases in nn-e8bac1c07a5a.nnue A: 6000, alpha: 0.602, gamma: 0.101 weights: [-127, 127], c_end = 6 biases: [-8192, 8192], c_end = 64 Among the 264 params, 189 weights and all 8 biases were changed. Changes in the weights: - mean: -0.111 +/- 3.57 - range: [-8, 8] Found with the same method as: official-stockfish#5459 Due to the original name (nn-ea8c9128c325.nnue) being too similar to the previous main net (nn-e8bac1c07a5a.nnue) and creating confusion, it was renamed by making non-functional changes to the .nnue file the same way as past nets with: https://github.com/linrock/nnue-namer To verify that bench is the same and view the modified non-functional bytes: ``` echo -e "setoption name EvalFile value nn-ea8c9128c325.nnue\nbench" | ./stockfish echo -e "setoption name EvalFile value nn-31337bea577c.nnue\nbench" | ./stockfish cmp -l nn-ea8c9128c325.nnue nn-31337bea577c.nnue diff <(xxd nn-ea8c9128c325.nnue) <(xxd nn-31337bea577c.nnue) ``` Passed STC: https://tests.stockfishchess.org/tests/view/669564154ff211be9d4ec080 LLR: 2.93 (-2.94,2.94) <0.00,2.00> Total: 57280 W: 15139 L: 14789 D: 27352 Ptnml(0-2): 209, 6685, 14522, 6995, 229 Passed LTC: https://tests.stockfishchess.org/tests/view/669694204ff211be9d4ec1b4 LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 63030 W: 16093 L: 15720 D: 31217 Ptnml(0-2): 47, 6766, 17516, 7139, 47 closes official-stockfish#5509 bench 1371485

Created from 2 distinct spsa tunes of the latest main net (nn-31337bea577c.nnue) and applying the params to the prior main net (nn-e8bac1c07a5a.nnue). This effectively reverts the modifications to output weights and biases in official-stockfish#5509 SPSA: A: 6000, alpha: 0.602, gamma: 0.101 1st - 437 feature transformer biases where values are < 25 54k / 120k games at 180+1.8 https://tests.stockfishchess.org/tests/view/66af98ac4ff211be9d4edad0 nn-808259761cca.nnue 2nd - 208 L2 weights where values are zero 112k / 120k games at 180+1.8 https://tests.stockfishchess.org/tests/view/66b0c3074ff211be9d4edbe5 nn-a56cb8c3d477.nnue When creating the above 2 nets (nn-808259761cca.nnue, nn-a56cb8c3d477.nnue), spsa params were unintentionally applied to nn-e8bac1c07a5a.nnue rather than nn-31337bea577c.nnue due to an issue in a script that creates nets by applying spsa results to base nets. Since they both passed STC and were neutral or slightly positive at LTC, they were combined to see if the elo from each set of params was additive. The 2 nets can be merged together with: https://github.com/linrock/nnue-tools/blob/90942d3/spsa/combine_nnue.py ``` python3 combine_nnue.py \ nn-e8bac1c07a5a.nnue \ nn-808259761cca.nnue \ nn-a56cb8c3d477.nnue ``` To print the spsa params with nnue-pytorch: ``` import features from serialize import NNUEReader feature_set = features.get_feature_set_from_name("HalfKAv2_hm") with open("nn-31337bea577c.nnue", "rb") as f: model = NNUEReader(f, feature_set).model c_end = 16 for i,ft_bias in enumerate(model.input.bias.data[:3072]): value = int(ft_bias * 254) if abs(value) < 25: print(f"ftB[{i}],{value},-1024,1024,{c_end},0.0020") c_end = 6 for i in range(8): for j in range(32): for k in range(30): value = int(model.layer_stacks.l2.weight.data[32 * i + j, k] * 64) if value == 0: print(f"twoW[{i}][{j}][{k}],{value},-127,127,{c_end},0.0020") ``` Prepared with the same method as: official-stockfish#5459 Passed STC: https://tests.stockfishchess.org/tests/view/66b4d4464ff211be9d4edf6e LLR: 2.94 (-2.94,2.94) <0.00,2.00> Total: 136416 W: 35753 L: 35283 D: 65380 Ptnml(0-2): 510, 16159, 34416, 16597, 526 Passed LTC: https://tests.stockfishchess.org/tests/view/66b76e814ff211be9d4ee1cc LLR: 2.95 (-2.94,2.94) <0.50,2.50> Total: 159336 W: 40753 L: 40178 D: 78405 Ptnml(0-2): 126, 17497, 43864, 18038, 143 bench 1517119

@MinetaS

Created from 2 distinct spsa tunes of the latest main net (nn-31337bea577c.nnue) and applying the params to the prior main net (nn-e8bac1c07a5a.nnue). This effectively reverts the modifications to output weights and biases in official-stockfish#5509 SPSA: A: 6000, alpha: 0.602, gamma: 0.101 1st - 437 feature transformer biases where values are < 25 54k / 120k games at 180+1.8 https://tests.stockfishchess.org/tests/view/66af98ac4ff211be9d4edad0 nn-808259761cca.nnue 2nd - 208 L2 weights where values are zero 112k / 120k games at 180+1.8 https://tests.stockfishchess.org/tests/view/66b0c3074ff211be9d4edbe5 nn-a56cb8c3d477.nnue When creating the above 2 nets (nn-808259761cca.nnue, nn-a56cb8c3d477.nnue), spsa params were unintentionally applied to nn-e8bac1c07a5a.nnue rather than nn-31337bea577c.nnue due to an issue in a script that creates nets by applying spsa results to base nets. Since they both passed STC and were neutral or slightly positive at LTC, they were combined to see if the elo from each set of params was additive. The 2 nets can be merged on top of nn-e8bac1c07a5a.nnue with: https://github.com/linrock/nnue-tools/blob/90942d3/spsa/combine_nnue.py ``` python3 combine_nnue.py \ nn-e8bac1c07a5a.nnue \ nn-808259761cca.nnue \ nn-a56cb8c3d477.nnue ``` Merging yields nn-87caa003fc6a.nnue which was renamed to nn-1111cefa1111.nnue with an updated nnue-namer around 10x faster than before by: - using a prefix trie for efficient prefix matches - modifying 4 non-functional bytes near the end of the file instead of 2 https://github.com/linrock/nnue-namer Thanks to @MinetaS for pointing out in #nnue-dev what the non-functional bytes are: L3 is 32, 4 bytes for biases, 32 bytes for weights. (fc_2) So -38 and -37 are technically -2 and -1 of fc_1 (type AffineTransform<30, 32>) And since InputDimension is padded to 32 there are total 32 of 2 adjacent bytes padding. So yes, it's non-functional whatever values are there. It's possible to tweak bytes at -38 - 32 * N and -37 - 32 * N given N = 0 ... 31 The net renamed with the new method passed non-regression STC vs. the original net: https://tests.stockfishchess.org/tests/view/66c0f0a821503a509c13b332 To print the spsa params with nnue-pytorch: ``` import features from serialize import NNUEReader feature_set = features.get_feature_set_from_name("HalfKAv2_hm") with open("nn-31337bea577c.nnue", "rb") as f: model = NNUEReader(f, feature_set).model c_end = 16 for i,ft_bias in enumerate(model.input.bias.data[:3072]): value = int(ft_bias * 254) if abs(value) < 25: print(f"ftB[{i}],{value},-1024,1024,{c_end},0.0020") c_end = 6 for i in range(8): for j in range(32): for k in range(30): value = int(model.layer_stacks.l2.weight.data[32 * i + j, k] * 64) if value == 0: print(f"twoW[{i}][{j}][{k}],{value},-127,127,{c_end},0.0020") ``` New params found with the same method as: official-stockfish#5459 Passed STC: https://tests.stockfishchess.org/tests/view/66b4d4464ff211be9d4edf6e LLR: 2.94 (-2.94,2.94) <0.00,2.00> Total: 136416 W: 35753 L: 35283 D: 65380 Ptnml(0-2): 510, 16159, 34416, 16597, 526 Passed LTC: https://tests.stockfishchess.org/tests/view/66b76e814ff211be9d4ee1cc LLR: 2.95 (-2.94,2.94) <0.50,2.50> Total: 159336 W: 40753 L: 40178 D: 78405 Ptnml(0-2): 126, 17497, 43864, 18038, 143 bench 1517119

@MinetaS

Created from 2 distinct spsa tunes of the latest main net (nn-31337bea577c.nnue) and applying the params to the prior main net (nn-e8bac1c07a5a.nnue). This effectively reverts the modifications to output weights and biases in official-stockfish#5509 SPSA: A: 6000, alpha: 0.602, gamma: 0.101 1st - 437 feature transformer biases where values are < 25 54k / 120k games at 180+1.8 https://tests.stockfishchess.org/tests/view/66af98ac4ff211be9d4edad0 nn-808259761cca.nnue 2nd - 208 L2 weights where values are zero 112k / 120k games at 180+1.8 https://tests.stockfishchess.org/tests/view/66b0c3074ff211be9d4edbe5 nn-a56cb8c3d477.nnue When creating the above 2 nets (nn-808259761cca.nnue, nn-a56cb8c3d477.nnue), spsa params were unintentionally applied to nn-e8bac1c07a5a.nnue rather than nn-31337bea577c.nnue due to an issue in a script that creates nets by applying spsa results to base nets. Since they both passed STC and were neutral or slightly positive at LTC, they were combined to see if the elo from each set of params was additive. The 2 nets can be merged on top of nn-e8bac1c07a5a.nnue with: https://github.com/linrock/nnue-tools/blob/90942d3/spsa/combine_nnue.py ``` python3 combine_nnue.py \ nn-e8bac1c07a5a.nnue \ nn-808259761cca.nnue \ nn-a56cb8c3d477.nnue ``` Merging yields nn-87caa003fc6a.nnue which was renamed to nn-1111cefa1111.nnue with an updated nnue-namer around 10x faster than before by: - using a prefix trie for efficient prefix matches - modifying 4 non-functional bytes near the end of the file instead of 2 https://github.com/linrock/nnue-namer Thanks to @MinetaS for pointing out in #nnue-dev what the non-functional bytes are: L3 is 32, 4 bytes for biases, 32 bytes for weights. (fc_2) So -38 and -37 are technically -2 and -1 of fc_1 (type AffineTransform<30, 32>) And since InputDimension is padded to 32 there are total 32 of 2 adjacent bytes padding. So yes, it's non-functional whatever values are there. It's possible to tweak bytes at -38 - 32 * N and -37 - 32 * N given N = 0 ... 31 The net renamed with the new method passed non-regression STC vs. the original net: https://tests.stockfishchess.org/tests/view/66c0f0a821503a509c13b332 To print the spsa params with nnue-pytorch: ``` import features from serialize import NNUEReader feature_set = features.get_feature_set_from_name("HalfKAv2_hm") with open("nn-31337bea577c.nnue", "rb") as f: model = NNUEReader(f, feature_set).model c_end = 16 for i,ft_bias in enumerate(model.input.bias.data[:3072]): value = int(ft_bias * 254) if abs(value) < 25: print(f"ftB[{i}],{value},-1024,1024,{c_end},0.0020") c_end = 6 for i in range(8): for j in range(32): for k in range(30): value = int(model.layer_stacks.l2.weight.data[32 * i + j, k] * 64) if value == 0: print(f"twoW[{i}][{j}][{k}],{value},-127,127,{c_end},0.0020") ``` New params found with the same method as: official-stockfish#5459 Passed STC: https://tests.stockfishchess.org/tests/view/66b4d4464ff211be9d4edf6e LLR: 2.94 (-2.94,2.94) <0.00,2.00> Total: 136416 W: 35753 L: 35283 D: 65380 Ptnml(0-2): 510, 16159, 34416, 16597, 526 Passed LTC: https://tests.stockfishchess.org/tests/view/66b76e814ff211be9d4ee1cc LLR: 2.95 (-2.94,2.94) <0.50,2.50> Total: 159336 W: 40753 L: 40178 D: 78405 Ptnml(0-2): 126, 17497, 43864, 18038, 143 closes official-stockfish#5534 bench 1613043

Created by setting output weights (256) and biases (8) of the previous main net nn-1111cefa1111.nnue to values found with spsa after 38k / 120k games at 120+1.2 using the same method as: official-stockfish#5459 nn-1111cefa1111.nnue -> nn-1cedc0ffeeee.nnue # weights changed: 185 mean: 0.0703 +/- 2.53 min: -6 max: 6 Passed STC: https://tests.stockfishchess.org/tests/view/6703589b86d5ee47d953bda1 LLR: 2.93 (-2.94,2.94) <0.00,2.00> Total: 101984 W: 26690 L: 26275 D: 49019 Ptnml(0-2): 375, 11944, 25926, 12385, 362 Passed LTC: https://tests.stockfishchess.org/tests/view/670542d286d5ee47d953befa LLR: 2.94 (-2.94,2.94) <0.50,2.50> Total: 106224 W: 27079 L: 26618 D: 52527 Ptnml(0-2): 71, 11508, 29487, 11981, 65 bench 1351413

vondele added the 🚀 gainer Gains elo label Jul 8, 2024

linrock mentioned this pull request Jul 9, 2024

Update default main net to nn-e8bac1c07a5a.nnue #5466

Closed

vondele added the to be merged Will be merged shortly label Jul 9, 2024

vondele closed this in 5752529 Jul 9, 2024

linrock mentioned this pull request Jul 21, 2024

Update default main net to nn-31337bea577c.nnue #5509

Closed

linrock mentioned this pull request Aug 14, 2024

Update default main net to nn-1111cefa1111.nnue #5534

Closed

linrock mentioned this pull request Oct 13, 2024

Update default main net to nn-1cedc0ffeeee.nnue #5632

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Update default main net to nn-74f1d263ae9a.nnue #5459

Update default main net to nn-74f1d263ae9a.nnue #5459

linrock commented Jul 6, 2024

Update default main net to nn-74f1d263ae9a.nnue #5459

Update default main net to nn-74f1d263ae9a.nnue #5459

Conversation

linrock commented Jul 6, 2024