cuda:0
----------------------------------------------------------------
Layer (type) Output Shape Param #
================================================================
LayerNorm-1 [-1, 60, 4000] 8,000
Linear-2 [-1, 60, 12000] 48,000,000
Dropout-3 [-1, 8, 60, 60] 0
Linear-4 [-1, 60, 4000] 16,004,000
Dropout-5 [-1, 60, 4000] 0
Attention-6 [-1, 60, 4000] 0
Identity-7 [-1, 60, 4000] 0
LayerNorm-8 [-1, 60, 4000] 8,000
Linear-9 [-1, 60, 16000] 64,016,000
GELU-10 [-1, 60, 16000] 0
Dropout-11 [-1, 60, 16000] 0
Linear-12 [-1, 60, 4000] 64,004,000
Dropout-13 [-1, 60, 4000] 0
Mlp-14 [-1, 60, 4000] 0
Identity-15 [-1, 60, 4000] 0
Block-16 [-1, 60, 4000] 0
Linear-17 [-1, 512] 2,048,512
ReLU-18 [-1, 512] 0
Linear-19 [-1, 3] 1,539
================================================================
Total params: 194,090,051
Trainable params: 194,090,051
Non-trainable params: 0
----------------------------------------------------------------
Input size (MB): 0.90
Forward/backward pass size (MB): 47.83
Params size (MB): 740.39
Estimated Total Size (MB): 789.13
----------------------------------------------------------------
None
data/BCICIV_calib_ds1e_1000Hz.mat
data/BCICIV_calib_ds1b_1000Hz.mat
data/BCICIV_calib_ds1g_1000Hz.mat
data/BCICIV_calib_ds1a_1000Hz.mat
data/BCICIV_calib_ds1f_1000Hz.mat
data/BCICIV_calib_ds1c_1000Hz.mat
data/BCICIV_calib_ds1d_1000Hz.mat
Epoch 1/50
----------
train Loss: 1.0211 Acc: 0.4820
val Loss: 0.9764 Acc: 0.5330
test Loss: 1.0240 Acc: 0.4750
Epoch 2/50
----------
train Loss: 0.9457 Acc: 0.5030
val Loss: 0.9285 Acc: 0.5127
test Loss: 0.9886 Acc: 0.4650
Epoch 3/50
----------
train Loss: 0.9041 Acc: 0.5280
val Loss: 0.9121 Acc: 0.5127
test Loss: 0.9752 Acc: 0.4850
Epoch 4/50
----------
train Loss: 0.8927 Acc: 0.5100
val Loss: 0.9110 Acc: 0.5076
test Loss: 0.9573 Acc: 0.4800
Epoch 5/50
----------
train Loss: 0.8719 Acc: 0.5450
val Loss: 0.9154 Acc: 0.4822
test Loss: 0.9598 Acc: 0.4850
Epoch 6/50
----------
train Loss: 0.8696 Acc: 0.5310
val Loss: 0.9168 Acc: 0.4924
test Loss: 0.9553 Acc: 0.4450
Epoch 7/50
----------
train Loss: 0.8606 Acc: 0.5490
val Loss: 0.9051 Acc: 0.4518
test Loss: 0.9460 Acc: 0.4650
Epoch 8/50
----------
train Loss: 0.8593 Acc: 0.5660
val Loss: 0.8955 Acc: 0.5127
test Loss: 0.9409 Acc: 0.4450
Epoch 9/50
----------
train Loss: 0.8580 Acc: 0.5500
val Loss: 0.8945 Acc: 0.4924
test Loss: 0.9295 Acc: 0.4850
Epoch 10/50
----------
train Loss: 0.8481 Acc: 0.5600
val Loss: 0.8820 Acc: 0.5025
test Loss: 0.9340 Acc: 0.5050
Epoch 11/50
----------
train Loss: 0.8378 Acc: 0.5670
val Loss: 0.8864 Acc: 0.5228
test Loss: 0.9374 Acc: 0.5100
Epoch 12/50
----------
train Loss: 0.8326 Acc: 0.5750
val Loss: 0.9086 Acc: 0.4721
test Loss: 0.9382 Acc: 0.4600
Epoch 13/50
----------
train Loss: 0.8372 Acc: 0.5630
val Loss: 0.8896 Acc: 0.5025
test Loss: 0.9276 Acc: 0.4700
Epoch 14/50
----------
train Loss: 0.8321 Acc: 0.5660
val Loss: 0.9059 Acc: 0.4772
test Loss: 0.9227 Acc: 0.4850
Epoch 15/50
----------
train Loss: 0.8005 Acc: 0.5850
val Loss: 0.9147 Acc: 0.4873
test Loss: 0.9468 Acc: 0.4600
Epoch 16/50
----------
train Loss: 0.8013 Acc: 0.5720
val Loss: 0.9398 Acc: 0.4416
test Loss: 0.9556 Acc: 0.4850
Epoch 17/50
----------
train Loss: 0.8102 Acc: 0.6000
val Loss: 0.9638 Acc: 0.4569
test Loss: 0.9491 Acc: 0.4650
Epoch 18/50
----------
train Loss: 0.8045 Acc: 0.5740
val Loss: 0.9362 Acc: 0.4772
test Loss: 0.9330 Acc: 0.4600
Epoch 19/50
----------
train Loss: 0.7965 Acc: 0.5840
val Loss: 0.9432 Acc: 0.4721
test Loss: 0.9457 Acc: 0.4550
Epoch 20/50
----------
train Loss: 0.8083 Acc: 0.5810
val Loss: 0.8763 Acc: 0.5228
test Loss: 0.9183 Acc: 0.5000
Epoch 21/50
----------
train Loss: 0.7892 Acc: 0.5940
val Loss: 0.9020 Acc: 0.4619
test Loss: 0.9112 Acc: 0.4800
Epoch 22/50
----------
train Loss: 0.7814 Acc: 0.6180
val Loss: 0.9187 Acc: 0.5076
test Loss: 0.9268 Acc: 0.4950
Epoch 23/50
----------
train Loss: 0.7667 Acc: 0.6120
val Loss: 0.9415 Acc: 0.4873
test Loss: 0.9303 Acc: 0.4850
Epoch 24/50
----------
train Loss: 0.7572 Acc: 0.6400
val Loss: 0.9322 Acc: 0.5076
test Loss: 0.9447 Acc: 0.5150
Epoch 25/50
----------
train Loss: 0.7489 Acc: 0.6350
val Loss: 0.9303 Acc: 0.5330
test Loss: 0.9085 Acc: 0.5100
Epoch 26/50
----------
train Loss: 0.7656 Acc: 0.6080
val Loss: 0.9567 Acc: 0.5025
test Loss: 0.9517 Acc: 0.4750
Epoch 27/50
----------
train Loss: 0.7568 Acc: 0.6180
val Loss: 0.9697 Acc: 0.4924
test Loss: 0.8937 Acc: 0.5100
Epoch 28/50
----------
train Loss: 0.7692 Acc: 0.6270
val Loss: 0.9502 Acc: 0.4670
test Loss: 0.9323 Acc: 0.5100
Epoch 29/50
----------
train Loss: 0.7802 Acc: 0.6140
val Loss: 0.9800 Acc: 0.4670
test Loss: 0.9779 Acc: 0.4750
Epoch 30/50
----------
train Loss: 0.7504 Acc: 0.6390
val Loss: 0.9169 Acc: 0.5076
test Loss: 0.9288 Acc: 0.5100
Epoch 31/50
----------
train Loss: 0.7724 Acc: 0.6240
val Loss: 0.9116 Acc: 0.5482
test Loss: 0.9396 Acc: 0.5100
Epoch 32/50
----------
train Loss: 0.7525 Acc: 0.6290
val Loss: 0.9480 Acc: 0.5279
test Loss: 0.9757 Acc: 0.4550
Epoch 33/50
----------
train Loss: 0.7287 Acc: 0.6590
val Loss: 0.9929 Acc: 0.4619
test Loss: 0.9229 Acc: 0.5450
Epoch 34/50
----------
train Loss: 0.7599 Acc: 0.6070
val Loss: 0.9612 Acc: 0.5178
test Loss: 0.9624 Acc: 0.4300
Epoch 35/50
----------
train Loss: 0.7325 Acc: 0.6460
val Loss: 0.9414 Acc: 0.5330
test Loss: 0.9693 Acc: 0.4600
Epoch 36/50
----------
train Loss: 0.7492 Acc: 0.6260
val Loss: 0.9598 Acc: 0.5279
test Loss: 1.0015 Acc: 0.4800
Epoch 37/50
----------
train Loss: 0.7447 Acc: 0.6240
val Loss: 0.9932 Acc: 0.4619
test Loss: 1.0366 Acc: 0.4500
Epoch 38/50
----------
train Loss: 0.6880 Acc: 0.6690
val Loss: 1.0724 Acc: 0.5279
test Loss: 1.0444 Acc: 0.4650
Epoch 39/50
----------
train Loss: 0.7417 Acc: 0.6480
val Loss: 1.0000 Acc: 0.5330
test Loss: 0.9902 Acc: 0.5250
Epoch 40/50
----------
train Loss: 0.7345 Acc: 0.6420
val Loss: 0.9834 Acc: 0.5685
test Loss: 1.0127 Acc: 0.4700
Epoch 41/50
----------
train Loss: 0.6990 Acc: 0.6630
val Loss: 1.0695 Acc: 0.4822
test Loss: 0.9737 Acc: 0.5100
Epoch 42/50
----------
train Loss: 0.7158 Acc: 0.6560
val Loss: 0.9503 Acc: 0.5584
test Loss: 1.0307 Acc: 0.4550
Epoch 43/50
----------
train Loss: 0.7379 Acc: 0.6430
val Loss: 1.0447 Acc: 0.4822
test Loss: 1.0054 Acc: 0.4300
Epoch 44/50
----------
train Loss: 0.7236 Acc: 0.6620
val Loss: 0.9764 Acc: 0.5076
test Loss: 0.9995 Acc: 0.4900
Epoch 45/50
----------
train Loss: 0.6670 Acc: 0.6650
val Loss: 0.9997 Acc: 0.5025
test Loss: 1.1000 Acc: 0.4400
Epoch 46/50
----------
train Loss: 0.6976 Acc: 0.6470
val Loss: 0.9293 Acc: 0.5279
test Loss: 1.0334 Acc: 0.4600
Epoch 47/50
----------
train Loss: 0.7013 Acc: 0.6780
val Loss: 0.9315 Acc: 0.5178
test Loss: 0.9961 Acc: 0.4650
Epoch 48/50
----------
train Loss: 0.7115 Acc: 0.6680
val Loss: 0.9609 Acc: 0.5178
test Loss: 1.0162 Acc: 0.4850
Epoch 49/50
----------
train Loss: 0.6997 Acc: 0.6680
val Loss: 0.9317 Acc: 0.5228
test Loss: 0.9562 Acc: 0.5100
Epoch 50/50
----------
train Loss: 0.6155 Acc: 0.7090
val Loss: 1.0608 Acc: 0.4670
test Loss: 1.0940 Acc: 0.4450
Training complete in 59m 59s
Best val Acc: 0.568528
Datasets
Models
