实验数据_cnnlstm_mape
2006-03-04
pred:
array([7958.4155, 7751.268 , 7527.7954, 7201.982 , 6891.795 , 6662.3774,
6567.015 , 6512.1772, 6521.0806, 6697.8184, 6918.5425, 7376.8047,
7873.748 , 8257.072 , 8664.629 , 8909.608 , 9099.593 , 9343.218 ,
9468.084 , 9583.169 , 9606.54 , 9620.329 , 9640.951 , 9613.612 ,
9600.95 , 9607.041 , 9601.413 , 9596.466 , 9567.325 , 9542.329 ,
9573.12 , 9588.498 , 9589.878 , 9513.756 , 9474.52 , 9308.211 ,
9253.097 , 9176.314 , 9170.35 , 9147.754 , 9026.866 , 8784.027 ,
8507.148 , 8512.837 , 8331.485 , 8332.714 , 8179.672 , 8026.828 ],
dtype=float32)true:
array([7763.64667, 7520.06833, 7242.73833, 6931.72833, 6589.28167,
6371.48167, 6227.40667, 6159.62333, 6142.65167, 6245.355 ,
6329.72167, 6571.29833, 6780.075 , 7031.895 , 7411.87167,
7845.47333, 8172.615 , 8561.64667, 8656.84333, 8791.97833,
8778.925 , 8757.92833, 8746.56167, 8688.47 , 8686.71167,
8624.335 , 8611.68333, 8566.765 , 8546.89333, 8572.79 ,
8569.62333, 8589.01833, 8617.93 , 8454.73667, 8506.32167,
8423.125 , 8387.81833, 8230.435 , 8371.435 , 8415.43167,
8304.89667, 8093.65667, 7916.065 , 7885.44833, 7734.59333,
7808.35333, 7578.37667, 7347.17667])bule-forcesing
2010-03-04
331 Model: "model_5" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= input_6 (InputLayer) [(None, 48, 3)] 0 _________________________________________________________________ conv1d_9 (Conv1D) (None, 48, 128) 512 _________________________________________________________________ conv1d_10 (Conv1D) (None, 48, 128) 16512 _________________________________________________________________ max_pooling1d_4 (MaxPooling1 (None, 1, 128) 0 _________________________________________________________________ lstm_13 (LSTM) (None, 1, 128) 131584 _________________________________________________________________ lstm_14 (LSTM) (None, 1, 64) 49408 _________________________________________________________________ lstm_15 (LSTM) (None, 48) 21696 _________________________________________________________________ dense_5 (Dense) (None, 48) 2352 ================================================================= Total params: 222,064 Trainable params: 222,064 Non-trainable params: 0 _________________________________________________________________ Train on 331 samples, validate on 143 samples Epoch 1/30 331/331 [==============================] - 3s 10ms/sample - loss: 0.1349 - accuracy: 0.0997 - val_loss: 0.1059 - val_accuracy: 0.1189 Epoch 2/30 331/331 [==============================] - 0s 275us/sample - loss: 0.0722 - accuracy: 0.1178 - val_loss: 0.0305 - val_accuracy: 0.1189 Epoch 3/30 331/331 [==============================] - 0s 212us/sample - loss: 0.0206 - accuracy: 0.1178 - val_loss: 0.0081 - val_accuracy: 0.1189 Epoch 4/30 331/331 [==============================] - 0s 205us/sample - loss: 0.0068 - accuracy: 0.0544 - val_loss: 0.0051 - val_accuracy: 0.1538 Epoch 5/30 331/331 [==============================] - 0s 202us/sample - loss: 0.0051 - accuracy: 0.1027 - val_loss: 0.0047 - val_accuracy: 0.0490 Epoch 6/30 331/331 [==============================] - 0s 196us/sample - loss: 0.0048 - accuracy: 0.0272 - val_loss: 0.0044 - val_accuracy: 0.0350 Epoch 7/30 331/331 [==============================] - 0s 199us/sample - loss: 0.0046 - accuracy: 0.0151 - val_loss: 0.0043 - val_accuracy: 0.0140 Epoch 8/30 331/331 [==============================] - 0s 196us/sample - loss: 0.0045 - accuracy: 0.0423 - val_loss: 0.0043 - val_accuracy: 0.1189 Epoch 9/30 331/331 [==============================] - 0s 190us/sample - loss: 0.0045 - accuracy: 0.1057 - val_loss: 0.0043 - val_accuracy: 0.1189 Epoch 10/30 331/331 [==============================] - 0s 212us/sample - loss: 0.0045 - accuracy: 0.0665 - val_loss: 0.0043 - val_accuracy: 0.0699 Epoch 11/30 331/331 [==============================] - 0s 221us/sample - loss: 0.0045 - accuracy: 0.0785 - val_loss: 0.0043 - val_accuracy: 0.0699 Epoch 12/30 331/331 [==============================] - 0s 218us/sample - loss: 0.0045 - accuracy: 0.0785 - val_loss: 0.0043 - val_accuracy: 0.0699 Epoch 13/30 331/331 [==============================] - 0s 209us/sample - loss: 0.0045 - accuracy: 0.0785 - val_loss: 0.0043 - val_accuracy: 0.0699 Epoch 14/30 331/331 [==============================] - 0s 205us/sample - loss: 0.0045 - accuracy: 0.0785 - val_loss: 0.0043 - val_accuracy: 0.0699 Epoch 15/30 331/331 [==============================] - 0s 202us/sample - loss: 0.0045 - accuracy: 0.0785 - val_loss: 0.0043 - val_accuracy: 0.0699 Epoch 16/30 331/331 [==============================] - 0s 209us/sample - loss: 0.0045 - accuracy: 0.0785 - val_loss: 0.0043 - val_accuracy: 0.0699 Epoch 17/30 331/331 [==============================] - 0s 202us/sample - loss: 0.0045 - accuracy: 0.0785 - val_loss: 0.0043 - val_accuracy: 0.0699 Epoch 18/30 331/331 [==============================] - 0s 199us/sample - loss: 0.0045 - accuracy: 0.0785 - val_loss: 0.0043 - val_accuracy: 0.0699 Epoch 19/30 331/331 [==============================] - 0s 205us/sample - loss: 0.0045 - accuracy: 0.0785 - val_loss: 0.0043 - val_accuracy: 0.0699 Epoch 20/30 331/331 [==============================] - 0s 202us/sample - loss: 0.0045 - accuracy: 0.0785 - val_loss: 0.0043 - val_accuracy: 0.0699 Epoch 21/30 331/331 [==============================] - 0s 202us/sample - loss: 0.0045 - accuracy: 0.0785 - val_loss: 0.0043 - val_accuracy: 0.0699 Epoch 22/30 331/331 [==============================] - 0s 202us/sample - loss: 0.0045 - accuracy: 0.0785 - val_loss: 0.0043 - val_accuracy: 0.0699 Epoch 23/30 331/331 [==============================] - 0s 202us/sample - loss: 0.0045 - accuracy: 0.0785 - val_loss: 0.0043 - val_accuracy: 0.0699 Epoch 24/30 331/331 [==============================] - 0s 202us/sample - loss: 0.0045 - accuracy: 0.0785 - val_loss: 0.0043 - val_accuracy: 0.0699 Epoch 25/30 331/331 [==============================] - 0s 212us/sample - loss: 0.0045 - accuracy: 0.0785 - val_loss: 0.0043 - val_accuracy: 0.0699 Epoch 26/30 331/331 [==============================] - 0s 205us/sample - loss: 0.0045 - accuracy: 0.0574 - val_loss: 0.0043 - val_accuracy: 0.0699 Epoch 27/30 331/331 [==============================] - 0s 205us/sample - loss: 0.0045 - accuracy: 0.0574 - val_loss: 0.0043 - val_accuracy: 0.0699 Epoch 28/30 331/331 [==============================] - 0s 199us/sample - loss: 0.0045 - accuracy: 0.0574 - val_loss: 0.0043 - val_accuracy: 0.0699 Epoch 29/30 331/331 [==============================] - 0s 199us/sample - loss: 0.0045 - accuracy: 0.0574 - val_loss: 0.0043 - val_accuracy: 0.0699 Epoch 30/30 331/331 [==============================] - 0s 187us/sample - loss: 0.0045 - accuracy: 0.0574 - val_loss: 0.0043 - val_accuracy: 0.0699
pred [8010.635 7809.1787 7599.0825 7302.1855 6991.1035 6733.2344 6591.7183 6527.3687 6531.6743 6670.868 6870.7544 7306.865 7780.8086 8179.305 8601.343 8851.851 9067.619 9320.703 9440.045 9570.206 9602.607 9615.256 9646.539 9607.961 9620.904 9622.764 9601.794 9606.234 9587.057 9560.2705 9572.083 9583.597 9581.549 9531.673 9499.453 9366.619 9340.422 9293.003 9290.076 9223.878 9086.14 8861.275 8592.88 8551.889 8357.849 8352.641 8208.616 8078.3433] true [ 7723.29 7567.75 7305.22 7004.35 6772.37 6579.59 6498.34 6493.7 6575.3 6798.57 7057.49 7710.26 8257.28 8719.72 9137.33 9268.66 9383.15 9618.23 9762.4 9840.94 9900.17 9917.14 9956.33 9918.44 9973.25 10046.95 10062.29 10109.92 10086.19 10085.09 10105.69 10110.17 10110.66 9997.81 9930.54 9655.64 9558.58 9484.46 9598.29 9603.49 9384.25 9011.44 8689.19 8711.33 8509.65 8401.14 8252.09 8051.83]
2010-04-04
100 Model: "model_6" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= input_7 (InputLayer) [(None, 48, 3)] 0 _________________________________________________________________ conv1d_11 (Conv1D) (None, 48, 128) 512 _________________________________________________________________ conv1d_12 (Conv1D) (None, 48, 128) 16512 _________________________________________________________________ max_pooling1d_5 (MaxPooling1 (None, 1, 128) 0 _________________________________________________________________ lstm_16 (LSTM) (None, 1, 128) 131584 _________________________________________________________________ lstm_17 (LSTM) (None, 1, 64) 49408 _________________________________________________________________ lstm_18 (LSTM) (None, 48) 21696 _________________________________________________________________ dense_6 (Dense) (None, 48) 2352 ================================================================= Total params: 222,064 Trainable params: 222,064 Non-trainable params: 0 _________________________________________________________________ Train on 100 samples, validate on 43 samples Epoch 1/30 100/100 [==============================] - 4s 36ms/sample - loss: 0.1144 - accuracy: 0.0100 - val_loss: 0.0954 - val_accuracy: 0.0698 Epoch 2/30 100/100 [==============================] - 0s 440us/sample - loss: 0.1058 - accuracy: 0.0200 - val_loss: 0.0833 - val_accuracy: 0.0698 Epoch 3/30 100/100 [==============================] - 0s 450us/sample - loss: 0.0906 - accuracy: 0.0200 - val_loss: 0.0631 - val_accuracy: 0.0698 Epoch 4/30 100/100 [==============================] - 0s 420us/sample - loss: 0.0664 - accuracy: 0.0200 - val_loss: 0.0397 - val_accuracy: 0.0698 Epoch 5/30 100/100 [==============================] - 0s 350us/sample - loss: 0.0419 - accuracy: 0.0200 - val_loss: 0.0289 - val_accuracy: 0.0698 Epoch 6/30 100/100 [==============================] - 0s 310us/sample - loss: 0.0301 - accuracy: 0.0200 - val_loss: 0.0222 - val_accuracy: 0.0233 Epoch 7/30 100/100 [==============================] - 0s 330us/sample - loss: 0.0209 - accuracy: 0.0000e+00 - val_loss: 0.0128 - val_accuracy: 0.0233 Epoch 8/30 100/100 [==============================] - 0s 300us/sample - loss: 0.0133 - accuracy: 0.2000 - val_loss: 0.0085 - val_accuracy: 0.1628 Epoch 9/30 100/100 [==============================] - 0s 320us/sample - loss: 0.0104 - accuracy: 0.2700 - val_loss: 0.0070 - val_accuracy: 0.1628 Epoch 10/30 100/100 [==============================] - 0s 310us/sample - loss: 0.0088 - accuracy: 0.2100 - val_loss: 0.0067 - val_accuracy: 0.0465 Epoch 11/30 100/100 [==============================] - 0s 300us/sample - loss: 0.0079 - accuracy: 0.0700 - val_loss: 0.0070 - val_accuracy: 0.0233 Epoch 12/30 100/100 [==============================] - 0s 290us/sample - loss: 0.0078 - accuracy: 0.0400 - val_loss: 0.0068 - val_accuracy: 0.0233 Epoch 13/30 100/100 [==============================] - 0s 290us/sample - loss: 0.0077 - accuracy: 0.0400 - val_loss: 0.0065 - val_accuracy: 0.0000e+00 Epoch 14/30 100/100 [==============================] - 0s 290us/sample - loss: 0.0077 - accuracy: 0.0200 - val_loss: 0.0063 - val_accuracy: 0.0000e+00 Epoch 15/30 100/100 [==============================] - 0s 280us/sample - loss: 0.0076 - accuracy: 0.0300 - val_loss: 0.0063 - val_accuracy: 0.1860 Epoch 16/30 100/100 [==============================] - 0s 280us/sample - loss: 0.0075 - accuracy: 0.1700 - val_loss: 0.0063 - val_accuracy: 0.1628 Epoch 17/30 100/100 [==============================] - 0s 310us/sample - loss: 0.0074 - accuracy: 0.2700 - val_loss: 0.0062 - val_accuracy: 0.1628 Epoch 18/30 100/100 [==============================] - 0s 300us/sample - loss: 0.0073 - accuracy: 0.1600 - val_loss: 0.0061 - val_accuracy: 0.1860 Epoch 19/30 100/100 [==============================] - 0s 300us/sample - loss: 0.0073 - accuracy: 0.2300 - val_loss: 0.0060 - val_accuracy: 0.1628 Epoch 20/30 100/100 [==============================] - 0s 300us/sample - loss: 0.0073 - accuracy: 0.2700 - val_loss: 0.0060 - val_accuracy: 0.1628 Epoch 21/30 100/100 [==============================] - 0s 310us/sample - loss: 0.0073 - accuracy: 0.2700 - val_loss: 0.0061 - val_accuracy: 0.1628 Epoch 22/30 100/100 [==============================] - 0s 290us/sample - loss: 0.0073 - accuracy: 0.2700 - val_loss: 0.0061 - val_accuracy: 0.1628 Epoch 23/30 100/100 [==============================] - 0s 300us/sample - loss: 0.0073 - accuracy: 0.2700 - val_loss: 0.0060 - val_accuracy: 0.1628 Epoch 24/30 100/100 [==============================] - 0s 290us/sample - loss: 0.0073 - accuracy: 0.2700 - val_loss: 0.0060 - val_accuracy: 0.1628 Epoch 25/30 100/100 [==============================] - 0s 290us/sample - loss: 0.0073 - accuracy: 0.2700 - val_loss: 0.0060 - val_accuracy: 0.1628 Epoch 26/30 100/100 [==============================] - 0s 280us/sample - loss: 0.0073 - accuracy: 0.2700 - val_loss: 0.0061 - val_accuracy: 0.1628 Epoch 27/30 100/100 [==============================] - 0s 310us/sample - loss: 0.0073 - accuracy: 0.2700 - val_loss: 0.0060 - val_accuracy: 0.1628 Epoch 28/30 100/100 [==============================] - 0s 290us/sample - loss: 0.0073 - accuracy: 0.2700 - val_loss: 0.0060 - val_accuracy: 0.1628 Epoch 29/30 100/100 [==============================] - 0s 300us/sample - loss: 0.0073 - accuracy: 0.2700 - val_loss: 0.0060 - val_accuracy: 0.1628 Epoch 30/30 100/100 [==============================] - 0s 280us/sample - loss: 0.0073 - accuracy: 0.2700 - val_loss: 0.0061 - val_accuracy: 0.1628
pred [7349.497 7179.998 7020.6714 6816.061 6613.731 6393.7188 6249.8213 6175.4307 6158.613 6246.8555 6362.7847 6669.811 6999.348 7368.3564 7751.2993 8045.674 8360.4 8637.023 8807.22 8977.349 9048.192 9079.813 9111.7295 9080.084 9068.504 9079.433 9045.489 9030.003 8980.91 8978.036 8964.549 8970.938 8978.405 8977.747 9007.412 8994.505 9147.034 9234.686 9187.403 9044.471 8913.489 8701.031 8471.27 8348.808 8121. 8157.4746 8010.4307 7913.78 ] true [7044.52 6840.32 6683.7 6475.78 6243.97 6047.35 5957.95 5921.97 5890.06 5893.66 5925.86 6033.7 6097.41 6290.21 6567.28 6901.81 7219.55 7515.17 7723.48 7855.98 7920.14 7902.77 7929.26 7823.64 7844.46 7776.33 7739.9 7615.65 7533.99 7493.34 7416.12 7454.71 7473.31 7505.29 7565.4 7641.99 7884.96 8240.28 8210.25 7974.3 7811.36 7668.71 7614.29 7492.89 7400.21 7369.07 7394.38 7306.22]
2010
12-02
256 Model: "model_7" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= input_8 (InputLayer) [(None, 48, 3)] 0 _________________________________________________________________ conv1d_13 (Conv1D) (None, 48, 128) 512 _________________________________________________________________ conv1d_14 (Conv1D) (None, 48, 128) 16512 _________________________________________________________________ max_pooling1d_6 (MaxPooling1 (None, 1, 128) 0 _________________________________________________________________ lstm_19 (LSTM) (None, 1, 128) 131584 _________________________________________________________________ lstm_20 (LSTM) (None, 1, 64) 49408 _________________________________________________________________ lstm_21 (LSTM) (None, 48) 21696 _________________________________________________________________ dense_7 (Dense) (None, 48) 2352 ================================================================= Total params: 222,064 Trainable params: 222,064 Non-trainable params: 0 _________________________________________________________________ Train on 256 samples, validate on 110 samples Epoch 1/30 256/256 [==============================] - 4s 14ms/sample - loss: 0.1284 - accuracy: 0.0078 - val_loss: 0.1102 - val_accuracy: 0.0091 Epoch 2/30 256/256 [==============================] - 0s 242us/sample - loss: 0.0958 - accuracy: 0.0078 - val_loss: 0.0600 - val_accuracy: 0.0091 Epoch 3/30 256/256 [==============================] - 0s 231us/sample - loss: 0.0428 - accuracy: 0.0078 - val_loss: 0.0253 - val_accuracy: 0.0091 Epoch 4/30 256/256 [==============================] - 0s 203us/sample - loss: 0.0182 - accuracy: 0.0078 - val_loss: 0.0091 - val_accuracy: 0.0091 Epoch 5/30 256/256 [==============================] - 0s 199us/sample - loss: 0.0072 - accuracy: 0.0078 - val_loss: 0.0048 - val_accuracy: 0.0182 Epoch 6/30 256/256 [==============================] - 0s 199us/sample - loss: 0.0052 - accuracy: 0.0195 - val_loss: 0.0045 - val_accuracy: 0.0182 Epoch 7/30 256/256 [==============================] - 0s 215us/sample - loss: 0.0048 - accuracy: 0.0234 - val_loss: 0.0041 - val_accuracy: 0.0364 Epoch 8/30 256/256 [==============================] - 0s 207us/sample - loss: 0.0045 - accuracy: 0.0117 - val_loss: 0.0038 - val_accuracy: 0.0364 Epoch 9/30 256/256 [==============================] - 0s 211us/sample - loss: 0.0043 - accuracy: 0.1055 - val_loss: 0.0038 - val_accuracy: 0.1364 Epoch 10/30 256/256 [==============================] - 0s 195us/sample - loss: 0.0042 - accuracy: 0.0977 - val_loss: 0.0037 - val_accuracy: 0.0909 Epoch 11/30 256/256 [==============================] - 0s 188us/sample - loss: 0.0042 - accuracy: 0.0781 - val_loss: 0.0037 - val_accuracy: 0.0182 Epoch 12/30 256/256 [==============================] - 0s 188us/sample - loss: 0.0042 - accuracy: 0.0664 - val_loss: 0.0037 - val_accuracy: 0.0909 Epoch 13/30 256/256 [==============================] - 0s 203us/sample - loss: 0.0042 - accuracy: 0.0586 - val_loss: 0.0037 - val_accuracy: 0.0364 Epoch 14/30 256/256 [==============================] - 0s 207us/sample - loss: 0.0042 - accuracy: 0.0312 - val_loss: 0.0037 - val_accuracy: 0.0364 Epoch 15/30 256/256 [==============================] - 0s 219us/sample - loss: 0.0042 - accuracy: 0.0625 - val_loss: 0.0037 - val_accuracy: 0.0909 Epoch 16/30 256/256 [==============================] - 0s 203us/sample - loss: 0.0042 - accuracy: 0.0781 - val_loss: 0.0037 - val_accuracy: 0.0909 Epoch 17/30 256/256 [==============================] - 0s 207us/sample - loss: 0.0042 - accuracy: 0.0781 - val_loss: 0.0037 - val_accuracy: 0.0909 Epoch 18/30 256/256 [==============================] - 0s 199us/sample - loss: 0.0042 - accuracy: 0.0781 - val_loss: 0.0037 - val_accuracy: 0.0909 Epoch 19/30 256/256 [==============================] - 0s 199us/sample - loss: 0.0042 - accuracy: 0.0781 - val_loss: 0.0037 - val_accuracy: 0.0909 Epoch 20/30 256/256 [==============================] - 0s 199us/sample - loss: 0.0042 - accuracy: 0.0781 - val_loss: 0.0037 - val_accuracy: 0.0909 Epoch 21/30 256/256 [==============================] - 0s 203us/sample - loss: 0.0042 - accuracy: 0.0781 - val_loss: 0.0037 - val_accuracy: 0.0909 Epoch 22/30 256/256 [==============================] - 0s 199us/sample - loss: 0.0042 - accuracy: 0.0781 - val_loss: 0.0037 - val_accuracy: 0.0909 Epoch 23/30 256/256 [==============================] - 0s 203us/sample - loss: 0.0042 - accuracy: 0.0781 - val_loss: 0.0037 - val_accuracy: 0.0909 Epoch 24/30 256/256 [==============================] - 0s 199us/sample - loss: 0.0042 - accuracy: 0.0781 - val_loss: 0.0037 - val_accuracy: 0.0909 Epoch 25/30 256/256 [==============================] - 0s 215us/sample - loss: 0.0042 - accuracy: 0.0781 - val_loss: 0.0037 - val_accuracy: 0.0909 Epoch 26/30 256/256 [==============================] - 0s 207us/sample - loss: 0.0042 - accuracy: 0.0781 - val_loss: 0.0037 - val_accuracy: 0.0909 Epoch 27/30 256/256 [==============================] - 0s 195us/sample - loss: 0.0042 - accuracy: 0.0781 - val_loss: 0.0037 - val_accuracy: 0.0909 Epoch 28/30 256/256 [==============================] - 0s 184us/sample - loss: 0.0042 - accuracy: 0.0781 - val_loss: 0.0037 - val_accuracy: 0.0909 Epoch 29/30 256/256 [==============================] - 0s 199us/sample - loss: 0.0042 - accuracy: 0.0781 - val_loss: 0.0037 - val_accuracy: 0.0909 Epoch 30/30 256/256 [==============================] - 0s 199us/sample - loss: 0.0042 - accuracy: 0.0781 - val_loss: 0.0037 - val_accuracy: 0.0909
pred [7923.696 7740.4424 7512.9272 7214.1787 6914.0386 6676.3384 6528.638 6453.856 6466.6724 6612.4575 6805.5674 7223.514 7664.2925 8048.367 8458.108 8704.794 8927.65 9173.978 9301.087 9425.839 9456.668 9467.464 9475.861 9464.989 9450.09 9445.564 9436.48 9437.943 9406.843 9396.314 9406.29 9415.468 9424.467 9367.473 9346.866 9223.6045 9201.393 9156.106 9155.85 9101.848 8980.621 8743.551 8489.39 8446.115 8266.466 8273.215 8120.446 7979.259 ] true [7656.58 7479.25 7253.48 6997.73 6697.31 6495.14 6385.37 6388.82 6469.15 6701.75 6914.02 7488.23 8049.46 8442.19 8901.89 9133.85 9224. 9456.13 9599.82 9696.93 9716.15 9683.64 9701.91 9698.27 9640.22 9691.94 9685.57 9730.46 9666.57 9489.31 9536.09 9513.78 9505.3 9472.51 9471.54 9214.27 9103.36 8967.74 8939.74 9037.84 8926.12 8672.38 8349.17 8373.84 8233.03 8222.77 8067.77 7873.98]
12-03
338 Model: "model_8" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= input_9 (InputLayer) [(None, 48, 3)] 0 _________________________________________________________________ conv1d_15 (Conv1D) (None, 48, 128) 512 _________________________________________________________________ conv1d_16 (Conv1D) (None, 48, 128) 16512 _________________________________________________________________ max_pooling1d_7 (MaxPooling1 (None, 1, 128) 0 _________________________________________________________________ lstm_22 (LSTM) (None, 1, 128) 131584 _________________________________________________________________ lstm_23 (LSTM) (None, 1, 64) 49408 _________________________________________________________________ lstm_24 (LSTM) (None, 48) 21696 _________________________________________________________________ dense_8 (Dense) (None, 48) 2352 ================================================================= Total params: 222,064 Trainable params: 222,064 Non-trainable params: 0 _________________________________________________________________ Train on 338 samples, validate on 145 samples Epoch 1/30 338/338 [==============================] - 4s 12ms/sample - loss: 0.1289 - accuracy: 0.0592 - val_loss: 0.0994 - val_accuracy: 0.0897 Epoch 2/30 338/338 [==============================] - 0s 243us/sample - loss: 0.0673 - accuracy: 0.0828 - val_loss: 0.0326 - val_accuracy: 0.0897 Epoch 3/30 338/338 [==============================] - 0s 284us/sample - loss: 0.0202 - accuracy: 0.0799 - val_loss: 0.0092 - val_accuracy: 0.0069 Epoch 4/30 338/338 [==============================] - 0s 210us/sample - loss: 0.0071 - accuracy: 0.0533 - val_loss: 0.0050 - val_accuracy: 0.1379 Epoch 5/30 338/338 [==============================] - 0s 183us/sample - loss: 0.0049 - accuracy: 0.1479 - val_loss: 0.0047 - val_accuracy: 0.0138 Epoch 6/30 338/338 [==============================] - 0s 195us/sample - loss: 0.0046 - accuracy: 0.0473 - val_loss: 0.0045 - val_accuracy: 0.0690 Epoch 7/30 338/338 [==============================] - 0s 186us/sample - loss: 0.0043 - accuracy: 0.0799 - val_loss: 0.0043 - val_accuracy: 0.0690 Epoch 8/30 338/338 [==============================] - 0s 186us/sample - loss: 0.0043 - accuracy: 0.0266 - val_loss: 0.0043 - val_accuracy: 0.0276 Epoch 9/30 338/338 [==============================] - 0s 207us/sample - loss: 0.0043 - accuracy: 0.0118 - val_loss: 0.0043 - val_accuracy: 0.0138 Epoch 10/30 338/338 [==============================] - 0s 210us/sample - loss: 0.0043 - accuracy: 0.0266 - val_loss: 0.0043 - val_accuracy: 0.0690 Epoch 11/30 338/338 [==============================] - 0s 204us/sample - loss: 0.0043 - accuracy: 0.0799 - val_loss: 0.0043 - val_accuracy: 0.0690 Epoch 12/30 338/338 [==============================] - 0s 207us/sample - loss: 0.0043 - accuracy: 0.0799 - val_loss: 0.0043 - val_accuracy: 0.0690 Epoch 13/30 338/338 [==============================] - 0s 201us/sample - loss: 0.0043 - accuracy: 0.0799 - val_loss: 0.0043 - val_accuracy: 0.0690 Epoch 14/30 338/338 [==============================] - 0s 204us/sample - loss: 0.0043 - accuracy: 0.0799 - val_loss: 0.0043 - val_accuracy: 0.0690 Epoch 15/30 338/338 [==============================] - 0s 195us/sample - loss: 0.0043 - accuracy: 0.0799 - val_loss: 0.0043 - val_accuracy: 0.0690 Epoch 16/30 338/338 [==============================] - 0s 210us/sample - loss: 0.0043 - accuracy: 0.0799 - val_loss: 0.0043 - val_accuracy: 0.0690 Epoch 17/30 338/338 [==============================] - 0s 204us/sample - loss: 0.0043 - accuracy: 0.0799 - val_loss: 0.0043 - val_accuracy: 0.0690 Epoch 18/30 338/338 [==============================] - 0s 201us/sample - loss: 0.0043 - accuracy: 0.0799 - val_loss: 0.0043 - val_accuracy: 0.0690 Epoch 19/30 338/338 [==============================] - 0s 216us/sample - loss: 0.0043 - accuracy: 0.0799 - val_loss: 0.0043 - val_accuracy: 0.0690 Epoch 20/30 338/338 [==============================] - 0s 222us/sample - loss: 0.0043 - accuracy: 0.0799 - val_loss: 0.0043 - val_accuracy: 0.0690 Epoch 21/30 338/338 [==============================] - 0s 210us/sample - loss: 0.0043 - accuracy: 0.0799 - val_loss: 0.0043 - val_accuracy: 0.0690 Epoch 22/30 338/338 [==============================] - 0s 207us/sample - loss: 0.0043 - accuracy: 0.0799 - val_loss: 0.0043 - val_accuracy: 0.0690 Epoch 23/30 338/338 [==============================] - 0s 204us/sample - loss: 0.0043 - accuracy: 0.0799 - val_loss: 0.0043 - val_accuracy: 0.0690 Epoch 24/30 338/338 [==============================] - 0s 204us/sample - loss: 0.0043 - accuracy: 0.0799 - val_loss: 0.0043 - val_accuracy: 0.0690 Epoch 25/30 338/338 [==============================] - 0s 204us/sample - loss: 0.0043 - accuracy: 0.0799 - val_loss: 0.0043 - val_accuracy: 0.0690 Epoch 26/30 338/338 [==============================] - 0s 207us/sample - loss: 0.0043 - accuracy: 0.0799 - val_loss: 0.0043 - val_accuracy: 0.0690 Epoch 27/30 338/338 [==============================] - 0s 207us/sample - loss: 0.0043 - accuracy: 0.0799 - val_loss: 0.0043 - val_accuracy: 0.0690 Epoch 28/30 338/338 [==============================] - 0s 204us/sample - loss: 0.0043 - accuracy: 0.0799 - val_loss: 0.0043 - val_accuracy: 0.0690 Epoch 29/30 338/338 [==============================] - 0s 201us/sample - loss: 0.0043 - accuracy: 0.0799 - val_loss: 0.0043 - val_accuracy: 0.0690 Epoch 30/30 338/338 [==============================] - 0s 207us/sample - loss: 0.0043 - accuracy: 0.0799 - val_loss: 0.0043 - val_accuracy: 0.0690
pred [8029.077 7832.7837 7621.4766 7335.655 7043.4087 6779.6904 6602.6035 6527.119 6526.795 6643.554 6819.8306 7246.94 7681.114 8106.089 8522.493 8767.697 9014.915 9283.784 9399.581 9527.989 9572.119 9575.893 9587.795 9572.996 9570.831 9570.688 9558.398 9551.962 9541.78 9503.843 9522.485 9532.224 9535.44 9463.853 9446.416 9334.089 9329.337 9317.327 9303.091 9227.795 9094.39 8865.05 8602.331 8546.524 8357.236 8373.456 8222.707 8099.8384] true [ 7699.36 7537.08 7392.4 7066.04 6756.79 6593.92 6474.21 6476.41 6564.8 6786.13 7053.51 7583.67 8147.22 8321.58 8765.5 9016.14 9362.55 9787.3 9932.43 10048.65 10067.54 10063.7 10123.17 10090.39 10065.33 10045.28 10056.09 10027.39 10000.66 9980.53 9951.47 9906.39 9815.59 9628.56 9459.65 9173.19 8980.21 8857.04 8848.53 8871.45 8792.61 8551.15 8375.27 8451.91 8385.28 8377.46 8299.01 8014.12]
12-04
296 Model: "model_9" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= input_10 (InputLayer) [(None, 48, 3)] 0 _________________________________________________________________ conv1d_17 (Conv1D) (None, 48, 128) 512 _________________________________________________________________ conv1d_18 (Conv1D) (None, 48, 128) 16512 _________________________________________________________________ max_pooling1d_8 (MaxPooling1 (None, 1, 128) 0 _________________________________________________________________ lstm_25 (LSTM) (None, 1, 128) 131584 _________________________________________________________________ lstm_26 (LSTM) (None, 1, 64) 49408 _________________________________________________________________ lstm_27 (LSTM) (None, 48) 21696 _________________________________________________________________ dense_9 (Dense) (None, 48) 2352 ================================================================= Total params: 222,064 Trainable params: 222,064 Non-trainable params: 0 _________________________________________________________________ Train on 296 samples, validate on 128 samples Epoch 1/30 296/296 [==============================] - 4s 13ms/sample - loss: 0.1391 - accuracy: 0.0304 - val_loss: 0.1114 - val_accuracy: 0.0547 Epoch 2/30 296/296 [==============================] - 0s 186us/sample - loss: 0.0845 - accuracy: 0.0372 - val_loss: 0.0381 - val_accuracy: 0.0547 Epoch 3/30 296/296 [==============================] - 0s 182us/sample - loss: 0.0287 - accuracy: 0.0372 - val_loss: 0.0139 - val_accuracy: 0.0547 Epoch 4/30 296/296 [==============================] - 0s 186us/sample - loss: 0.0098 - accuracy: 0.0372 - val_loss: 0.0065 - val_accuracy: 0.0000e+00 Epoch 5/30 296/296 [==============================] - 0s 182us/sample - loss: 0.0058 - accuracy: 0.0236 - val_loss: 0.0049 - val_accuracy: 0.0156 Epoch 6/30 296/296 [==============================] - 0s 182us/sample - loss: 0.0049 - accuracy: 0.1520 - val_loss: 0.0045 - val_accuracy: 0.1875 Epoch 7/30 296/296 [==============================] - 0s 182us/sample - loss: 0.0048 - accuracy: 0.1318 - val_loss: 0.0045 - val_accuracy: 0.0078 Epoch 8/30 296/296 [==============================] - 0s 182us/sample - loss: 0.0047 - accuracy: 0.0135 - val_loss: 0.0043 - val_accuracy: 0.0312 Epoch 9/30 296/296 [==============================] - 0s 182us/sample - loss: 0.0046 - accuracy: 0.0743 - val_loss: 0.0043 - val_accuracy: 0.0625 Epoch 10/30 296/296 [==============================] - 0s 182us/sample - loss: 0.0046 - accuracy: 0.0811 - val_loss: 0.0043 - val_accuracy: 0.0547 Epoch 11/30 296/296 [==============================] - 0s 182us/sample - loss: 0.0046 - accuracy: 0.0372 - val_loss: 0.0043 - val_accuracy: 0.0156 Epoch 12/30 296/296 [==============================] - 0s 182us/sample - loss: 0.0046 - accuracy: 0.0878 - val_loss: 0.0043 - val_accuracy: 0.0625 Epoch 13/30 296/296 [==============================] - 0s 179us/sample - loss: 0.0046 - accuracy: 0.1351 - val_loss: 0.0043 - val_accuracy: 0.0625 Epoch 14/30 296/296 [==============================] - 0s 186us/sample - loss: 0.0046 - accuracy: 0.1419 - val_loss: 0.0043 - val_accuracy: 0.0625 Epoch 15/30 296/296 [==============================] - 0s 182us/sample - loss: 0.0046 - accuracy: 0.1486 - val_loss: 0.0043 - val_accuracy: 0.0625 Epoch 16/30 296/296 [==============================] - 0s 182us/sample - loss: 0.0046 - accuracy: 0.1385 - val_loss: 0.0043 - val_accuracy: 0.0625 Epoch 17/30 296/296 [==============================] - 0s 182us/sample - loss: 0.0046 - accuracy: 0.1385 - val_loss: 0.0043 - val_accuracy: 0.0625 Epoch 18/30 296/296 [==============================] - 0s 186us/sample - loss: 0.0046 - accuracy: 0.1385 - val_loss: 0.0043 - val_accuracy: 0.0625 Epoch 19/30 296/296 [==============================] - 0s 196us/sample - loss: 0.0046 - accuracy: 0.1385 - val_loss: 0.0043 - val_accuracy: 0.0625 Epoch 20/30 296/296 [==============================] - 0s 189us/sample - loss: 0.0046 - accuracy: 0.1385 - val_loss: 0.0043 - val_accuracy: 0.0625 Epoch 21/30 296/296 [==============================] - 0s 186us/sample - loss: 0.0046 - accuracy: 0.1385 - val_loss: 0.0043 - val_accuracy: 0.0625 Epoch 22/30 296/296 [==============================] - 0s 182us/sample - loss: 0.0046 - accuracy: 0.1385 - val_loss: 0.0043 - val_accuracy: 0.0625 Epoch 23/30 296/296 [==============================] - 0s 186us/sample - loss: 0.0046 - accuracy: 0.1385 - val_loss: 0.0043 - val_accuracy: 0.0625 Epoch 24/30 296/296 [==============================] - 0s 189us/sample - loss: 0.0046 - accuracy: 0.1385 - val_loss: 0.0043 - val_accuracy: 0.0625 Epoch 25/30 296/296 [==============================] - 0s 186us/sample - loss: 0.0046 - accuracy: 0.1385 - val_loss: 0.0043 - val_accuracy: 0.0625 Epoch 26/30 296/296 [==============================] - 0s 210us/sample - loss: 0.0046 - accuracy: 0.1385 - val_loss: 0.0043 - val_accuracy: 0.0625 Epoch 27/30 296/296 [==============================] - 0s 186us/sample - loss: 0.0046 - accuracy: 0.1385 - val_loss: 0.0043 - val_accuracy: 0.0625 Epoch 28/30 296/296 [==============================] - 0s 189us/sample - loss: 0.0046 - accuracy: 0.1385 - val_loss: 0.0043 - val_accuracy: 0.0625 Epoch 29/30 296/296 [==============================] - 0s 186us/sample - loss: 0.0046 - accuracy: 0.1385 - val_loss: 0.0043 - val_accuracy: 0.0625 Epoch 30/30 296/296 [==============================] - 0s 182us/sample - loss: 0.0046 - accuracy: 0.1318 - val_loss: 0.0043 - val_accuracy: 0.0625
pred [7958.0576 7759.34 7539.0767 7212.7817 6907.1313 6678.9155 6552.6836 6511.8564 6528.0137 6714.446 6925.062 7394.0566 7890.836 8268.435 8681.149 8906.056 9120.451 9388.612 9510.767 9614.055 9668.039 9675.767 9709.273 9691.607 9687.969 9700.235 9685.791 9709.01 9681.103 9660.03 9686.174 9686.691 9696.266 9629.146 9574.18 9401.561 9327.356 9242.614 9240.816 9213.33 9083.897 8838.83 8564.487 8539.373 8357.995 8344.04 8199.021 8029.7124] true [7855.8 7695.78 7386.18 7085.98 6814.34 6615.06 6463.05 6418.7 6422.93 6515.57 6520.47 6708.32 7025.89 7360.76 7733.97 8128.16 8451.18 8810.48 8874.04 8992.48 8987.8 8916.44 8932.77 8875.14 8795.33 8741.9 8755.06 8739.13 8734.63 8716.26 8664.24 8683.48 8658.94 8507.08 8543.99 8514.78 8485.41 8352.18 8330.75 8346.79 8305.49 8117.21 7998.15 7955.43 7823.99 7858.36 7662.11 7422.66]
12-05
131 Model: "model_10" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= input_11 (InputLayer) [(None, 48, 3)] 0 _________________________________________________________________ conv1d_19 (Conv1D) (None, 48, 128) 512 _________________________________________________________________ conv1d_20 (Conv1D) (None, 48, 128) 16512 _________________________________________________________________ max_pooling1d_9 (MaxPooling1 (None, 1, 128) 0 _________________________________________________________________ lstm_28 (LSTM) (None, 1, 128) 131584 _________________________________________________________________ lstm_29 (LSTM) (None, 1, 64) 49408 _________________________________________________________________ lstm_30 (LSTM) (None, 48) 21696 _________________________________________________________________ dense_10 (Dense) (None, 48) 2352 ================================================================= Total params: 222,064 Trainable params: 222,064 Non-trainable params: 0 _________________________________________________________________ Train on 131 samples, validate on 57 samples Epoch 1/30 131/131 [==============================] - 4s 29ms/sample - loss: 0.1047 - accuracy: 0.0000e+00 - val_loss: 0.0989 - val_accuracy: 0.0175 Epoch 2/30 131/131 [==============================] - 0s 275us/sample - loss: 0.0932 - accuracy: 0.0153 - val_loss: 0.0816 - val_accuracy: 0.0175 Epoch 3/30 131/131 [==============================] - 0s 290us/sample - loss: 0.0726 - accuracy: 0.0153 - val_loss: 0.0552 - val_accuracy: 0.0175 Epoch 4/30 131/131 [==============================] - 0s 252us/sample - loss: 0.0469 - accuracy: 0.0153 - val_loss: 0.0359 - val_accuracy: 0.0175 Epoch 5/30 131/131 [==============================] - 0s 260us/sample - loss: 0.0310 - accuracy: 0.0153 - val_loss: 0.0230 - val_accuracy: 0.0175 Epoch 6/30 131/131 [==============================] - 0s 275us/sample - loss: 0.0186 - accuracy: 0.0305 - val_loss: 0.0138 - val_accuracy: 0.1228 Epoch 7/30 131/131 [==============================] - 0s 283us/sample - loss: 0.0115 - accuracy: 0.1069 - val_loss: 0.0101 - val_accuracy: 0.1228 Epoch 8/30 131/131 [==============================] - 0s 275us/sample - loss: 0.0085 - accuracy: 0.1069 - val_loss: 0.0083 - val_accuracy: 0.0000e+00 Epoch 9/30 131/131 [==============================] - 0s 275us/sample - loss: 0.0072 - accuracy: 0.0229 - val_loss: 0.0078 - val_accuracy: 0.0000e+00 Epoch 10/30 131/131 [==============================] - 0s 260us/sample - loss: 0.0069 - accuracy: 0.0611 - val_loss: 0.0077 - val_accuracy: 0.0877 Epoch 11/30 131/131 [==============================] - 0s 252us/sample - loss: 0.0069 - accuracy: 0.1221 - val_loss: 0.0077 - val_accuracy: 0.0702 Epoch 12/30 131/131 [==============================] - 0s 260us/sample - loss: 0.0068 - accuracy: 0.0611 - val_loss: 0.0075 - val_accuracy: 0.0702 Epoch 13/30 131/131 [==============================] - 0s 260us/sample - loss: 0.0067 - accuracy: 0.0611 - val_loss: 0.0074 - val_accuracy: 0.2105 Epoch 14/30 131/131 [==============================] - 0s 267us/sample - loss: 0.0065 - accuracy: 0.0687 - val_loss: 0.0074 - val_accuracy: 0.0351 Epoch 15/30 131/131 [==============================] - 0s 267us/sample - loss: 0.0065 - accuracy: 0.0153 - val_loss: 0.0074 - val_accuracy: 0.0351 Epoch 16/30 131/131 [==============================] - 0s 275us/sample - loss: 0.0065 - accuracy: 0.0992 - val_loss: 0.0073 - val_accuracy: 0.0877 Epoch 17/30 131/131 [==============================] - 0s 283us/sample - loss: 0.0065 - accuracy: 0.1298 - val_loss: 0.0073 - val_accuracy: 0.0877 Epoch 18/30 131/131 [==============================] - 0s 283us/sample - loss: 0.0065 - accuracy: 0.1298 - val_loss: 0.0073 - val_accuracy: 0.0877 Epoch 19/30 131/131 [==============================] - 0s 260us/sample - loss: 0.0065 - accuracy: 0.1298 - val_loss: 0.0073 - val_accuracy: 0.0877 Epoch 20/30 131/131 [==============================] - 0s 267us/sample - loss: 0.0065 - accuracy: 0.1298 - val_loss: 0.0073 - val_accuracy: 0.0877 Epoch 21/30 131/131 [==============================] - 0s 252us/sample - loss: 0.0065 - accuracy: 0.1298 - val_loss: 0.0073 - val_accuracy: 0.0877 Epoch 22/30 131/131 [==============================] - 0s 260us/sample - loss: 0.0065 - accuracy: 0.1069 - val_loss: 0.0073 - val_accuracy: 0.0877 Epoch 23/30 131/131 [==============================] - 0s 252us/sample - loss: 0.0065 - accuracy: 0.1069 - val_loss: 0.0073 - val_accuracy: 0.0877 Epoch 24/30 131/131 [==============================] - 0s 252us/sample - loss: 0.0065 - accuracy: 0.1069 - val_loss: 0.0073 - val_accuracy: 0.0877 Epoch 25/30 131/131 [==============================] - 0s 237us/sample - loss: 0.0065 - accuracy: 0.1298 - val_loss: 0.0073 - val_accuracy: 0.0877 Epoch 26/30 131/131 [==============================] - 0s 244us/sample - loss: 0.0065 - accuracy: 0.1298 - val_loss: 0.0073 - val_accuracy: 0.0877 Epoch 27/30 131/131 [==============================] - 0s 260us/sample - loss: 0.0065 - accuracy: 0.1298 - val_loss: 0.0073 - val_accuracy: 0.0877 Epoch 28/30 131/131 [==============================] - 0s 244us/sample - loss: 0.0065 - accuracy: 0.1298 - val_loss: 0.0073 - val_accuracy: 0.0877 Epoch 29/30 131/131 [==============================] - 0s 267us/sample - loss: 0.0065 - accuracy: 0.1298 - val_loss: 0.0073 - val_accuracy: 0.0877 Epoch 30/30 131/131 [==============================] - 0s 260us/sample - loss: 0.0065 - accuracy: 0.1298 - val_loss: 0.0073 - val_accuracy: 0.0877
pred [7468.0015 7285.739 7111.7686 6854.704 6618.753 6416.906 6289.9585 6213.242 6205.834 6275.3574 6364.411 6588.532 6818.9316 7098.265 7404.6235 7734.5225 8051.6675 8384.511 8600.304 8783.04 8846.781 8889.047 8908.35 8886.795 8877.337 8869.199 8844.105 8823.986 8800.736 8796.681 8797.683 8815.177 8827.009 8812.868 8855.23 8831.113 8929.532 8987.874 8969.738 8892.95 8802.79 8597.65 8361.995 8250.179 8041.98 8008.451 7846.6953 7713.182 ] true [7214.25 7077.09 6917.48 6689.79 6477.29 6276.32 6197.87 6187.15 6118.35 6181.4 6171.95 6235.12 6437.94 6705.62 7056.8 7474.55 7829.64 8123.71 8328.35 8516.35 8567.91 8617.11 8614.07 8563.02 8574.88 8522.64 8432.17 8448.92 8383.17 8395.01 8431.68 8447.05 8493.35 8393.05 8468.13 8468.94 8448.69 8355.89 8419.57 8511.51 8562.32 8415.76 8136.01 8077.22 7828.02 7755.79 7510.79 7312.72]
12-06
166 Model: "model_11" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= input_12 (InputLayer) [(None, 48, 3)] 0 _________________________________________________________________ conv1d_21 (Conv1D) (None, 48, 128) 512 _________________________________________________________________ conv1d_22 (Conv1D) (None, 48, 128) 16512 _________________________________________________________________ max_pooling1d_10 (MaxPooling (None, 1, 128) 0 _________________________________________________________________ lstm_31 (LSTM) (None, 1, 128) 131584 _________________________________________________________________ lstm_32 (LSTM) (None, 1, 64) 49408 _________________________________________________________________ lstm_33 (LSTM) (None, 48) 21696 _________________________________________________________________ dense_11 (Dense) (None, 48) 2352 ================================================================= Total params: 222,064 Trainable params: 222,064 Non-trainable params: 0 _________________________________________________________________ Train on 166 samples, validate on 72 samples Epoch 1/30 166/166 [==============================] - 3s 19ms/sample - loss: 0.1174 - accuracy: 0.0120 - val_loss: 0.0989 - val_accuracy: 0.0000e+00 Epoch 2/30 166/166 [==============================] - 0s 229us/sample - loss: 0.0984 - accuracy: 0.0120 - val_loss: 0.0699 - val_accuracy: 0.0000e+00 Epoch 3/30 166/166 [==============================] - 0s 229us/sample - loss: 0.0617 - accuracy: 0.0602 - val_loss: 0.0356 - val_accuracy: 0.0972 Epoch 4/30 166/166 [==============================] - 0s 223us/sample - loss: 0.0335 - accuracy: 0.1265 - val_loss: 0.0249 - val_accuracy: 0.0000e+00 Epoch 5/30 166/166 [==============================] - 0s 223us/sample - loss: 0.0189 - accuracy: 0.0120 - val_loss: 0.0116 - val_accuracy: 0.0000e+00 Epoch 6/30 166/166 [==============================] - 0s 229us/sample - loss: 0.0107 - accuracy: 0.0120 - val_loss: 0.0090 - val_accuracy: 0.0139 Epoch 7/30 166/166 [==============================] - 0s 229us/sample - loss: 0.0084 - accuracy: 0.0060 - val_loss: 0.0083 - val_accuracy: 0.0139 Epoch 8/30 166/166 [==============================] - 0s 229us/sample - loss: 0.0077 - accuracy: 0.0843 - val_loss: 0.0081 - val_accuracy: 0.2361 Epoch 9/30 166/166 [==============================] - 0s 223us/sample - loss: 0.0074 - accuracy: 0.1687 - val_loss: 0.0075 - val_accuracy: 0.2361 Epoch 10/30 166/166 [==============================] - 0s 229us/sample - loss: 0.0073 - accuracy: 0.1506 - val_loss: 0.0073 - val_accuracy: 0.0278 Epoch 11/30 166/166 [==============================] - 0s 229us/sample - loss: 0.0070 - accuracy: 0.0120 - val_loss: 0.0074 - val_accuracy: 0.0278 Epoch 12/30 166/166 [==============================] - 0s 229us/sample - loss: 0.0070 - accuracy: 0.0120 - val_loss: 0.0073 - val_accuracy: 0.0417 Epoch 13/30 166/166 [==============================] - 0s 235us/sample - loss: 0.0069 - accuracy: 0.0361 - val_loss: 0.0072 - val_accuracy: 0.0972 Epoch 14/30 166/166 [==============================] - 0s 223us/sample - loss: 0.0069 - accuracy: 0.1265 - val_loss: 0.0073 - val_accuracy: 0.0139 Epoch 15/30 166/166 [==============================] - 0s 247us/sample - loss: 0.0069 - accuracy: 0.0060 - val_loss: 0.0073 - val_accuracy: 0.0000e+00 Epoch 16/30 166/166 [==============================] - 0s 235us/sample - loss: 0.0069 - accuracy: 0.0120 - val_loss: 0.0072 - val_accuracy: 0.0000e+00 Epoch 17/30 166/166 [==============================] - 0s 223us/sample - loss: 0.0069 - accuracy: 0.0120 - val_loss: 0.0073 - val_accuracy: 0.0000e+00 Epoch 18/30 166/166 [==============================] - 0s 235us/sample - loss: 0.0069 - accuracy: 0.0181 - val_loss: 0.0073 - val_accuracy: 0.0417 Epoch 19/30 166/166 [==============================] - 0s 241us/sample - loss: 0.0069 - accuracy: 0.0361 - val_loss: 0.0072 - val_accuracy: 0.0417 Epoch 20/30 166/166 [==============================] - 0s 247us/sample - loss: 0.0069 - accuracy: 0.0361 - val_loss: 0.0073 - val_accuracy: 0.0417 Epoch 21/30 166/166 [==============================] - 0s 247us/sample - loss: 0.0069 - accuracy: 0.0361 - val_loss: 0.0073 - val_accuracy: 0.0417 Epoch 22/30 166/166 [==============================] - 0s 235us/sample - loss: 0.0069 - accuracy: 0.0361 - val_loss: 0.0073 - val_accuracy: 0.0417 Epoch 23/30 166/166 [==============================] - 0s 241us/sample - loss: 0.0069 - accuracy: 0.0361 - val_loss: 0.0073 - val_accuracy: 0.0417 Epoch 24/30 166/166 [==============================] - 0s 229us/sample - loss: 0.0069 - accuracy: 0.0361 - val_loss: 0.0073 - val_accuracy: 0.0417 Epoch 25/30 166/166 [==============================] - 0s 229us/sample - loss: 0.0069 - accuracy: 0.0361 - val_loss: 0.0073 - val_accuracy: 0.0417 Epoch 26/30 166/166 [==============================] - 0s 229us/sample - loss: 0.0069 - accuracy: 0.0361 - val_loss: 0.0073 - val_accuracy: 0.0417 Epoch 27/30 166/166 [==============================] - 0s 241us/sample - loss: 0.0069 - accuracy: 0.0361 - val_loss: 0.0073 - val_accuracy: 0.0417 Epoch 28/30 166/166 [==============================] - 0s 229us/sample - loss: 0.0069 - accuracy: 0.0361 - val_loss: 0.0073 - val_accuracy: 0.0417 Epoch 29/30 166/166 [==============================] - 0s 229us/sample - loss: 0.0069 - accuracy: 0.0361 - val_loss: 0.0073 - val_accuracy: 0.0417 Epoch 30/30 166/166 [==============================] - 0s 229us/sample - loss: 0.0069 - accuracy: 0.0361 - val_loss: 0.0073 - val_accuracy: 0.0417
pred [7354.5415 7182.1494 7023.1646 6786.9297 6575.576 6383.8877 6265.299 6217.471 6225.808 6324.8413 6465.7065 6781.5767 7123.324 7488.344 7885.414 8183.9663 8475.529 8767.779 8939.7705 9089.954 9161.981 9205.077 9236.669 9222.855 9224.209 9227.723 9209.451 9200.444 9182.419 9181.256 9187.504 9200.718 9202.423 9174.872 9196.694 9130.557 9190.921 9206.005 9180.409 9075.652 8950.305 8722.285 8462.343 8361.873 8143.2144 8144.484 7988.845 7862.783 ] true [ 7163.05 7026.16 6907.79 6608.58 6414.14 6387.98 6332.67 6344.56 6390.92 6646.32 6932.74 7525.56 8191.29 8655. 9120.43 9367.3 9579.16 9913.47 10099.72 10200.93 10286.04 10252.95 10332.86 10340.67 10322.2 10308.35 10352.2 10331.84 10309.15 10244.07 10274.3 10298.24 10257.93 10166.79 9939.89 9578.21 9406.9 9190.17 9094.8 9258.12 9218.7 9014.7 8716.05 8661.35 8453.62 8341.08 8195.69 7920.77]
12-07
249 Model: "model_12" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= input_13 (InputLayer) [(None, 48, 3)] 0 _________________________________________________________________ conv1d_23 (Conv1D) (None, 48, 128) 512 _________________________________________________________________ conv1d_24 (Conv1D) (None, 48, 128) 16512 _________________________________________________________________ max_pooling1d_11 (MaxPooling (None, 1, 128) 0 _________________________________________________________________ lstm_34 (LSTM) (None, 1, 128) 131584 _________________________________________________________________ lstm_35 (LSTM) (None, 1, 64) 49408 _________________________________________________________________ lstm_36 (LSTM) (None, 48) 21696 _________________________________________________________________ dense_12 (Dense) (None, 48) 2352 ================================================================= Total params: 222,064 Trainable params: 222,064 Non-trainable params: 0 _________________________________________________________________ Train on 249 samples, validate on 107 samples Epoch 1/30 249/249 [==============================] - 3s 12ms/sample - loss: 0.1597 - accuracy: 0.0442 - val_loss: 0.1422 - val_accuracy: 0.0000e+00 Epoch 2/30 249/249 [==============================] - 0s 221us/sample - loss: 0.1295 - accuracy: 0.0000e+00 - val_loss: 0.0909 - val_accuracy: 0.0000e+00 Epoch 3/30 249/249 [==============================] - 0s 221us/sample - loss: 0.0688 - accuracy: 0.0040 - val_loss: 0.0425 - val_accuracy: 0.0000e+00 Epoch 4/30 249/249 [==============================] - 0s 221us/sample - loss: 0.0294 - accuracy: 0.0080 - val_loss: 0.0129 - val_accuracy: 0.0000e+00 Epoch 5/30 249/249 [==============================] - 0s 221us/sample - loss: 0.0093 - accuracy: 0.0602 - val_loss: 0.0061 - val_accuracy: 0.0935 Epoch 6/30 249/249 [==============================] - 0s 217us/sample - loss: 0.0057 - accuracy: 0.0723 - val_loss: 0.0053 - val_accuracy: 0.0280 Epoch 7/30 249/249 [==============================] - 0s 233us/sample - loss: 0.0049 - accuracy: 0.0120 - val_loss: 0.0048 - val_accuracy: 0.0280 Epoch 8/30 249/249 [==============================] - 0s 225us/sample - loss: 0.0046 - accuracy: 0.0120 - val_loss: 0.0044 - val_accuracy: 0.0280 Epoch 9/30 249/249 [==============================] - 0s 225us/sample - loss: 0.0042 - accuracy: 0.0402 - val_loss: 0.0042 - val_accuracy: 0.0280 Epoch 10/30 249/249 [==============================] - 0s 217us/sample - loss: 0.0042 - accuracy: 0.1365 - val_loss: 0.0041 - val_accuracy: 0.1121 Epoch 11/30 249/249 [==============================] - 0s 217us/sample - loss: 0.0042 - accuracy: 0.1807 - val_loss: 0.0041 - val_accuracy: 0.1121 Epoch 12/30 249/249 [==============================] - 0s 217us/sample - loss: 0.0041 - accuracy: 0.1847 - val_loss: 0.0041 - val_accuracy: 0.2523 Epoch 13/30 249/249 [==============================] - 0s 217us/sample - loss: 0.0041 - accuracy: 0.1968 - val_loss: 0.0041 - val_accuracy: 0.2523 Epoch 14/30 249/249 [==============================] - 0s 217us/sample - loss: 0.0041 - accuracy: 0.1968 - val_loss: 0.0041 - val_accuracy: 0.1121 Epoch 15/30 249/249 [==============================] - 0s 217us/sample - loss: 0.0041 - accuracy: 0.1807 - val_loss: 0.0041 - val_accuracy: 0.1121 Epoch 16/30 249/249 [==============================] - 0s 217us/sample - loss: 0.0041 - accuracy: 0.1566 - val_loss: 0.0041 - val_accuracy: 0.2523 Epoch 17/30 249/249 [==============================] - 0s 217us/sample - loss: 0.0041 - accuracy: 0.1968 - val_loss: 0.0041 - val_accuracy: 0.1121 Epoch 18/30 249/249 [==============================] - 0s 221us/sample - loss: 0.0041 - accuracy: 0.1807 - val_loss: 0.0041 - val_accuracy: 0.1121 Epoch 19/30 249/249 [==============================] - 0s 217us/sample - loss: 0.0041 - accuracy: 0.1807 - val_loss: 0.0041 - val_accuracy: 0.1121 Epoch 20/30 249/249 [==============================] - 0s 217us/sample - loss: 0.0041 - accuracy: 0.1807 - val_loss: 0.0041 - val_accuracy: 0.1121 Epoch 21/30 249/249 [==============================] - 0s 221us/sample - loss: 0.0041 - accuracy: 0.1807 - val_loss: 0.0041 - val_accuracy: 0.1121 Epoch 22/30 249/249 [==============================] - 0s 229us/sample - loss: 0.0041 - accuracy: 0.1807 - val_loss: 0.0041 - val_accuracy: 0.1121 Epoch 23/30 249/249 [==============================] - 0s 213us/sample - loss: 0.0041 - accuracy: 0.1807 - val_loss: 0.0041 - val_accuracy: 0.1121 Epoch 24/30 249/249 [==============================] - 0s 225us/sample - loss: 0.0041 - accuracy: 0.1807 - val_loss: 0.0041 - val_accuracy: 0.1121 Epoch 25/30 249/249 [==============================] - 0s 241us/sample - loss: 0.0041 - accuracy: 0.1807 - val_loss: 0.0041 - val_accuracy: 0.1121 Epoch 26/30 249/249 [==============================] - 0s 229us/sample - loss: 0.0041 - accuracy: 0.1807 - val_loss: 0.0041 - val_accuracy: 0.1121 Epoch 27/30 249/249 [==============================] - 0s 241us/sample - loss: 0.0041 - accuracy: 0.1807 - val_loss: 0.0041 - val_accuracy: 0.1121 Epoch 28/30 249/249 [==============================] - 0s 225us/sample - loss: 0.0041 - accuracy: 0.1807 - val_loss: 0.0041 - val_accuracy: 0.1121 Epoch 29/30 249/249 [==============================] - 0s 217us/sample - loss: 0.0041 - accuracy: 0.1807 - val_loss: 0.0041 - val_accuracy: 0.1121 Epoch 30/30 249/249 [==============================] - 0s 217us/sample - loss: 0.0041 - accuracy: 0.1807 - val_loss: 0.0041 - val_accuracy: 0.1121
pred [ 7950.193 7752.6665 7509.118 7175.239 6858.6357 6656.558 6547.883 6516.3267 6567.1587 6757.177 7002.997 7536.818 8058.975 8442.772 8856.219 9099.122 9299.801 9566.696 9714.75 9822.131 9881.78 9907.257 9956.303 9942.771 9949.823 9984.191 10001.972 10012.741 10009.141 9993.645 10024.178 10048.361 10039.312 9939.174 9861.79 9632.454 9504.164 9341.0205 9316.258 9310.65 9220.555 8952.673 8659.018 8647.736 8458.635 8429.165 8263.59 8075.7935] true [ 7766.2 7617.96 7374.47 7021.17 6807.34 6672.05 6590.83 6575.89 6639.22 6807.09 7069.38 7606.75 8210.51 8694.29 9132.02 9381.14 9588.18 9836.12 9983.45 10149.58 10162.29 10159.52 10194.68 10203.19 10242.13 10289.77 10324.48 10372.78 10388.92 10338.68 10375.31 10367.85 10403.75 10288.48 10120.55 9782.41 9533.65 9305.94 9190.81 9279.69 9226.3 9068.28 8717.32 8689.82 8425.41 8369.89 8204.88 7994.47]
12-09
220 Model: "model_13" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= input_14 (InputLayer) [(None, 48, 3)] 0 _________________________________________________________________ conv1d_25 (Conv1D) (None, 48, 128) 512 _________________________________________________________________ conv1d_26 (Conv1D) (None, 48, 128) 16512 _________________________________________________________________ max_pooling1d_12 (MaxPooling (None, 1, 128) 0 _________________________________________________________________ lstm_37 (LSTM) (None, 1, 128) 131584 _________________________________________________________________ lstm_38 (LSTM) (None, 1, 64) 49408 _________________________________________________________________ lstm_39 (LSTM) (None, 48) 21696 _________________________________________________________________ dense_13 (Dense) (None, 48) 2352 ================================================================= Total params: 222,064 Trainable params: 222,064 Non-trainable params: 0 _________________________________________________________________ Train on 220 samples, validate on 95 samples Epoch 1/30 220/220 [==============================] - 4s 18ms/sample - loss: 0.1735 - accuracy: 0.0455 - val_loss: 0.1582 - val_accuracy: 0.0000e+00 Epoch 2/30 220/220 [==============================] - 0s 214us/sample - loss: 0.1397 - accuracy: 0.0000e+00 - val_loss: 0.1031 - val_accuracy: 0.0000e+00 Epoch 3/30 220/220 [==============================] - 0s 227us/sample - loss: 0.0733 - accuracy: 0.0000e+00 - val_loss: 0.0429 - val_accuracy: 0.0000e+00 Epoch 4/30 220/220 [==============================] - 0s 236us/sample - loss: 0.0320 - accuracy: 0.0227 - val_loss: 0.0217 - val_accuracy: 0.0211 Epoch 5/30 220/220 [==============================] - 0s 218us/sample - loss: 0.0119 - accuracy: 0.0136 - val_loss: 0.0078 - val_accuracy: 0.0211 Epoch 6/30 220/220 [==============================] - 0s 223us/sample - loss: 0.0062 - accuracy: 0.0591 - val_loss: 0.0073 - val_accuracy: 0.0316 Epoch 7/30 220/220 [==============================] - 0s 227us/sample - loss: 0.0057 - accuracy: 0.0364 - val_loss: 0.0069 - val_accuracy: 0.0316 Epoch 8/30 220/220 [==============================] - 0s 218us/sample - loss: 0.0052 - accuracy: 0.0318 - val_loss: 0.0066 - val_accuracy: 0.0526 Epoch 9/30 220/220 [==============================] - 0s 236us/sample - loss: 0.0047 - accuracy: 0.1091 - val_loss: 0.0062 - val_accuracy: 0.1789 Epoch 10/30 220/220 [==============================] - 0s 232us/sample - loss: 0.0045 - accuracy: 0.1909 - val_loss: 0.0060 - val_accuracy: 0.1789 Epoch 11/30 220/220 [==============================] - 0s 227us/sample - loss: 0.0044 - accuracy: 0.1909 - val_loss: 0.0060 - val_accuracy: 0.0842 Epoch 12/30 220/220 [==============================] - 0s 227us/sample - loss: 0.0043 - accuracy: 0.0955 - val_loss: 0.0060 - val_accuracy: 0.0632 Epoch 13/30 220/220 [==============================] - 0s 227us/sample - loss: 0.0043 - accuracy: 0.0818 - val_loss: 0.0060 - val_accuracy: 0.0632 Epoch 14/30 220/220 [==============================] - 0s 223us/sample - loss: 0.0043 - accuracy: 0.1864 - val_loss: 0.0060 - val_accuracy: 0.1789 Epoch 15/30 220/220 [==============================] - 0s 218us/sample - loss: 0.0043 - accuracy: 0.1909 - val_loss: 0.0060 - val_accuracy: 0.1789 Epoch 16/30 220/220 [==============================] - 0s 223us/sample - loss: 0.0043 - accuracy: 0.1909 - val_loss: 0.0060 - val_accuracy: 0.1789 Epoch 17/30 220/220 [==============================] - 0s 223us/sample - loss: 0.0043 - accuracy: 0.1909 - val_loss: 0.0060 - val_accuracy: 0.1789 Epoch 18/30 220/220 [==============================] - 0s 223us/sample - loss: 0.0043 - accuracy: 0.1909 - val_loss: 0.0060 - val_accuracy: 0.1789 Epoch 19/30 220/220 [==============================] - 0s 223us/sample - loss: 0.0043 - accuracy: 0.1909 - val_loss: 0.0060 - val_accuracy: 0.1789 Epoch 20/30 220/220 [==============================] - 0s 227us/sample - loss: 0.0043 - accuracy: 0.1909 - val_loss: 0.0060 - val_accuracy: 0.1789 Epoch 21/30 220/220 [==============================] - 0s 227us/sample - loss: 0.0043 - accuracy: 0.1909 - val_loss: 0.0060 - val_accuracy: 0.1789 Epoch 22/30 220/220 [==============================] - 0s 223us/sample - loss: 0.0043 - accuracy: 0.1909 - val_loss: 0.0060 - val_accuracy: 0.1789 Epoch 23/30 220/220 [==============================] - 0s 227us/sample - loss: 0.0043 - accuracy: 0.1909 - val_loss: 0.0060 - val_accuracy: 0.1789 Epoch 24/30 220/220 [==============================] - 0s 223us/sample - loss: 0.0043 - accuracy: 0.1909 - val_loss: 0.0060 - val_accuracy: 0.1789 Epoch 25/30 220/220 [==============================] - 0s 232us/sample - loss: 0.0043 - accuracy: 0.1909 - val_loss: 0.0060 - val_accuracy: 0.1789 Epoch 26/30 220/220 [==============================] - 0s 223us/sample - loss: 0.0043 - accuracy: 0.1909 - val_loss: 0.0060 - val_accuracy: 0.1789 Epoch 27/30 220/220 [==============================] - 0s 236us/sample - loss: 0.0043 - accuracy: 0.1909 - val_loss: 0.0060 - val_accuracy: 0.1789 Epoch 28/30 220/220 [==============================] - 0s 223us/sample - loss: 0.0043 - accuracy: 0.1909 - val_loss: 0.0060 - val_accuracy: 0.1789 Epoch 29/30 220/220 [==============================] - 0s 218us/sample - loss: 0.0043 - accuracy: 0.1909 - val_loss: 0.0060 - val_accuracy: 0.1789 Epoch 30/30 220/220 [==============================] - 0s 223us/sample - loss: 0.0043 - accuracy: 0.1909 - val_loss: 0.0060 - val_accuracy: 0.1789
pred [ 8016.648 7805.978 7565.331 7230.298 6919.1523 6718.987 6612.395 6582.264 6642.0503 6825.507 7071.503 7592.791 8105.245 8503.763 8918.699 9190.54 9414.387 9704.001 9857.393 9987.211 10060.728 10099.812 10158.307 10168.561 10185.43 10232.804 10262.596 10288.372 10289.455 10274.563 10311.14 10325.333 10309.139 10204.239 10100.845 9852.403 9704.055 9517.541 9475.692 9465.053 9359.09 9088.985 8783.88 8770.908 8567.945 8523.855 8342.21 8139.722 ] true [ 7851.48 7690.51 7506.92 7194.91 6880.82 6697.45 6590.59 6611.04 6667.32 6994.37 7313.09 7882.93 8454.02 8989.95 9480.06 9754.4 9945.64 10205.44 10561.68 10735.26 10858.15 10972.1 11077.14 11138.77 11105.14 11229.09 11354.06 11342.74 11301.26 11240.85 11358.58 11354.11 11337.73 11077. 10948.37 10644.45 10397.09 10175.02 10005.1 10062.6 9950.54 9583.81 9215.56 9072.67 8819.12 8642.37 8404.12 8150.69]
12-10
141 Model: "model_14" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= input_15 (InputLayer) [(None, 48, 3)] 0 _________________________________________________________________ conv1d_27 (Conv1D) (None, 48, 128) 512 _________________________________________________________________ conv1d_28 (Conv1D) (None, 48, 128) 16512 _________________________________________________________________ max_pooling1d_13 (MaxPooling (None, 1, 128) 0 _________________________________________________________________ lstm_40 (LSTM) (None, 1, 128) 131584 _________________________________________________________________ lstm_41 (LSTM) (None, 1, 64) 49408 _________________________________________________________________ lstm_42 (LSTM) (None, 48) 21696 _________________________________________________________________ dense_14 (Dense) (None, 48) 2352 ================================================================= Total params: 222,064 Trainable params: 222,064 Non-trainable params: 0 _________________________________________________________________ Train on 141 samples, validate on 61 samples Epoch 1/30 141/141 [==============================] - 3s 20ms/sample - loss: 0.1910 - accuracy: 0.0213 - val_loss: 0.1854 - val_accuracy: 0.0164 Epoch 2/30 141/141 [==============================] - 0s 277us/sample - loss: 0.1731 - accuracy: 0.0426 - val_loss: 0.1569 - val_accuracy: 0.0164 Epoch 3/30 141/141 [==============================] - 0s 248us/sample - loss: 0.1370 - accuracy: 0.0426 - val_loss: 0.1055 - val_accuracy: 0.0164 Epoch 4/30 141/141 [==============================] - 0s 255us/sample - loss: 0.0827 - accuracy: 0.0426 - val_loss: 0.0546 - val_accuracy: 0.0164 Epoch 5/30 141/141 [==============================] - 0s 248us/sample - loss: 0.0446 - accuracy: 0.0426 - val_loss: 0.0349 - val_accuracy: 0.0164 Epoch 6/30 141/141 [==============================] - 0s 255us/sample - loss: 0.0265 - accuracy: 0.0426 - val_loss: 0.0170 - val_accuracy: 0.0656 Epoch 7/30 141/141 [==============================] - 0s 255us/sample - loss: 0.0123 - accuracy: 0.0638 - val_loss: 0.0112 - val_accuracy: 0.0164 Epoch 8/30 141/141 [==============================] - 0s 305us/sample - loss: 0.0089 - accuracy: 0.0000e+00 - val_loss: 0.0095 - val_accuracy: 0.0164 Epoch 9/30 141/141 [==============================] - 0s 284us/sample - loss: 0.0071 - accuracy: 0.0142 - val_loss: 0.0080 - val_accuracy: 0.0000e+00 Epoch 10/30 141/141 [==============================] - 0s 284us/sample - loss: 0.0062 - accuracy: 0.0496 - val_loss: 0.0076 - val_accuracy: 0.0820 Epoch 11/30 141/141 [==============================] - 0s 262us/sample - loss: 0.0058 - accuracy: 0.1560 - val_loss: 0.0073 - val_accuracy: 0.2295 Epoch 12/30 141/141 [==============================] - 0s 262us/sample - loss: 0.0055 - accuracy: 0.2057 - val_loss: 0.0072 - val_accuracy: 0.2295 Epoch 13/30 141/141 [==============================] - 0s 255us/sample - loss: 0.0054 - accuracy: 0.2057 - val_loss: 0.0070 - val_accuracy: 0.2295 Epoch 14/30 141/141 [==============================] - 0s 255us/sample - loss: 0.0053 - accuracy: 0.2057 - val_loss: 0.0068 - val_accuracy: 0.2295 Epoch 15/30 141/141 [==============================] - 0s 270us/sample - loss: 0.0051 - accuracy: 0.1135 - val_loss: 0.0068 - val_accuracy: 0.0164 Epoch 16/30 141/141 [==============================] - 0s 248us/sample - loss: 0.0051 - accuracy: 0.0426 - val_loss: 0.0067 - val_accuracy: 0.0164 Epoch 17/30 141/141 [==============================] - 0s 262us/sample - loss: 0.0051 - accuracy: 0.0355 - val_loss: 0.0067 - val_accuracy: 0.0820 Epoch 18/30 141/141 [==============================] - 0s 255us/sample - loss: 0.0051 - accuracy: 0.0851 - val_loss: 0.0067 - val_accuracy: 0.0820 Epoch 19/30 141/141 [==============================] - 0s 255us/sample - loss: 0.0051 - accuracy: 0.0851 - val_loss: 0.0067 - val_accuracy: 0.0820 Epoch 20/30 141/141 [==============================] - 0s 262us/sample - loss: 0.0050 - accuracy: 0.1773 - val_loss: 0.0067 - val_accuracy: 0.2131 Epoch 21/30 141/141 [==============================] - 0s 262us/sample - loss: 0.0050 - accuracy: 0.2340 - val_loss: 0.0067 - val_accuracy: 0.2131 Epoch 22/30 141/141 [==============================] - 0s 255us/sample - loss: 0.0050 - accuracy: 0.2340 - val_loss: 0.0067 - val_accuracy: 0.2131 Epoch 23/30 141/141 [==============================] - 0s 270us/sample - loss: 0.0050 - accuracy: 0.2340 - val_loss: 0.0067 - val_accuracy: 0.2131 Epoch 24/30 141/141 [==============================] - 0s 255us/sample - loss: 0.0050 - accuracy: 0.2340 - val_loss: 0.0067 - val_accuracy: 0.2295 Epoch 25/30 141/141 [==============================] - 0s 255us/sample - loss: 0.0050 - accuracy: 0.1631 - val_loss: 0.0067 - val_accuracy: 0.2131 Epoch 26/30 141/141 [==============================] - 0s 262us/sample - loss: 0.0050 - accuracy: 0.2340 - val_loss: 0.0067 - val_accuracy: 0.2131 Epoch 27/30 141/141 [==============================] - 0s 270us/sample - loss: 0.0050 - accuracy: 0.2340 - val_loss: 0.0067 - val_accuracy: 0.2131 Epoch 28/30 141/141 [==============================] - 0s 262us/sample - loss: 0.0050 - accuracy: 0.2340 - val_loss: 0.0067 - val_accuracy: 0.2131 Epoch 29/30 141/141 [==============================] - 0s 262us/sample - loss: 0.0050 - accuracy: 0.2340 - val_loss: 0.0067 - val_accuracy: 0.2131 Epoch 30/30 141/141 [==============================] - 0s 248us/sample - loss: 0.0050 - accuracy: 0.2340 - val_loss: 0.0067 - val_accuracy: 0.2131
pred [ 8100.379 7865.2554 7619.7754 7276.385 6973.5347 6773.982 6673.3857 6649.74 6698.5664 6889.2485 7125.77 7637.74 8140.2935 8540.239 8965.953 9274.014 9529.929 9845.902 10007.664 10151.896 10240.472 10280.724 10356.417 10382.71 10418.244 10478.212 10513.614 10547.599 10557.547 10550.114 10585.956 10600.247 10574.437 10452.327 10347.073 10070.917 9900.528 9676.019 9598.458 9578.768 9478.127 9207.914 8900.773 8874.115 8658.45 8589.455 8392.926 8180.1553] true [ 7885.41 7664.28 7440.68 7107.98 6876.49 6682.53 6607.52 6631.62 6694.8 6931.47 7230.92 7782.65 8384.73 8785.7 9295.35 9540.97 9715.03 9909.28 10038.93 10118.85 9867.44 9968.02 10288.14 10332.53 10314.62 10365.4 10391.8 10378.99 10404.25 10357.03 10454.88 10463.79 10407.06 10267.94 10102.7 9729.57 9550.78 9345.59 9231.96 9241.71 9302.13 9111.91 8834.14 8776.39 8604.26 8438. 8267.3 7948.22]
12-11
286 Model: "model_15" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= input_16 (InputLayer) [(None, 48, 3)] 0 _________________________________________________________________ conv1d_29 (Conv1D) (None, 48, 128) 512 _________________________________________________________________ conv1d_30 (Conv1D) (None, 48, 128) 16512 _________________________________________________________________ max_pooling1d_14 (MaxPooling (None, 1, 128) 0 _________________________________________________________________ lstm_43 (LSTM) (None, 1, 128) 131584 _________________________________________________________________ lstm_44 (LSTM) (None, 1, 64) 49408 _________________________________________________________________ lstm_45 (LSTM) (None, 48) 21696 _________________________________________________________________ dense_15 (Dense) (None, 48) 2352 ================================================================= Total params: 222,064 Trainable params: 222,064 Non-trainable params: 0 _________________________________________________________________ Train on 286 samples, validate on 123 samples Epoch 1/30 286/286 [==============================] - 4s 15ms/sample - loss: 0.1541 - accuracy: 0.0874 - val_loss: 0.1345 - val_accuracy: 0.0976 Epoch 2/30 286/286 [==============================] - 0s 238us/sample - loss: 0.1053 - accuracy: 0.0979 - val_loss: 0.0608 - val_accuracy: 0.0976 Epoch 3/30 286/286 [==============================] - 0s 245us/sample - loss: 0.0416 - accuracy: 0.0979 - val_loss: 0.0240 - val_accuracy: 0.2195 Epoch 4/30 286/286 [==============================] - 0s 231us/sample - loss: 0.0151 - accuracy: 0.1958 - val_loss: 0.0077 - val_accuracy: 0.2195 Epoch 5/30 286/286 [==============================] - 0s 241us/sample - loss: 0.0069 - accuracy: 0.1329 - val_loss: 0.0049 - val_accuracy: 0.0813 Epoch 6/30 286/286 [==============================] - 0s 255us/sample - loss: 0.0055 - accuracy: 0.0455 - val_loss: 0.0045 - val_accuracy: 0.0000e+00 Epoch 7/30 286/286 [==============================] - 0s 252us/sample - loss: 0.0051 - accuracy: 0.0070 - val_loss: 0.0043 - val_accuracy: 0.0000e+00 Epoch 8/30 286/286 [==============================] - 0s 252us/sample - loss: 0.0049 - accuracy: 0.0070 - val_loss: 0.0042 - val_accuracy: 0.0000e+00 Epoch 9/30 286/286 [==============================] - 0s 248us/sample - loss: 0.0048 - accuracy: 0.0699 - val_loss: 0.0040 - val_accuracy: 0.1463 Epoch 10/30 286/286 [==============================] - 0s 248us/sample - loss: 0.0047 - accuracy: 0.0734 - val_loss: 0.0040 - val_accuracy: 0.2195 Epoch 11/30 286/286 [==============================] - 0s 245us/sample - loss: 0.0047 - accuracy: 0.1958 - val_loss: 0.0040 - val_accuracy: 0.2195 Epoch 12/30 286/286 [==============================] - 0s 238us/sample - loss: 0.0047 - accuracy: 0.1958 - val_loss: 0.0040 - val_accuracy: 0.2195 Epoch 13/30 286/286 [==============================] - 0s 231us/sample - loss: 0.0047 - accuracy: 0.1958 - val_loss: 0.0040 - val_accuracy: 0.2195 Epoch 14/30 286/286 [==============================] - 0s 224us/sample - loss: 0.0047 - accuracy: 0.1853 - val_loss: 0.0040 - val_accuracy: 0.1463 Epoch 15/30 286/286 [==============================] - 0s 227us/sample - loss: 0.0047 - accuracy: 0.1538 - val_loss: 0.0040 - val_accuracy: 0.1463 Epoch 16/30 286/286 [==============================] - 0s 224us/sample - loss: 0.0047 - accuracy: 0.1538 - val_loss: 0.0040 - val_accuracy: 0.1463 Epoch 17/30 286/286 [==============================] - 0s 231us/sample - loss: 0.0047 - accuracy: 0.1538 - val_loss: 0.0040 - val_accuracy: 0.1463 Epoch 18/30 286/286 [==============================] - 0s 234us/sample - loss: 0.0047 - accuracy: 0.1818 - val_loss: 0.0040 - val_accuracy: 0.1463 Epoch 19/30 286/286 [==============================] - 0s 241us/sample - loss: 0.0047 - accuracy: 0.1818 - val_loss: 0.0040 - val_accuracy: 0.1463 Epoch 20/30 286/286 [==============================] - 0s 245us/sample - loss: 0.0047 - accuracy: 0.1538 - val_loss: 0.0040 - val_accuracy: 0.1463 Epoch 21/30 286/286 [==============================] - 0s 245us/sample - loss: 0.0047 - accuracy: 0.1538 - val_loss: 0.0040 - val_accuracy: 0.1463 Epoch 22/30 286/286 [==============================] - 0s 245us/sample - loss: 0.0047 - accuracy: 0.1538 - val_loss: 0.0040 - val_accuracy: 0.1463 Epoch 23/30 286/286 [==============================] - 0s 245us/sample - loss: 0.0047 - accuracy: 0.1538 - val_loss: 0.0040 - val_accuracy: 0.1463 Epoch 24/30 286/286 [==============================] - 0s 259us/sample - loss: 0.0047 - accuracy: 0.1538 - val_loss: 0.0040 - val_accuracy: 0.1463 Epoch 25/30 286/286 [==============================] - 0s 241us/sample - loss: 0.0047 - accuracy: 0.1538 - val_loss: 0.0040 - val_accuracy: 0.1463 Epoch 26/30 286/286 [==============================] - 0s 238us/sample - loss: 0.0047 - accuracy: 0.1538 - val_loss: 0.0040 - val_accuracy: 0.1463 Epoch 27/30 286/286 [==============================] - 0s 245us/sample - loss: 0.0047 - accuracy: 0.1538 - val_loss: 0.0040 - val_accuracy: 0.1463 Epoch 28/30 286/286 [==============================] - 0s 241us/sample - loss: 0.0047 - accuracy: 0.1538 - val_loss: 0.0040 - val_accuracy: 0.1463 Epoch 29/30 286/286 [==============================] - 0s 231us/sample - loss: 0.0047 - accuracy: 0.1538 - val_loss: 0.0040 - val_accuracy: 0.1463 Epoch 30/30 286/286 [==============================] - 0s 238us/sample - loss: 0.0047 - accuracy: 0.1538 - val_loss: 0.0040 - val_accuracy: 0.1463
pred [8003.1416 7789.0356 7551.5386 7223.4604 6904.8945 6691.962 6576.3594 6549.2715 6590.569 6776.643 6999.612 7522.885 8020.3594 8403.926 8824.659 9078.801 9282.94 9558.638 9687.96 9799.985 9857.573 9878.223 9918.466 9909.272 9913.162 9938.115 9952.026 9969.912 9952.414 9943.165 9985.121 9994.311 9988.686 9893.041 9827.561 9627.074 9507.73 9362. 9333.248 9337.677 9232.361 8970.84 8682.373 8675.927 8469.995 8448.734 8279.1045 8104.3647] true [7708.56 7473.72 7213.36 6893.25 6629.62 6442.55 6346.8 6324.64 6325.32 6422.41 6432.17 6642. 6960.28 7320.87 7700.4 8026.47 8340.17 8599.07 8607.47 8766.32 8726.94 8725.8 8739.91 8730.05 8676.62 8706.94 8673.62 8689.01 8722.71 8758.53 8800.54 8825.8 8841.9 8718.52 8692.14 8643.35 8510.24 8348.76 8257.95 8244.88 8309.18 8222.97 8068.83 7946.74 7815.45 7753.92 7586.01 7321.49]
12-12
140 Model: "model_16" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= input_17 (InputLayer) [(None, 48, 3)] 0 _________________________________________________________________ conv1d_31 (Conv1D) (None, 48, 128) 512 _________________________________________________________________ conv1d_32 (Conv1D) (None, 48, 128) 16512 _________________________________________________________________ max_pooling1d_15 (MaxPooling (None, 1, 128) 0 _________________________________________________________________ lstm_46 (LSTM) (None, 1, 128) 131584 _________________________________________________________________ lstm_47 (LSTM) (None, 1, 64) 49408 _________________________________________________________________ lstm_48 (LSTM) (None, 48) 21696 _________________________________________________________________ dense_16 (Dense) (None, 48) 2352 ================================================================= Total params: 222,064 Trainable params: 222,064 Non-trainable params: 0 _________________________________________________________________ Train on 140 samples, validate on 60 samples Epoch 1/30 140/140 [==============================] - 3s 22ms/sample - loss: 0.1152 - accuracy: 0.0000e+00 - val_loss: 0.1050 - val_accuracy: 0.0833 Epoch 2/30 140/140 [==============================] - 0s 243us/sample - loss: 0.1046 - accuracy: 0.1000 - val_loss: 0.0893 - val_accuracy: 0.0333 Epoch 3/30 140/140 [==============================] - 0s 236us/sample - loss: 0.0841 - accuracy: 0.0429 - val_loss: 0.0613 - val_accuracy: 0.0333 Epoch 4/30 140/140 [==============================] - 0s 243us/sample - loss: 0.0535 - accuracy: 0.0429 - val_loss: 0.0372 - val_accuracy: 0.0333 Epoch 5/30 140/140 [==============================] - 0s 250us/sample - loss: 0.0343 - accuracy: 0.0429 - val_loss: 0.0272 - val_accuracy: 0.0333 Epoch 6/30 140/140 [==============================] - 0s 250us/sample - loss: 0.0221 - accuracy: 0.0429 - val_loss: 0.0154 - val_accuracy: 0.0000e+00 Epoch 7/30 140/140 [==============================] - 0s 250us/sample - loss: 0.0134 - accuracy: 0.0143 - val_loss: 0.0116 - val_accuracy: 0.0000e+00 Epoch 8/30 140/140 [==============================] - 0s 250us/sample - loss: 0.0107 - accuracy: 0.0429 - val_loss: 0.0097 - val_accuracy: 0.0000e+00 Epoch 9/30 140/140 [==============================] - 0s 257us/sample - loss: 0.0088 - accuracy: 0.0357 - val_loss: 0.0089 - val_accuracy: 0.0000e+00 Epoch 10/30 140/140 [==============================] - 0s 250us/sample - loss: 0.0081 - accuracy: 0.0643 - val_loss: 0.0085 - val_accuracy: 0.0000e+00 Epoch 11/30 140/140 [==============================] - 0s 271us/sample - loss: 0.0077 - accuracy: 0.0714 - val_loss: 0.0082 - val_accuracy: 0.1833 Epoch 12/30 140/140 [==============================] - 0s 243us/sample - loss: 0.0077 - accuracy: 0.1357 - val_loss: 0.0082 - val_accuracy: 0.0333 Epoch 13/30 140/140 [==============================] - 0s 264us/sample - loss: 0.0076 - accuracy: 0.0214 - val_loss: 0.0082 - val_accuracy: 0.0333 Epoch 14/30 140/140 [==============================] - 0s 264us/sample - loss: 0.0075 - accuracy: 0.0214 - val_loss: 0.0082 - val_accuracy: 0.0333 Epoch 15/30 140/140 [==============================] - 0s 257us/sample - loss: 0.0075 - accuracy: 0.0214 - val_loss: 0.0080 - val_accuracy: 0.0333 Epoch 16/30 140/140 [==============================] - 0s 257us/sample - loss: 0.0074 - accuracy: 0.0214 - val_loss: 0.0080 - val_accuracy: 0.1333 Epoch 17/30 140/140 [==============================] - 0s 264us/sample - loss: 0.0074 - accuracy: 0.1500 - val_loss: 0.0080 - val_accuracy: 0.1333 Epoch 18/30 140/140 [==============================] - 0s 257us/sample - loss: 0.0074 - accuracy: 0.1500 - val_loss: 0.0080 - val_accuracy: 0.1333 Epoch 19/30 140/140 [==============================] - 0s 250us/sample - loss: 0.0074 - accuracy: 0.1214 - val_loss: 0.0080 - val_accuracy: 0.0833 Epoch 20/30 140/140 [==============================] - 0s 264us/sample - loss: 0.0074 - accuracy: 0.0929 - val_loss: 0.0080 - val_accuracy: 0.0833 Epoch 21/30 140/140 [==============================] - 0s 271us/sample - loss: 0.0074 - accuracy: 0.0929 - val_loss: 0.0080 - val_accuracy: 0.0833 Epoch 22/30 140/140 [==============================] - 0s 279us/sample - loss: 0.0074 - accuracy: 0.0929 - val_loss: 0.0080 - val_accuracy: 0.0833 Epoch 23/30 140/140 [==============================] - 0s 257us/sample - loss: 0.0074 - accuracy: 0.1429 - val_loss: 0.0080 - val_accuracy: 0.1333 Epoch 24/30 140/140 [==============================] - 0s 271us/sample - loss: 0.0074 - accuracy: 0.1500 - val_loss: 0.0080 - val_accuracy: 0.1333 Epoch 25/30 140/140 [==============================] - 0s 250us/sample - loss: 0.0074 - accuracy: 0.1500 - val_loss: 0.0080 - val_accuracy: 0.1333 Epoch 26/30 140/140 [==============================] - 0s 250us/sample - loss: 0.0074 - accuracy: 0.1500 - val_loss: 0.0080 - val_accuracy: 0.1333 Epoch 27/30 140/140 [==============================] - 0s 257us/sample - loss: 0.0074 - accuracy: 0.1500 - val_loss: 0.0080 - val_accuracy: 0.1333 Epoch 28/30 140/140 [==============================] - 0s 243us/sample - loss: 0.0074 - accuracy: 0.1500 - val_loss: 0.0080 - val_accuracy: 0.1333 Epoch 29/30 140/140 [==============================] - 0s 236us/sample - loss: 0.0074 - accuracy: 0.1500 - val_loss: 0.0080 - val_accuracy: 0.1333 Epoch 30/30 140/140 [==============================] - 0s 229us/sample - loss: 0.0074 - accuracy: 0.1500 - val_loss: 0.0080 - val_accuracy: 0.1333
pred [7369.9585 7200.9834 7028.761 6761.938 6531.9375 6364.6978 6268.64 6232.3154 6252.644 6370.538 6502.516 6804.556 7121.7554 7435.1484 7800.4897 8127.2964 8407.719 8717.083 8905.924 9054.688 9125.381 9162.089 9204.826 9209.171 9215.575 9228.746 9223.391 9230.476 9224.799 9234.984 9263.72 9284.136 9293.766 9248.098 9252.37 9147.094 9106.476 9030.707 9013.392 8997.841 8918.276 8694.034 8413.436 8341.0625 8138.644 8098.1133 7925.825 7763.4053] true [7107.15 6905.44 6685.05 6420.49 6235.78 6065.24 5992.03 5974.2 5968.51 6016.01 6005.39 6104.04 6310.36 6617.23 6986.72 7313.32 7618.65 7902.21 8118.4 8240.1 8269.32 8339.51 8393.03 8496.69 8481.54 8521.15 8493.49 8519.37 8552.41 8632.61 8756.16 8774.66 8778.73 8775.16 8886.59 8863.23 8849.15 8777.13 8628.47 8714.02 8820.83 8704.29 8432.96 8294.08 8008.26 7824.74 7572.87 7315.62]
12-13
100 Model: "model_17" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= input_18 (InputLayer) [(None, 48, 3)] 0 _________________________________________________________________ conv1d_33 (Conv1D) (None, 48, 128) 512 _________________________________________________________________ conv1d_34 (Conv1D) (None, 48, 128) 16512 _________________________________________________________________ max_pooling1d_16 (MaxPooling (None, 1, 128) 0 _________________________________________________________________ lstm_49 (LSTM) (None, 1, 128) 131584 _________________________________________________________________ lstm_50 (LSTM) (None, 1, 64) 49408 _________________________________________________________________ lstm_51 (LSTM) (None, 48) 21696 _________________________________________________________________ dense_17 (Dense) (None, 48) 2352 ================================================================= Total params: 222,064 Trainable params: 222,064 Non-trainable params: 0 _________________________________________________________________ Train on 100 samples, validate on 43 samples Epoch 1/30 100/100 [==============================] - 3s 30ms/sample - loss: 0.1430 - accuracy: 0.0200 - val_loss: 0.1326 - val_accuracy: 0.0698 Epoch 2/30 100/100 [==============================] - 0s 270us/sample - loss: 0.1336 - accuracy: 0.0000e+00 - val_loss: 0.1189 - val_accuracy: 0.0698 Epoch 3/30 100/100 [==============================] - 0s 260us/sample - loss: 0.1171 - accuracy: 0.0000e+00 - val_loss: 0.0952 - val_accuracy: 0.0698 Epoch 4/30 100/100 [==============================] - 0s 260us/sample - loss: 0.0900 - accuracy: 0.0000e+00 - val_loss: 0.0634 - val_accuracy: 0.0698 Epoch 5/30 100/100 [==============================] - 0s 270us/sample - loss: 0.0577 - accuracy: 0.0400 - val_loss: 0.0409 - val_accuracy: 0.0000e+00 Epoch 6/30 100/100 [==============================] - 0s 270us/sample - loss: 0.0393 - accuracy: 0.0400 - val_loss: 0.0342 - val_accuracy: 0.0000e+00 Epoch 7/30 100/100 [==============================] - 0s 270us/sample - loss: 0.0311 - accuracy: 0.0400 - val_loss: 0.0226 - val_accuracy: 0.0000e+00 Epoch 8/30 100/100 [==============================] - 0s 270us/sample - loss: 0.0194 - accuracy: 0.0400 - val_loss: 0.0135 - val_accuracy: 0.0000e+00 Epoch 9/30 100/100 [==============================] - 0s 270us/sample - loss: 0.0122 - accuracy: 0.0400 - val_loss: 0.0096 - val_accuracy: 0.0000e+00 Epoch 10/30 100/100 [==============================] - 0s 280us/sample - loss: 0.0090 - accuracy: 0.0400 - val_loss: 0.0074 - val_accuracy: 0.0000e+00 Epoch 11/30 100/100 [==============================] - 0s 280us/sample - loss: 0.0069 - accuracy: 0.0400 - val_loss: 0.0067 - val_accuracy: 0.0000e+00 Epoch 12/30 100/100 [==============================] - 0s 290us/sample - loss: 0.0063 - accuracy: 0.0800 - val_loss: 0.0069 - val_accuracy: 0.0233 Epoch 13/30 100/100 [==============================] - 0s 270us/sample - loss: 0.0064 - accuracy: 0.0100 - val_loss: 0.0067 - val_accuracy: 0.0233 Epoch 14/30 100/100 [==============================] - 0s 270us/sample - loss: 0.0062 - accuracy: 0.0100 - val_loss: 0.0065 - val_accuracy: 0.0233 Epoch 15/30 100/100 [==============================] - 0s 270us/sample - loss: 0.0061 - accuracy: 0.0100 - val_loss: 0.0063 - val_accuracy: 0.0000e+00 Epoch 16/30 100/100 [==============================] - 0s 280us/sample - loss: 0.0060 - accuracy: 0.0200 - val_loss: 0.0062 - val_accuracy: 0.0000e+00 Epoch 17/30 100/100 [==============================] - 0s 270us/sample - loss: 0.0059 - accuracy: 0.0100 - val_loss: 0.0062 - val_accuracy: 0.0233 Epoch 18/30 100/100 [==============================] - 0s 300us/sample - loss: 0.0058 - accuracy: 0.0100 - val_loss: 0.0062 - val_accuracy: 0.0233 Epoch 19/30 100/100 [==============================] - 0s 270us/sample - loss: 0.0058 - accuracy: 0.0900 - val_loss: 0.0061 - val_accuracy: 0.1628 Epoch 20/30 100/100 [==============================] - 0s 270us/sample - loss: 0.0057 - accuracy: 0.1700 - val_loss: 0.0060 - val_accuracy: 0.1628 Epoch 21/30 100/100 [==============================] - 0s 270us/sample - loss: 0.0057 - accuracy: 0.1600 - val_loss: 0.0060 - val_accuracy: 0.0698 Epoch 22/30 100/100 [==============================] - 0s 280us/sample - loss: 0.0057 - accuracy: 0.0300 - val_loss: 0.0061 - val_accuracy: 0.0000e+00 Epoch 23/30 100/100 [==============================] - 0s 270us/sample - loss: 0.0057 - accuracy: 0.0400 - val_loss: 0.0061 - val_accuracy: 0.0000e+00 Epoch 24/30 100/100 [==============================] - 0s 270us/sample - loss: 0.0057 - accuracy: 0.0400 - val_loss: 0.0061 - val_accuracy: 0.0000e+00 Epoch 25/30 100/100 [==============================] - 0s 280us/sample - loss: 0.0057 - accuracy: 0.0400 - val_loss: 0.0061 - val_accuracy: 0.0000e+00 Epoch 26/30 100/100 [==============================] - 0s 280us/sample - loss: 0.0057 - accuracy: 0.0400 - val_loss: 0.0061 - val_accuracy: 0.0000e+00 Epoch 27/30 100/100 [==============================] - 0s 280us/sample - loss: 0.0057 - accuracy: 0.0400 - val_loss: 0.0061 - val_accuracy: 0.0000e+00 Epoch 28/30 100/100 [==============================] - 0s 270us/sample - loss: 0.0057 - accuracy: 0.0400 - val_loss: 0.0061 - val_accuracy: 0.0465 Epoch 29/30 100/100 [==============================] - 0s 270us/sample - loss: 0.0057 - accuracy: 0.1100 - val_loss: 0.0061 - val_accuracy: 0.0465 Epoch 30/30 100/100 [==============================] - 0s 260us/sample - loss: 0.0057 - accuracy: 0.1100 - val_loss: 0.0061 - val_accuracy: 0.1628
pred [7325.5684 7168.5063 7002.6133 6755.878 6544.169 6401.701 6299.653 6283.3438 6315.724 6475.9585 6681.091 7138.377 7583.3677 8009.2446 8450.552 8738.074 8983.585 9264.755 9423.79 9569.858 9637.64 9665.849 9721.047 9722.32 9724.004 9768.039 9787.563 9786.6 9788.899 9771.91 9786.657 9809.235 9812.62 9767.638 9721.595 9554.51 9504.496 9381.819 9338.447 9277.919 9150.202 8891.067 8593.608 8530.443 8333.852 8312.344 8154.748 8011.0864] true [ 7169.06 7028.37 6882.56 6675.72 6509.74 6433.63 6337.44 6391.6 6430.49 6699.06 6909.91 7506.36 8082.77 8525.79 9019.16 9284.05 9491.5 9730.94 9878.81 9983.39 10091.39 10110.35 10147.24 10175.82 10173.61 10206.99 10277.8 10316.92 10343.71 10320.83 10394.41 10425.31 10389.64 10280.01 10109.29 9720.02 9469.66 9099.85 8935.96 8966.7 9020.98 8850.09 8575.67 8547.27 8319.89 8234.43 8131.17 7867.71]
12-16
219 Model: "model_18" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= input_19 (InputLayer) [(None, 48, 3)] 0 _________________________________________________________________ conv1d_35 (Conv1D) (None, 48, 128) 512 _________________________________________________________________ conv1d_36 (Conv1D) (None, 48, 128) 16512 _________________________________________________________________ max_pooling1d_17 (MaxPooling (None, 1, 128) 0 _________________________________________________________________ lstm_52 (LSTM) (None, 1, 128) 131584 _________________________________________________________________ lstm_53 (LSTM) (None, 1, 64) 49408 _________________________________________________________________ lstm_54 (LSTM) (None, 48) 21696 _________________________________________________________________ dense_18 (Dense) (None, 48) 2352 ================================================================= Total params: 222,064 Trainable params: 222,064 Non-trainable params: 0 _________________________________________________________________ Train on 219 samples, validate on 95 samples Epoch 1/30 219/219 [==============================] - 3s 14ms/sample - loss: 0.1732 - accuracy: 0.0046 - val_loss: 0.1575 - val_accuracy: 0.0000e+00 Epoch 2/30 219/219 [==============================] - 0s 215us/sample - loss: 0.1380 - accuracy: 0.0000e+00 - val_loss: 0.1020 - val_accuracy: 0.0000e+00 Epoch 3/30 219/219 [==============================] - 0s 206us/sample - loss: 0.0732 - accuracy: 0.0000e+00 - val_loss: 0.0443 - val_accuracy: 0.0842 Epoch 4/30 219/219 [==============================] - 0s 206us/sample - loss: 0.0333 - accuracy: 0.1050 - val_loss: 0.0212 - val_accuracy: 0.0842 Epoch 5/30 219/219 [==============================] - 0s 210us/sample - loss: 0.0123 - accuracy: 0.1005 - val_loss: 0.0098 - val_accuracy: 0.1895 Epoch 6/30 219/219 [==============================] - 0s 210us/sample - loss: 0.0076 - accuracy: 0.1918 - val_loss: 0.0077 - val_accuracy: 0.1895 Epoch 7/30 219/219 [==============================] - 0s 210us/sample - loss: 0.0055 - accuracy: 0.1826 - val_loss: 0.0066 - val_accuracy: 0.0526 Epoch 8/30 219/219 [==============================] - 0s 210us/sample - loss: 0.0048 - accuracy: 0.0594 - val_loss: 0.0064 - val_accuracy: 0.0947 Epoch 9/30 219/219 [==============================] - 0s 219us/sample - loss: 0.0046 - accuracy: 0.0776 - val_loss: 0.0062 - val_accuracy: 0.0947 Epoch 10/30 219/219 [==============================] - 0s 210us/sample - loss: 0.0044 - accuracy: 0.0776 - val_loss: 0.0060 - val_accuracy: 0.0947 Epoch 11/30 219/219 [==============================] - 0s 215us/sample - loss: 0.0043 - accuracy: 0.0868 - val_loss: 0.0060 - val_accuracy: 0.0526 Epoch 12/30 219/219 [==============================] - 0s 206us/sample - loss: 0.0042 - accuracy: 0.1233 - val_loss: 0.0059 - val_accuracy: 0.1895 Epoch 13/30 219/219 [==============================] - 0s 210us/sample - loss: 0.0042 - accuracy: 0.1826 - val_loss: 0.0059 - val_accuracy: 0.2947 Epoch 14/30 219/219 [==============================] - 0s 210us/sample - loss: 0.0042 - accuracy: 0.1735 - val_loss: 0.0059 - val_accuracy: 0.2947 Epoch 15/30 219/219 [==============================] - 0s 210us/sample - loss: 0.0042 - accuracy: 0.1872 - val_loss: 0.0059 - val_accuracy: 0.1895 Epoch 16/30 219/219 [==============================] - 0s 215us/sample - loss: 0.0042 - accuracy: 0.1918 - val_loss: 0.0059 - val_accuracy: 0.1895 Epoch 17/30 219/219 [==============================] - 0s 219us/sample - loss: 0.0042 - accuracy: 0.1918 - val_loss: 0.0059 - val_accuracy: 0.0947 Epoch 18/30 219/219 [==============================] - 0s 224us/sample - loss: 0.0042 - accuracy: 0.1005 - val_loss: 0.0059 - val_accuracy: 0.0947 Epoch 19/30 219/219 [==============================] - 0s 215us/sample - loss: 0.0042 - accuracy: 0.1233 - val_loss: 0.0059 - val_accuracy: 0.1895 Epoch 20/30 219/219 [==============================] - 0s 219us/sample - loss: 0.0042 - accuracy: 0.1918 - val_loss: 0.0059 - val_accuracy: 0.1895 Epoch 21/30 219/219 [==============================] - 0s 219us/sample - loss: 0.0042 - accuracy: 0.1918 - val_loss: 0.0059 - val_accuracy: 0.1895 Epoch 22/30 219/219 [==============================] - 0s 215us/sample - loss: 0.0042 - accuracy: 0.1918 - val_loss: 0.0059 - val_accuracy: 0.1895 Epoch 23/30 219/219 [==============================] - 0s 215us/sample - loss: 0.0042 - accuracy: 0.1918 - val_loss: 0.0059 - val_accuracy: 0.1895 Epoch 24/30 219/219 [==============================] - 0s 219us/sample - loss: 0.0042 - accuracy: 0.1918 - val_loss: 0.0059 - val_accuracy: 0.1895 Epoch 25/30 219/219 [==============================] - 0s 219us/sample - loss: 0.0042 - accuracy: 0.1918 - val_loss: 0.0059 - val_accuracy: 0.1895 Epoch 26/30 219/219 [==============================] - 0s 210us/sample - loss: 0.0042 - accuracy: 0.1918 - val_loss: 0.0059 - val_accuracy: 0.1895 Epoch 27/30 219/219 [==============================] - 0s 215us/sample - loss: 0.0042 - accuracy: 0.1918 - val_loss: 0.0059 - val_accuracy: 0.1895 Epoch 28/30 219/219 [==============================] - 0s 215us/sample - loss: 0.0042 - accuracy: 0.1918 - val_loss: 0.0059 - val_accuracy: 0.1895 Epoch 29/30 219/219 [==============================] - 0s 215us/sample - loss: 0.0042 - accuracy: 0.1918 - val_loss: 0.0059 - val_accuracy: 0.1895 Epoch 30/30 219/219 [==============================] - 0s 210us/sample - loss: 0.0042 - accuracy: 0.1918 - val_loss: 0.0059 - val_accuracy: 0.1895
pred [ 8010.2812 7795.0703 7564.422 7225.707 6919.785 6712.253 6607.061 6586.321 6634.1426 6830.8213 7080.0835 7604.8574 8131.903 8530.349 8947.871 9210.863 9431.09 9719.999 9876.513 9997.637 10072.125 10105.951 10164.23 10174.132 10195.184 10236.515 10265.087 10291.155 10299.1875 10285.224 10323.4375 10334.313 10327.895 10218.452 10122.957 9863.107 9719.938 9526.191 9484.489 9480.636 9370.866 9098.046 8800.417 8780.977 8572.676 8534.667 8355.627 8160.165 ] true [ 7883.01 7697.8 7454.8 7124.84 6924.24 6775.86 6658.5 6700.81 6729.53 6968.24 7165.2 7764.98 8301.17 8652.23 9112. 9473.62 9749.28 10047.51 10199.95 10421.91 10504.25 10467.54 10474.48 10425.8 10434.05 10387.62 10255.64 10079.61 10005.6 9878.18 9705.36 9582.39 9615.68 9572.51 9594.13 9403.11 9213.98 8993.86 8949.15 8978.68 8994.21 8781.87 8530.68 8483.23 8245.1 8215.07 8083.06 7847.47]
12-17
pred [7953.6826 7764.2075 7525.9487 7193.199 6884.777 6641.638 6540.9766 6501.481 6541.193 6738.235 6985.7256 7513.686 8028.8506 8388.727 8816.574 9045.857 9238.705 9511.588 9636.9 9736.143 9784.97 9792.08 9819.831 9803.825 9798.889 9808.947 9821.329 9825.073 9804.305 9783.639 9813.6455 9838.6045 9832.7295 9741.81 9677.087 9479.081 9365.7705 9237.208 9227.098 9215.348 9102.898 8842.39 8565.39 8563.052 8382.6455 8378.317 8209.04 8044.8027] true [7655.29 7456.22 7226.49 6964.52 6727.05 6546.17 6485.81 6449.02 6545.72 6793.83 7014.99 7557.14 8063.95 8404.65 8775.34 9023.61 9192.9 9427.04 9507.14 9528.89 9494.33 9470.4 9437.37 9383.72 9375.24 9385.37 9357.67 9340.12 9305.39 9321.9 9260.88 9277.83 9270.07 9165.36 9076.46 8773.58 8617.1 8501.93 8360.43 8420.27 8542.9 8353.83 8175.34 8191.65 8097.69 8080.61 7946.84 7741.89]
12-18
pred [7865.2236 7677.1333 7456.1416 7154.1284 6850.285 6626.4424 6489.7876 6419.662 6430.68 6571.7866 6756.8037 7149.8203 7573.453 7936.213 8327.337 8582.044 8804.69 9057.268 9184.601 9315.578 9339.636 9345.116 9357.485 9340.651 9316.209 9318.807 9310.985 9302.852 9274.648 9256.042 9284.229 9294.672 9316.4 9251.395 9233.683 9121.313 9090.658 9043.04 9055.084 9015.442 8912.394 8674.6455 8423.779 8391.762 8221.767 8217.742 8056.0825 7911.022 ] true [7536.2 7320.33 7097.55 6795.59 6507.03 6277.95 6164.62 6115.29 6125.64 6234.22 6269.43 6477.05 6708.51 7009.05 7362.78 7761.16 8027.79 8333.36 8363.98 8438.6 8425.72 8417.9 8372.52 8343.37 8242.85 8191.22 8085.53 8095.87 8022.95 8001.18 7968.75 7915.96 7924.51 7878.38 7902.41 7880.48 7825.89 7750.67 7784.41 7887.17 8004.28 7936.47 7801.48 7805.52 7691.82 7689.13 7490.91 7292.07]
11-19
pred [7967.773 7780.1353 7578.1963 7293.3896 6998.7544 6734.992 6570.999 6492.248 6488.9717 6597.7173 6776.7847 7177.6133 7602.7163 8011.8896 8439.06 8677.58 8916.004 9165.357 9291.33 9420.282 9458.338 9475.561 9491.976 9461.701 9454.392 9461.733 9439.987 9437.684 9413.306 9397.864 9409.572 9406.601 9421.963 9364.303 9351.47 9237.589 9257.813 9243.763 9238.902 9152.531 9017.155 8782.299 8539.245 8467.33 8282.052 8310.971 8160.825 8034.3843] true [7540.35 7358.8 7176.6 6890.91 6623.68 6442.35 6355.86 6334.19 6432.11 6616.44 6924.92 7509.92 8086.14 8486.14 8901.87 9028.32 9097.9 9257.81 9311.52 9341.68 9330.75 9252.11 9174.87 9296.21 9184.21 9127.45 9104.33 9037.21 8908.58 8776.77 8776.51 8841.54 8851.47 8798.11 8733.41 8542.93 8544.56 8468.44 8507.41 8575.36 8434.61 8201.8 7990.06 8040.75 7989.99 8099.74 8030.74 7861.39]
11-20
pred [7658.116 7461.5674 7246.294 6941.2827 6662.401 6459.3604 6351.2134 6302.537 6319.0854 6437.2007 6557.7256 6861.8833 7187.0464 7481.572 7855.0415 8174.0874 8437.384 8719.736 8881.039 9015.198 9058.812 9069.197 9085.506 9077.307 9070.739 9071.924 9054.071 9046.765 9033.168 9035.119 9065.985 9087.845 9114.164 9062.605 9075.727 8981.485 8927.712 8852.261 8839.681 8849.244 8807.555 8598.439 8350.126 8318.162 8157.8555 8114.532 7946.8 7772.266 ] true [7698.96 7498.86 7281.45 6936.39 6676.8 6396.83 6314.37 6267.42 6277.04 6376.66 6391.01 6606.38 6879.9 7200.98 7572.37 7989.62 8264.51 8421.6 8495.47 8544.23 8481.18 8466.99 8380.54 8307.48 8208.27 8103.64 8019.13 8052.48 7998.17 7995.55 7981.04 8033.03 8018.03 7973.22 7977.28 7982.32 7905.84 7847.67 7901.97 7984.08 7928.23 7755.08 7653.65 7576.1 7483.01 7566.4 7456.93 7278.91]
11-21
pred [7335.7456 7160.5 7008.125 6778.8853 6570.6484 6347.791 6231.724 6143.876 6145.4414 6211.49 6343.6177 6595.224 6878.6084 7233.2134 7594.5986 7895.8594 8212.867 8482.55 8653.988 8810.846 8902.306 8922.246 8942.578 8907.583 8874.332 8876.025 8836.406 8817.289 8785.119 8764.53 8757. 8757.213 8775.332 8762.126 8826.847 8838.953 8965.05 9024.0205 8996.978 8893.513 8795.878 8592.542 8351.354 8233.171 8034.6646 8053.8403 7903.154 7786.9644] true [7090.83 6986.06 6760.39 6527.12 6294.61 6139.96 6051.96 6005.29 5980.31 6036. 6021.97 6129.97 6259.3 6515.63 6865.08 7278.56 7571.23 7880.97 8005.78 8125.06 8128.52 8082.69 8052.26 8077.55 8058.23 7995.52 7957.18 7926.25 7888.81 7942.18 7995.95 8030.31 8015.66 7924.17 8007.57 8083.29 8066.84 8028.86 8022.77 8140.54 8081.94 7891.74 7735.92 7670.76 7536.78 7442.11 7272.63 7106.86]
11-22
pred [7326.1045 7127.653 7001.1797 6760.385 6587.647 6357.289 6190.7124 6182.9023 6210.8623 6298.155 6419.5874 6756.05 7086.5205 7490.285 7928.6377 8116.4995 8491.8 8780.574 8960.463 9121.699 9203.788 9246.216 9218.818 9276.804 9183.858 9252.85 9255.539 9259.722 9255.282 9210.178 9192.959 9289.753 9175.495 9242.834 9230.407 9194.7 9254.693 9265.059 9271.058 9078.975 8944.088 8751.835 8494.521 8412.966 8142.4116 8209.342 8005.757 7942.297 ] true [6965.14 6852.4 6719.32 6534.4 6381.95 6278.91 6218.12 6246.06 6341.62 6626.39 6874.83 7428.62 7990.92 8429.33 8758.04 9021.75 9174.34 9379.13 9501.78 9662.41 9684.4 9700.28 9750.52 9687.44 9643.01 9673.54 9754.38 9766.75 9736.53 9736.18 9713.49 9726.78 9691.35 9706.76 9565.39 9381.76 9175.94 8997.81 8910.26 8961.54 8881.13 8608.68 8331.93 8326.58 8210.31 8095.65 7990.08 7809.96]
11-23
219
pred [7920.0327 7727.453 7498.005 7185.2114 6878.8813 6647.562 6509.2666 6460.4077 6487.224 6645.835 6849.7173 7306.449 7780.017 8165.4346 8575.308 8824.612 9033.209 9284.637 9409.561 9526.89 9556.26 9569.138 9598.808 9577.252 9565.444 9579.406 9571.225 9572.5205 9553.926 9536.858 9558.368 9571.38 9585.237 9507.746 9473.623 9326.139 9278.574 9203.591 9185.488 9143.21 9029.894 8775.27 8515.4 8487.965 8315.512 8309.635 8150.843 7995.8057] true [7676.74 7488.34 7284.98 6955.98 6695.74 6565.37 6467.37 6419.44 6535.9 6693.26 6915.04 7480.59 8073.68 8496.22 8859.66 9040.25 9218.58 9422.82 9549.47 9631.52 9691.09 9694.25 9714.88 9721.51 9793.06 9760.63 9741.99 9808.03 9826.22 9778.31 9820.92 9880.44 9872.48 9783.38 9720.19 9433.17 9303.74 9061.45 9019.82 9066.22 8970.96 8717.53 8420.32 8422.73 8254.52 8194.7 8091.06 7880.16]
11-24
345
pred [8004.741 7836.458 7603.202 7314.055 7014.709 6749.5996 6595.0283 6525.7344 6519.7524 6669.235 6852.7886 7258.991 7708.597 8117.655 8555.022 8792.569 9031.577 9288.715 9415.42 9547.265 9597.754 9636.981 9641.2705 9629.287 9617.367 9627.957 9622.893 9613.755 9593.704 9573.136 9595.104 9597.0205 9601.174 9541.243 9515.239 9390.875 9377.776 9336.862 9309.265 9246.4795 9109.968 8886.664 8630.598 8561.573 8355.295 8379.891 8237.742 8093.9297] true [ 7719.93 7555.17 7363.01 7047.69 6746.96 6545.98 6463.73 6456.23 6490.03 6737.62 6967.2 7567.01 8137.38 8556.23 8966.92 9197.37 9357.05 9614.87 9787.37 9877.89 9929.24 9991.64 10069.66 10115.05 10063.61 10120.29 10125.07 10195.46 10183.76 10173.81 10198.73 10228.02 10180.88 10039.71 9911.18 9599.43 9435.21 9241.65 9194.97 9289.21 9153.82 8898.44 8589.48 8558.74 8384.7 8342.53 8145.04 7909.66]
11-27
288
pred [7978.2227 7762.1753 7535.09 7190.468 6873.919 6672.794 6571.751 6531.8135 6578.506 6758.31 6998.714 7524.4707 8029.8286 8400.463 8815.995 9061.025 9268.963 9531.245 9665.14 9779.905 9835.957 9847.31 9885.594 9884.188 9881.698 9909.587 9928.226 9932.07 9915.323 9912.073 9937.959 9955.917 9946.1455 9853.647 9794.922 9575.495 9459.749 9313.353 9293.79 9295.67 9187.785 8927.393 8635.85 8632.782 8445.346 8413.299 8259.434 8080.411 ] true [7775.14 7621.58 7352.58 7055.24 6785.56 6631.69 6515.74 6447.65 6450.86 6504.6 6583.8 6820.93 7074.62 7480.55 7848.92 8329.09 8608.05 8961.58 9017.24 9193.34 9212.99 9222.8 9235.72 9268.29 9166.05 9208.1 9229.47 9191.43 9211.02 9105.96 9056.33 9102.75 9107.86 8917. 8822.83 8712.19 8554.41 8419.46 8377.18 8537.98 8434.52 8251.53 8073.58 8015.31 7965.3 7921.39 7696.89 7437.95]
11-28
165
pred [7622.089 7433.514 7227.8193 6961.206 6706.786 6493.731 6359.1245 6290.1704 6273.657 6345.671 6424.8203 6650.1357 6900.9785 7216.72 7585.784 7916.0874 8222.613 8535.923 8728.1875 8892.483 8960.711 8996.679 9024.557 9025.254 9023.9 9029.322 9012.389 9007.668 8991.39 8992.334 9005.018 9012.712 9023.311 8988.801 9014.838 8950.001 8952.094 8919.148 8881.271 8848.96 8795.48 8595.546 8356.173 8273.891 8101.968 8070.0713 7900.942 7747.8354] true [7239.79 7042.03 6899.01 6665.75 6421.44 6247.71 6241.86 6191.66 6158.63 6208.94 6166.45 6243.24 6440.46 6751.08 7064.03 7437.11 7727.5 7984.53 8152.66 8279.98 8321.88 8311.75 8375.56 8283.17 8275.1 8161.08 8094.9 8105.09 8096.23 8120.49 8081.84 8109.79 8197.5 8205.53 8266.91 8327.65 8263.45 8156.93 8184.06 8324.47 8254.33 8071.07 7900.53 7789.66 7572.89 7470.29 7255.79 7066.29]
11-29
128
pred [7354.361 7192.362 7026.5654 6798.1035 6588.299 6385.22 6259.58 6211.2954 6221.9624 6327.96 6477.664 6815.372 7176.6865 7559.209 7974.569 8252.226 8548.225 8815.762 8991.436 9139.384 9207.862 9247.233 9289.931 9257.233 9256.092 9270.941 9253.998 9245.9795 9226.661 9231.085 9231.876 9241.795 9247.777 9219.032 9247.188 9173.0205 9248.124 9260.431 9225.302 9114.342 8992.472 8779.633 8508.688 8398.28 8183.7065 8188.5264 8042.2544 7928.874 ] true [6992.16 6843.24 6716.09 6553.76 6352.7 6281.24 6249.09 6297.8 6281.89 6527.21 6776.26 7331.34 7877.59 8387.56 8791.79 9068.34 9195.86 9357.81 9482.36 9557.18 9506.76 9494.25 9547.87 9446.35 9366.44 9367.7 9388.01 9300.02 9268.14 9263.74 9228.2 9271.1 9342.75 9400.46 9359.57 9250.46 9093.62 8921.77 8819.34 8845.23 8627.71 8368.82 8059.55 8030.84 7823.82 7809.72 7743.59 7599.99]
11-30
148
pred [7841.304 7647.1177 7420.666 7113.2344 6810.6753 6582.8794 6456.209 6397.4062 6414.656 6564.199 6744.2104 7152.8975 7581.1265 7936.505 8336.705 8600.568 8823.452 9084.268 9217.222 9339.518 9379.849 9387.69 9405.911 9389.394 9383.671 9394.063 9385.995 9378.815 9359.981 9353.944 9380.819 9390.11 9409.569 9350.326 9320.608 9178.522 9126.161 9050.496 9030.646 9005.092 8924.07 8697.836 8448.34 8419.407 8254.421 8237.489 8079.025 7921.785 ] true [7486.74 7365.91 7196.1 6904.89 6589.38 6427.24 6344.64 6309.51 6446.28 6672.93 6930.36 7518.19 8137.44 8502.02 8907.83 9053.67 9127.21 9290.15 9398.54 9471.39 9518.63 9512.94 9475.56 9524.75 9459.6 9450.12 9433.43 9424.38 9422.29 9362.09 9414.97 9402.16 9475.24 9436.82 9434.39 9152.26 9108.13 8907.29 8900.34 8932.27 8759.71 8487. 8211.04 8264.15 8009.86 8072.83 7969.69 7781.7 ]