DL_eq_picking.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "e1a28737-d045-4b6e-a723-154458d02c88",
   "metadata": {},
   "source": [
    "# Here is a small-scale example for training deep-learning models for P/S wave arrival picking"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "e6628b03-11e1-4ae5-820c-d9360626946f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# References:\n",
    "# Cui, Y., M. Bai, J. Wu, and Y. Chen, 2025, Earthquake signal detection using a multiscale feature fusion network with hybrid attention mechanism, GJI, 988-1008\n",
    "# Saad, O.M., Chen, Y.F., Siervo, D., Zhang, F., Savvaidis, A., Huang, G., Igonin, N., Fomel, S., and Chen, Y., (2023), as a highly accurate and generalizable machine-learning seismic event detection algorithm, IEEE TGRS.\n",
    "# Chen et al., 2024, Near Real‐Time Earthquake Monitoring in Texas Using the Highly Precise Deep Learning Phase Picker, Earth and Space Science, 11, e2024EA003890.\n",
    "# "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "64d6d1e7-0c7e-4890-b529-1b74a8dee8f8",
   "metadata": {},
   "outputs": [],
   "source": [
    "#conda activate eqp (eqpolarity or eqpicking environment)\n",
    "\n",
    "# First set up the environment and install the dependency packages (tested on CPU machine)\n",
    "# \n",
    "# conda create -n eqp python=3.11.7\n",
    "# conda activate eqp\n",
    "# conda install ipython notebook\n",
    "# pip install matplotlib==3.8.0 tensorflow==2.14.0 scikit-learn==1.2.2 seaborn==0.13.2 \n",
    "# \n",
    "# Or, on machine with a GPU setup, use the following commands\n",
    "# conda create -n eqp python=3.10.14\n",
    "# conda activate eqp\n",
    "# conda install ipython notebook\n",
    "# pip install matplotlib==3.8.0 tensorflow==2.11.0 scikit-learn==1.2.2 seaborn==0.13.2 \n",
    "# conda install -c anaconda cudnn==8.2.1\n",
    "# Make sure LD_LIBRARY_PATH includes ~/anaconda3/envs/eqp/lib (libcudnn.so.8)\n",
    "\n",
    "\n",
    "# import library\n",
    "import numpy as np\n",
    "import h5py\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e25fc895-c20c-451e-87d4-476140f4bb40",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['texnet2022nlpv_PB31_EV' 'texnet2021bbcd_POST_EV'\n",
      " 'texnet2020znsu_OG02_EV' 'texnet2021ojgr_PB16_EV'\n",
      " 'texnet2021ghba_EF04_EV' 'texnet2020yuzo_MB07_EV'\n",
      " 'texnet2021gqdj_MB07_EV' 'texnet2022mqlw_PB03_EV'\n",
      " 'texnet2023ihss_PB44_EV' 'texnet2021wmup_PB09_EV']\n"
     ]
    }
   ],
   "source": [
    "# Load the TXED dataset\n",
    "# please down load the TXED from: https://drive.google.com/drive/folders/1WXVB8ytNB4bOaZ97oq6OmMRyAEg95trp?usp=sharing\n",
    "f_txed = h5py.File(\"/Users/chenyk/DATALIB/TXED/TXED_20231111.h5\", 'r')\n",
    "txed_signal_id = np.load(\"/Users/chenyk/DATALIB/TXED/ID_20231111.npy\", 'r')[0:100000].copy()\n",
    "np.random.seed(20232425)\n",
    "np.random.shuffle(txed_signal_id)\n",
    "txed_signal_id=txed_signal_id[0:10000]\n",
    "print(txed_signal_id[0:10]) #see if the random IDs are the same\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "167f0c57-4789-475b-9b2b-f93bc511baf4",
   "metadata": {},
   "source": [
    "# Design a DL architecture for P/S arrival picking"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "3080486a-b06f-4a9e-9bb3-4390a4f7015b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "__________________________________________________________________________________________________\n",
      " Layer (type)                Output Shape                 Param #   Connected to                  \n",
      "==================================================================================================\n",
      " input_1 (InputLayer)        [(None, 6000, 3)]            0         []                            \n",
      "                                                                                                  \n",
      " dense (Dense)               (None, 6000, 64)             256       ['input_1[0][0]']             \n",
      "                                                                                                  \n",
      " conv1d (Conv1D)             (None, 3000, 64)             12352     ['dense[0][0]']               \n",
      "                                                                                                  \n",
      " conv1d_1 (Conv1D)           (None, 3000, 64)             24640     ['dense[0][0]']               \n",
      "                                                                                                  \n",
      " conv1d_2 (Conv1D)           (None, 3000, 64)             36928     ['dense[0][0]']               \n",
      "                                                                                                  \n",
      " concatenate (Concatenate)   (None, 3000, 192)            0         ['conv1d[0][0]',              \n",
      "                                                                     'conv1d_1[0][0]',            \n",
      "                                                                     'conv1d_2[0][0]']            \n",
      "                                                                                                  \n",
      " leaky_re_lu (LeakyReLU)     (None, 3000, 192)            0         ['concatenate[0][0]']         \n",
      "                                                                                                  \n",
      " dropout (Dropout)           (None, 3000, 192)            0         ['leaky_re_lu[0][0]']         \n",
      "                                                                                                  \n",
      " global_average_pooling1d (  (None, 192)                  0         ['dropout[0][0]']             \n",
      " GlobalAveragePooling1D)                                                                          \n",
      "                                                                                                  \n",
      " dense_1 (Dense)             (None, 12)                   2316      ['global_average_pooling1d[0][\n",
      "                                                                    0]']                          \n",
      "                                                                                                  \n",
      " dense_2 (Dense)             (None, 192)                  2496      ['dense_1[0][0]']             \n",
      "                                                                                                  \n",
      " reshape (Reshape)           (None, 1, 192)               0         ['dense_2[0][0]']             \n",
      "                                                                                                  \n",
      " multiply (Multiply)         (None, 3000, 192)            0         ['dropout[0][0]',             \n",
      "                                                                     'reshape[0][0]']             \n",
      "                                                                                                  \n",
      " lambda (Lambda)             (None, 3000)                 0         ['multiply[0][0]']            \n",
      "                                                                                                  \n",
      " lambda_1 (Lambda)           (None, 3000)                 0         ['multiply[0][0]']            \n",
      "                                                                                                  \n",
      " concatenate_1 (Concatenate  (None, 6000)                 0         ['lambda[0][0]',              \n",
      " )                                                                   'lambda_1[0][0]']            \n",
      "                                                                                                  \n",
      " reshape_1 (Reshape)         (None, 3000, 2)              0         ['concatenate_1[0][0]']       \n",
      "                                                                                                  \n",
      " conv1d_3 (Conv1D)           (None, 3000, 1)              7         ['reshape_1[0][0]']           \n",
      "                                                                                                  \n",
      " reshape_2 (Reshape)         (None, 3000, 1)              0         ['conv1d_3[0][0]']            \n",
      "                                                                                                  \n",
      " multiply_1 (Multiply)       (None, 3000, 192)            0         ['multiply[0][0]',            \n",
      "                                                                     'reshape_2[0][0]']           \n",
      "                                                                                                  \n",
      " add (Add)                   (None, 3000, 192)            0         ['multiply[0][0]',            \n",
      "                                                                     'multiply_1[0][0]']          \n",
      "                                                                                                  \n",
      " dense_3 (Dense)             (None, 3000, 32)             6176      ['add[0][0]']                 \n",
      "                                                                                                  \n",
      " conv1d_4 (Conv1D)           (None, 1500, 32)             3104      ['dense_3[0][0]']             \n",
      "                                                                                                  \n",
      " conv1d_5 (Conv1D)           (None, 1500, 32)             6176      ['dense_3[0][0]']             \n",
      "                                                                                                  \n",
      " conv1d_6 (Conv1D)           (None, 1500, 32)             9248      ['dense_3[0][0]']             \n",
      "                                                                                                  \n",
      " concatenate_2 (Concatenate  (None, 1500, 96)             0         ['conv1d_4[0][0]',            \n",
      " )                                                                   'conv1d_5[0][0]',            \n",
      "                                                                     'conv1d_6[0][0]']            \n",
      "                                                                                                  \n",
      " leaky_re_lu_1 (LeakyReLU)   (None, 1500, 96)             0         ['concatenate_2[0][0]']       \n",
      "                                                                                                  \n",
      " dropout_1 (Dropout)         (None, 1500, 96)             0         ['leaky_re_lu_1[0][0]']       \n",
      "                                                                                                  \n",
      " global_average_pooling1d_1  (None, 96)                   0         ['dropout_1[0][0]']           \n",
      "  (GlobalAveragePooling1D)                                                                        \n",
      "                                                                                                  \n",
      " dense_4 (Dense)             (None, 6)                    582       ['global_average_pooling1d_1[0\n",
      "                                                                    ][0]']                        \n",
      "                                                                                                  \n",
      " dense_5 (Dense)             (None, 96)                   672       ['dense_4[0][0]']             \n",
      "                                                                                                  \n",
      " reshape_3 (Reshape)         (None, 1, 96)                0         ['dense_5[0][0]']             \n",
      "                                                                                                  \n",
      " multiply_2 (Multiply)       (None, 1500, 96)             0         ['dropout_1[0][0]',           \n",
      "                                                                     'reshape_3[0][0]']           \n",
      "                                                                                                  \n",
      " lambda_2 (Lambda)           (None, 1500)                 0         ['multiply_2[0][0]']          \n",
      "                                                                                                  \n",
      " lambda_3 (Lambda)           (None, 1500)                 0         ['multiply_2[0][0]']          \n",
      "                                                                                                  \n",
      " concatenate_3 (Concatenate  (None, 3000)                 0         ['lambda_2[0][0]',            \n",
      " )                                                                   'lambda_3[0][0]']            \n",
      "                                                                                                  \n",
      " reshape_4 (Reshape)         (None, 1500, 2)              0         ['concatenate_3[0][0]']       \n",
      "                                                                                                  \n",
      " conv1d_7 (Conv1D)           (None, 1500, 1)              7         ['reshape_4[0][0]']           \n",
      "                                                                                                  \n",
      " reshape_5 (Reshape)         (None, 1500, 1)              0         ['conv1d_7[0][0]']            \n",
      "                                                                                                  \n",
      " multiply_3 (Multiply)       (None, 1500, 96)             0         ['multiply_2[0][0]',          \n",
      "                                                                     'reshape_5[0][0]']           \n",
      "                                                                                                  \n",
      " add_1 (Add)                 (None, 1500, 96)             0         ['multiply_2[0][0]',          \n",
      "                                                                     'multiply_3[0][0]']          \n",
      "                                                                                                  \n",
      " dense_6 (Dense)             (None, 1500, 16)             1552      ['add_1[0][0]']               \n",
      "                                                                                                  \n",
      " conv1d_8 (Conv1D)           (None, 750, 16)              784       ['dense_6[0][0]']             \n",
      "                                                                                                  \n",
      " conv1d_9 (Conv1D)           (None, 750, 16)              1552      ['dense_6[0][0]']             \n",
      "                                                                                                  \n",
      " conv1d_10 (Conv1D)          (None, 750, 16)              2320      ['dense_6[0][0]']             \n",
      "                                                                                                  \n",
      " concatenate_4 (Concatenate  (None, 750, 48)              0         ['conv1d_8[0][0]',            \n",
      " )                                                                   'conv1d_9[0][0]',            \n",
      "                                                                     'conv1d_10[0][0]']           \n",
      "                                                                                                  \n",
      " leaky_re_lu_2 (LeakyReLU)   (None, 750, 48)              0         ['concatenate_4[0][0]']       \n",
      "                                                                                                  \n",
      " dropout_2 (Dropout)         (None, 750, 48)              0         ['leaky_re_lu_2[0][0]']       \n",
      "                                                                                                  \n",
      " dense_9 (Dense)             (None, 750, 8)               392       ['dropout_2[0][0]']           \n",
      "                                                                                                  \n",
      " conv1d_12 (Conv1D)          (None, 375, 8)               200       ['dense_9[0][0]']             \n",
      "                                                                                                  \n",
      " conv1d_13 (Conv1D)          (None, 375, 8)               392       ['dense_9[0][0]']             \n",
      "                                                                                                  \n",
      " conv1d_14 (Conv1D)          (None, 375, 8)               584       ['dense_9[0][0]']             \n",
      "                                                                                                  \n",
      " concatenate_6 (Concatenate  (None, 375, 24)              0         ['conv1d_12[0][0]',           \n",
      " )                                                                   'conv1d_13[0][0]',           \n",
      "                                                                     'conv1d_14[0][0]']           \n",
      "                                                                                                  \n",
      " leaky_re_lu_3 (LeakyReLU)   (None, 375, 24)              0         ['concatenate_6[0][0]']       \n",
      "                                                                                                  \n",
      " dropout_3 (Dropout)         (None, 375, 24)              0         ['leaky_re_lu_3[0][0]']       \n",
      "                                                                                                  \n",
      " global_average_pooling1d_3  (None, 24)                   0         ['dropout_3[0][0]']           \n",
      "  (GlobalAveragePooling1D)                                                                        \n",
      "                                                                                                  \n",
      " dense_10 (Dense)            (None, 1)                    25        ['global_average_pooling1d_3[0\n",
      "                                                                    ][0]']                        \n",
      "                                                                                                  \n",
      " dense_11 (Dense)            (None, 24)                   48        ['dense_10[0][0]']            \n",
      "                                                                                                  \n",
      " reshape_9 (Reshape)         (None, 1, 24)                0         ['dense_11[0][0]']            \n",
      "                                                                                                  \n",
      " multiply_6 (Multiply)       (None, 375, 24)              0         ['dropout_3[0][0]',           \n",
      "                                                                     'reshape_9[0][0]']           \n",
      "                                                                                                  \n",
      " lambda_6 (Lambda)           (None, 375)                  0         ['multiply_6[0][0]']          \n",
      "                                                                                                  \n",
      " lambda_7 (Lambda)           (None, 375)                  0         ['multiply_6[0][0]']          \n",
      "                                                                                                  \n",
      " concatenate_7 (Concatenate  (None, 750)                  0         ['lambda_6[0][0]',            \n",
      " )                                                                   'lambda_7[0][0]']            \n",
      "                                                                                                  \n",
      " dense_13 (Dense)            (None, 375, 8)               200       ['dropout_3[0][0]']           \n",
      "                                                                                                  \n",
      " reshape_10 (Reshape)        (None, 375, 2)               0         ['concatenate_7[0][0]']       \n",
      "                                                                                                  \n",
      " global_average_pooling1d_2  (None, 48)                   0         ['dropout_2[0][0]']           \n",
      "  (GlobalAveragePooling1D)                                                                        \n",
      "                                                                                                  \n",
      " conv1d_16 (Conv1D)          (None, 375, 8)               200       ['dense_13[0][0]']            \n",
      "                                                                                                  \n",
      " conv1d_17 (Conv1D)          (None, 375, 8)               392       ['dense_13[0][0]']            \n",
      "                                                                                                  \n",
      " conv1d_18 (Conv1D)          (None, 375, 8)               584       ['dense_13[0][0]']            \n",
      "                                                                                                  \n",
      " conv1d_15 (Conv1D)          (None, 375, 1)               7         ['reshape_10[0][0]']          \n",
      "                                                                                                  \n",
      " dense_7 (Dense)             (None, 3)                    147       ['global_average_pooling1d_2[0\n",
      "                                                                    ][0]']                        \n",
      "                                                                                                  \n",
      " concatenate_8 (Concatenate  (None, 375, 24)              0         ['conv1d_16[0][0]',           \n",
      " )                                                                   'conv1d_17[0][0]',           \n",
      "                                                                     'conv1d_18[0][0]']           \n",
      "                                                                                                  \n",
      " reshape_11 (Reshape)        (None, 375, 1)               0         ['conv1d_15[0][0]']           \n",
      "                                                                                                  \n",
      " dense_8 (Dense)             (None, 48)                   192       ['dense_7[0][0]']             \n",
      "                                                                                                  \n",
      " leaky_re_lu_4 (LeakyReLU)   (None, 375, 24)              0         ['concatenate_8[0][0]']       \n",
      "                                                                                                  \n",
      " multiply_7 (Multiply)       (None, 375, 24)              0         ['multiply_6[0][0]',          \n",
      "                                                                     'reshape_11[0][0]']          \n",
      "                                                                                                  \n",
      " reshape_6 (Reshape)         (None, 1, 48)                0         ['dense_8[0][0]']             \n",
      "                                                                                                  \n",
      " batch_normalization (Batch  (None, 375, 24)              96        ['leaky_re_lu_4[0][0]']       \n",
      " Normalization)                                                                                   \n",
      "                                                                                                  \n",
      " add_3 (Add)                 (None, 375, 24)              0         ['multiply_6[0][0]',          \n",
      "                                                                     'multiply_7[0][0]']          \n",
      "                                                                                                  \n",
      " multiply_4 (Multiply)       (None, 750, 48)              0         ['dropout_2[0][0]',           \n",
      "                                                                     'reshape_6[0][0]']           \n",
      "                                                                                                  \n",
      " concatenate_9 (Concatenate  (None, 375, 48)              0         ['batch_normalization[0][0]', \n",
      " )                                                                   'add_3[0][0]']               \n",
      "                                                                                                  \n",
      " lambda_4 (Lambda)           (None, 750)                  0         ['multiply_4[0][0]']          \n",
      "                                                                                                  \n",
      " lambda_5 (Lambda)           (None, 750)                  0         ['multiply_4[0][0]']          \n",
      "                                                                                                  \n",
      " up_sampling1d (UpSampling1  (None, 750, 48)              0         ['concatenate_9[0][0]']       \n",
      " D)                                                                                               \n",
      "                                                                                                  \n",
      " concatenate_5 (Concatenate  (None, 1500)                 0         ['lambda_4[0][0]',            \n",
      " )                                                                   'lambda_5[0][0]']            \n",
      "                                                                                                  \n",
      " dense_14 (Dense)            (None, 750, 16)              784       ['up_sampling1d[0][0]']       \n",
      "                                                                                                  \n",
      " reshape_7 (Reshape)         (None, 750, 2)               0         ['concatenate_5[0][0]']       \n",
      "                                                                                                  \n",
      " conv1d_19 (Conv1D)          (None, 750, 16)              784       ['dense_14[0][0]']            \n",
      "                                                                                                  \n",
      " conv1d_20 (Conv1D)          (None, 750, 16)              1552      ['dense_14[0][0]']            \n",
      "                                                                                                  \n",
      " conv1d_21 (Conv1D)          (None, 750, 16)              2320      ['dense_14[0][0]']            \n",
      "                                                                                                  \n",
      " conv1d_11 (Conv1D)          (None, 750, 1)               7         ['reshape_7[0][0]']           \n",
      "                                                                                                  \n",
      " concatenate_10 (Concatenat  (None, 750, 48)              0         ['conv1d_19[0][0]',           \n",
      " e)                                                                  'conv1d_20[0][0]',           \n",
      "                                                                     'conv1d_21[0][0]']           \n",
      "                                                                                                  \n",
      " reshape_8 (Reshape)         (None, 750, 1)               0         ['conv1d_11[0][0]']           \n",
      "                                                                                                  \n",
      " leaky_re_lu_5 (LeakyReLU)   (None, 750, 48)              0         ['concatenate_10[0][0]']      \n",
      "                                                                                                  \n",
      " multiply_5 (Multiply)       (None, 750, 48)              0         ['multiply_4[0][0]',          \n",
      "                                                                     'reshape_8[0][0]']           \n",
      "                                                                                                  \n",
      " batch_normalization_1 (Bat  (None, 750, 48)              192       ['leaky_re_lu_5[0][0]']       \n",
      " chNormalization)                                                                                 \n",
      "                                                                                                  \n",
      " add_2 (Add)                 (None, 750, 48)              0         ['multiply_4[0][0]',          \n",
      "                                                                     'multiply_5[0][0]']          \n",
      "                                                                                                  \n",
      " concatenate_11 (Concatenat  (None, 750, 96)              0         ['batch_normalization_1[0][0]'\n",
      " e)                                                                 , 'add_2[0][0]']              \n",
      "                                                                                                  \n",
      " up_sampling1d_1 (UpSamplin  (None, 1500, 96)             0         ['concatenate_11[0][0]']      \n",
      " g1D)                                                                                             \n",
      "                                                                                                  \n",
      " dense_15 (Dense)            (None, 1500, 32)             3104      ['up_sampling1d_1[0][0]']     \n",
      "                                                                                                  \n",
      " conv1d_22 (Conv1D)          (None, 1500, 32)             3104      ['dense_15[0][0]']            \n",
      "                                                                                                  \n",
      " conv1d_23 (Conv1D)          (None, 1500, 32)             6176      ['dense_15[0][0]']            \n",
      "                                                                                                  \n",
      " conv1d_24 (Conv1D)          (None, 1500, 32)             9248      ['dense_15[0][0]']            \n",
      "                                                                                                  \n",
      " concatenate_12 (Concatenat  (None, 1500, 96)             0         ['conv1d_22[0][0]',           \n",
      " e)                                                                  'conv1d_23[0][0]',           \n",
      "                                                                     'conv1d_24[0][0]']           \n",
      "                                                                                                  \n",
      " leaky_re_lu_6 (LeakyReLU)   (None, 1500, 96)             0         ['concatenate_12[0][0]']      \n",
      "                                                                                                  \n",
      " batch_normalization_2 (Bat  (None, 1500, 96)             384       ['leaky_re_lu_6[0][0]']       \n",
      " chNormalization)                                                                                 \n",
      "                                                                                                  \n",
      " concatenate_13 (Concatenat  (None, 1500, 192)            0         ['batch_normalization_2[0][0]'\n",
      " e)                                                                 , 'add_1[0][0]']              \n",
      "                                                                                                  \n",
      " up_sampling1d_2 (UpSamplin  (None, 3000, 192)            0         ['concatenate_13[0][0]']      \n",
      " g1D)                                                                                             \n",
      "                                                                                                  \n",
      " dense_16 (Dense)            (None, 3000, 64)             12352     ['up_sampling1d_2[0][0]']     \n",
      "                                                                                                  \n",
      " conv1d_25 (Conv1D)          (None, 3000, 64)             12352     ['dense_16[0][0]']            \n",
      "                                                                                                  \n",
      " conv1d_26 (Conv1D)          (None, 3000, 64)             24640     ['dense_16[0][0]']            \n",
      "                                                                                                  \n",
      " conv1d_27 (Conv1D)          (None, 3000, 64)             36928     ['dense_16[0][0]']            \n",
      "                                                                                                  \n",
      " concatenate_14 (Concatenat  (None, 3000, 192)            0         ['conv1d_25[0][0]',           \n",
      " e)                                                                  'conv1d_26[0][0]',           \n",
      "                                                                     'conv1d_27[0][0]']           \n",
      "                                                                                                  \n",
      " leaky_re_lu_7 (LeakyReLU)   (None, 3000, 192)            0         ['concatenate_14[0][0]']      \n",
      "                                                                                                  \n",
      " batch_normalization_3 (Bat  (None, 3000, 192)            768       ['leaky_re_lu_7[0][0]']       \n",
      " chNormalization)                                                                                 \n",
      "                                                                                                  \n",
      " concatenate_15 (Concatenat  (None, 3000, 384)            0         ['batch_normalization_3[0][0]'\n",
      " e)                                                                 , 'add[0][0]']                \n",
      "                                                                                                  \n",
      " up_sampling1d_3 (UpSamplin  (None, 6000, 384)            0         ['concatenate_15[0][0]']      \n",
      " g1D)                                                                                             \n",
      "                                                                                                  \n",
      " dense_17 (Dense)            (None, 6000, 64)             24640     ['up_sampling1d_3[0][0]']     \n",
      "                                                                                                  \n",
      " conv1d_28 (Conv1D)          (None, 6000, 2)              386       ['dense_17[0][0]']            \n",
      "                                                                                                  \n",
      "==================================================================================================\n",
      "Total params: 254348 (993.55 KB)\n",
      "Trainable params: 253628 (990.73 KB)\n",
      "Non-trainable params: 720 (2.81 KB)\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# Below example is from \n",
    "# Cui, Y., M. Bai, J. Wu, and Y. Chen, 2025, Earthquake signal detection using a multiscale feature fusion network with hybrid attention mechanism, GJI, 988-1008\n",
    "#\n",
    "\n",
    "from keras.layers import Input, Dense, Conv1D, Dropout, Activation, Flatten, RepeatVector\n",
    "from keras.callbacks import EarlyStopping, ModelCheckpoint, ReduceLROnPlateau,LearningRateScheduler, CSVLogger\n",
    "from keras.layers import concatenate,LeakyReLU,GlobalAveragePooling1D,Multiply,Reshape,Lambda,Add,GRU\n",
    "from keras.layers import BatchNormalization,UpSampling1D\n",
    "from keras.models import Model\n",
    "# from keras.optimizers import Adam\n",
    "from keras.optimizers.legacy import Adam\n",
    "from keras import backend as K\n",
    "\n",
    "D1, D2, D3, D4 = 64, 32, 16, 8 # number of filters in different layers\n",
    "k1, k2, k3 = 3, 6, 9 # kernel sizes in MSFF model\n",
    "\n",
    "# define hybrids attention model\n",
    "def channel_attention(input_tensor, reduction_ratio=16):\n",
    "    channel_avg = GlobalAveragePooling1D()(input_tensor)\n",
    "    channel_dense1 = Dense(units=int(input_tensor.shape[-1]) // reduction_ratio, activation='relu')(channel_avg)\n",
    "    channel_dense2 = Dense(units=int(input_tensor.shape[-1]), activation='sigmoid')(channel_dense1)\n",
    "    channel_attention = Multiply()([input_tensor, Reshape(target_shape=(1, -1))(channel_dense2)])\n",
    "    return channel_attention\n",
    "\n",
    "def spatial_attention(input_tensor):\n",
    "    spatial_avg = Lambda(lambda x: K.mean(x, axis=2))(input_tensor)\n",
    "    spatial_max = Lambda(lambda x: K.max(x, axis=2))(input_tensor)\n",
    "    spatial_concat = Reshape(target_shape=(-1, 2))(concatenate([spatial_avg, spatial_max],axis=1))\n",
    "    spatial_conv1 = Conv1D(filters=1, kernel_size=3, padding='same', activation='sigmoid')(spatial_concat)\n",
    "    spatial_attention = Multiply()([input_tensor, Reshape(target_shape=(-1, 1))(spatial_conv1)])\n",
    "    return spatial_attention\n",
    "\n",
    "def cbam_block(input_tensor):\n",
    "    channel_att = channel_attention(input_tensor)\n",
    "    spatial_att = spatial_attention(channel_att)\n",
    "    cbam_output = Add()([channel_att, spatial_att])\n",
    "    return cbam_output\n",
    "    \n",
    "# define encoder block\n",
    "inp = Input(shape=(6000, 3))\n",
    "\n",
    "# encoder block\n",
    "encoder1 = Dense(D1, activation='relu')(inp)\n",
    "mixed_1 = Conv1D(D1, kernel_size=k1, strides=2, padding='same', activation='linear')(encoder1)\n",
    "mixed_2 = Conv1D(D1, kernel_size=k2, strides=2, padding='same', activation='linear')(encoder1)\n",
    "mixed_3 = Conv1D(D1, kernel_size=k3, strides=2, padding='same', activation='linear')(encoder1)\n",
    "encoder1 = concatenate([mixed_1, mixed_2, mixed_3], axis=-1)\n",
    "encoder1 = LeakyReLU(0.2)(encoder1)\n",
    "encoder1 = Dropout(0.1)(encoder1)\n",
    "\n",
    "attention1 = cbam_block(encoder1)\n",
    "\n",
    "encoder2 = Dense(D2, activation='relu')(attention1)\n",
    "encoder2_0 = Conv1D(D2, kernel_size=k1, strides=2, padding='same')(encoder2)\n",
    "encoder2_1 = Conv1D(D2, kernel_size=k2, strides=2, padding='same')(encoder2)\n",
    "encoder2_2 = Conv1D(D2, kernel_size=k3, strides=2, padding='same')(encoder2)\n",
    "encoder2 = concatenate([encoder2_0, encoder2_1, encoder2_2], axis=-1)\n",
    "encoder2 = LeakyReLU(0.2)(encoder2)\n",
    "encoder2 = Dropout(0.1)(encoder2)\n",
    "\n",
    "attention2 = cbam_block(encoder2)\n",
    "\n",
    "encoder3 = Dense(D3, activation='relu')(attention2)\n",
    "encoder3_0 = Conv1D(D3, kernel_size=k1, strides=2, padding='same')(encoder3)\n",
    "encoder3_1 = Conv1D(D3, kernel_size=k2, strides=2, padding='same')(encoder3)\n",
    "encoder3_2 = Conv1D(D3, kernel_size=k3, strides=2, padding='same')(encoder3)\n",
    "encoder3 = concatenate([encoder3_0, encoder3_1, encoder3_2], axis=-1)\n",
    "encoder3 = LeakyReLU(0.2)(encoder3)\n",
    "encoder3 = Dropout(0.1)(encoder3)\n",
    "\n",
    "attention3 = cbam_block(encoder3)\n",
    "\n",
    "encoder4 = Dense(D4, activation='relu')(encoder3)\n",
    "encoder4_0 = Conv1D(D4, kernel_size=k1, strides=2, padding='same')(encoder4)\n",
    "encoder4_1 = Conv1D(D4, kernel_size=k2, strides=2, padding='same')(encoder4)\n",
    "encoder4_2 = Conv1D(D4, kernel_size=k3, strides=2, padding='same')(encoder4)\n",
    "encoder4 = concatenate([encoder4_0, encoder4_1, encoder4_2], axis=-1)\n",
    "encoder4 = LeakyReLU(0.2)(encoder4)\n",
    "encoder4 = Dropout(0.1)(encoder4)\n",
    "\n",
    "attention4 = cbam_block(encoder4)\n",
    "\n",
    "# define bottleneck block\n",
    "bottleneck = GRU(units=32, activation='tanh', return_sequences=True)(attention4) \n",
    "bottleneck = GRU(units=8, activation='tanh', return_sequences=True)(bottleneck)  \n",
    "bottleneck = Flatten()(bottleneck)\n",
    "encoder = Dense(2, activation='softmax')(bottleneck)\n",
    "\n",
    "\n",
    "# define the classification model in latent layer\n",
    "# Clas_model = Model(inputs=inp, outputs=encoder, name='encoder')\n",
    "# Clas_model.summary()\n",
    "\n",
    "# define decoder blocks\n",
    "decoder4 = Dense(D4, activation='relu')(encoder4)\n",
    "decoder4_0 = Conv1D(D4, k1, padding='same')(decoder4)\n",
    "decoder4_1 = Conv1D(D4, k2, padding='same')(decoder4)\n",
    "decoder4_2 = Conv1D(D4, k3, padding='same')(decoder4)\n",
    "decoder4 = concatenate([decoder4_0, decoder4_1, decoder4_2], axis=-1)\n",
    "decoder4 = LeakyReLU(0.2)(decoder4)\n",
    "decoder4 = BatchNormalization()(decoder4)\n",
    "decoder4 = concatenate([decoder4, attention4], axis=-1)\n",
    "decoder4 = UpSampling1D(2)(decoder4)\n",
    "\n",
    "decoder3 = Dense(D3, activation='relu')(decoder4)\n",
    "decoder3_0 = Conv1D(D3, k1, padding='same')(decoder3)\n",
    "decoder3_1 = Conv1D(D3, k2, padding='same')(decoder3)\n",
    "decoder3_2 = Conv1D(D3, k3, padding='same')(decoder3)\n",
    "decoder3 = concatenate([decoder3_0, decoder3_1, decoder3_2], axis=-1)\n",
    "decoder3 = LeakyReLU(0.2)(decoder3)\n",
    "decoder3 = BatchNormalization()(decoder3)\n",
    "decoder3 = concatenate([decoder3, attention3], axis=-1)\n",
    "decoder3 = UpSampling1D(2)(decoder3)\n",
    "\n",
    "decoder2 = Dense(D2, activation='relu')(decoder3)\n",
    "decoder2_0 = Conv1D(D2, k1, padding='same')(decoder2)\n",
    "decoder2_1 = Conv1D(D2, k2, padding='same')(decoder2)\n",
    "decoder2_2 = Conv1D(D2, k3, padding='same')(decoder2)\n",
    "decoder2 = concatenate([decoder2_0, decoder2_1, decoder2_2], axis=-1)\n",
    "decoder2 = LeakyReLU(0.2)(decoder2)\n",
    "decoder2 = BatchNormalization()(decoder2)\n",
    "decoder2 = concatenate([decoder2, attention2], axis=-1)\n",
    "decoder2 = UpSampling1D(2)(decoder2)\n",
    "\n",
    "decoder1 = Dense(D1, activation='relu')(decoder2)\n",
    "decoder1_0 = Conv1D(D1, k1, padding='same')(decoder1)\n",
    "decoder1_1 = Conv1D(D1, k2, padding='same')(decoder1)\n",
    "decoder1_2 = Conv1D(D1, k3, padding='same')(decoder1)\n",
    "decoder1 = concatenate([decoder1_0, decoder1_1, decoder1_2], axis=-1)\n",
    "decoder1 = LeakyReLU(0.2)(decoder1)\n",
    "decoder1 = BatchNormalization()(decoder1)\n",
    "decoder1 = concatenate([decoder1, attention1], axis=-1)\n",
    "decoder1 = UpSampling1D(2)(decoder1)\n",
    "\n",
    "\n",
    "\n",
    "# define the output block of the phase-picking network\n",
    "decoder1 = Dense(D1, activation='relu')(decoder1)\n",
    "decoder = Conv1D(2, 3, padding='same', activation='sigmoid')(decoder1)\n",
    "\n",
    "# define the picking model using all layers\n",
    "picking_model = Model(inputs=inp, outputs=decoder)\n",
    "picking_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "d8c3a297",
   "metadata": {},
   "outputs": [],
   "source": [
    "## Define a band-pass filter\n",
    "\n",
    "from scipy.signal import butter, lfilter, lfilter_zi\n",
    "\n",
    "def butter_bandpass(lowcut, highcut, fs, order=5):\n",
    "    nyq = 0.5 * fs\n",
    "    low = lowcut / nyq\n",
    "    high = highcut / nyq\n",
    "    b, a = butter(order, [low, high], btype='band')\n",
    "    return b, a\n",
    "    \n",
    "def bandpass_filter(data, lowcut, highcut, fs, order=5):\n",
    "    b, a = butter_bandpass(lowcut, highcut, fs, order=order)\n",
    "    zi = lfilter_zi(b, a)\n",
    "    y,zo = lfilter(b, a, data, zi=zi*data[0])\n",
    "    return y\n",
    "\n",
    "def generate_first_arrival_labels(p_indx, s_indx, sample_num):\n",
    "    \"\"\"\n",
    "    Generate labels for P-wave and S-wave first arrival picking.\n",
    "\n",
    "    Parameters:\n",
    "        p_indx (float): Arrival time of P-wave.\n",
    "        s_indx (float): Arrival time of S-wave.\n",
    "        sample_num (float): Total duration of the seismic signal.\n",
    "\n",
    "    Returns:\n",
    "        np.array: Array of labels indicating P-wave, S-wave, or neither.\n",
    "    \"\"\"\n",
    "    all_time = np.linspace(0, sample_num, sample_num)\n",
    "    #print(sample_num, all_time.shape)\n",
    "    p_wave_arrival_time = p_indx  # P-wave arrival time (in seconds)\n",
    "    s_wave_arrival_time = s_indx  # S-wave arrival time (in seconds)\n",
    "    \n",
    "    # Create Gaussian-shaped labels for P-wave and S-wave arrivals\n",
    "    sigma = 10 # Standard deviation of the Gaussian\n",
    "    p_wave_label = np.exp(-(all_time - p_wave_arrival_time)**2 / (2 * sigma**2))  # Gaussian label for P-wave\n",
    "    s_wave_label = np.exp(-(all_time - s_wave_arrival_time)**2 / (2 * sigma**2))  # Gaussian label for S-wave\n",
    "\n",
    "    return p_wave_label, s_wave_label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "bede1813",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-----------arrival time calculation begin-------------------\n",
      "-----------arrival time calculation end-------------------\n",
      "shape of arrivals: (10000,)\n",
      "-----------signal format convert begin-------------------\n",
      "-----------signal format convert finish-------------------\n",
      "(10000, 6000, 3)\n",
      "Processed_num: 0\t P_arrival_time: 1023\t S_arrival_time: 1563\t sample_num: 6000\n",
      "Processed_num: 5000\t P_arrival_time: 524\t S_arrival_time: 707\t sample_num: 6000\n"
     ]
    }
   ],
   "source": [
    "# obtain the P- and S-wave arrivals\n",
    "P_arrival_list = []\n",
    "S_arrival_list = []\n",
    "print('-----------arrival time calculation begin-------------------')\n",
    "for key in txed_signal_id:\n",
    "    if key in f_txed:   \n",
    "        dataset = f_txed.get(key)\n",
    "        P_arrival_list.append(int(dataset.attrs['p_arrival_sample']))\n",
    "        S_arrival_list.append(int(dataset.attrs['s_arrival_sample']))\n",
    "P_arrival_list = np.array(P_arrival_list)\n",
    "S_arrival_list = np.array(S_arrival_list)\n",
    "P_phase_label_indx = P_arrival_list\n",
    "S_phase_label_indx = S_arrival_list\n",
    "P_phase_label_indx = np.reshape(P_phase_label_indx, [P_phase_label_indx.shape[0], 1])\n",
    "S_phase_label_indx = np.reshape(S_phase_label_indx, [S_phase_label_indx.shape[0], 1])\n",
    "phase_label_indx = np.concatenate([P_phase_label_indx, S_phase_label_indx], axis=-1)\n",
    "print('-----------arrival time calculation end-------------------')\n",
    "print(f'shape of arrivals: {P_arrival_list.shape}')\n",
    "\n",
    "\n",
    "# apply the bandpass and normalize to each waveform\n",
    "signal_list = []\n",
    "print('-----------signal format convert begin-------------------')\n",
    "for key in txed_signal_id:\n",
    "    if key in f_txed:   \n",
    "        dataset = f_txed.get(key)\n",
    "        datas = dataset['data']\n",
    "        datas = np.array(datas)\n",
    "        datas_0 = bandpass_filter(datas[:,0], 1, 45, 100, order=3)\n",
    "        datas_1 = bandpass_filter(datas[:,1], 1, 45, 100, order=3)\n",
    "        datas_2 = bandpass_filter(datas[:,2], 1, 45, 100, order=3)\n",
    "        datas = np.vstack([datas_0, datas_1, datas_2])\n",
    "        signal_list.append(datas) \n",
    "signal_values = np.array(signal_list)\n",
    "bp_signal= np.transpose(signal_values, [0, 2, 1])\n",
    "\n",
    "\n",
    "#Normalized trace-by-trace\n",
    "max_values_per_event = np.max(bp_signal, axis=1)\n",
    "# Normalize each component of each event by dividing by its maximum value\n",
    "normalized_phase_data = bp_signal / max_values_per_event[:, np.newaxis, :]\n",
    "print('-----------signal format convert finish-------------------')\n",
    "print(bp_signal.shape)\n",
    "\n",
    "p_wave_label = []\n",
    "s_wave_label = []\n",
    "for i in range (normalized_phase_data.shape[0]):\n",
    "    # Example usage:\n",
    "    p_indx = P_arrival_list[i]  # Arrival time of P-wave (in seconds)\n",
    "    s_indx = S_arrival_list[i]  # Arrival time of S-wave (in seconds)    \n",
    "    sample_num = 6000  # Sampling rate of seismic signal (in Hz)\n",
    "    #print(f'Processed_num:{i}\\t P_arrival_time: {p_indx}\\t S_arrival_time: {s_indx}\\t sample_num: {sample_num}')\n",
    "    if i % 5000 == 0:\n",
    "        print(f'Processed_num: {i}\\t P_arrival_time: {p_indx}\\t S_arrival_time: {s_indx}\\t sample_num: {sample_num}')\n",
    "    \n",
    "    # Generate labels for P-wave and S-wave first arrival picking\n",
    "    p_labels, s_labels = generate_first_arrival_labels(p_indx, s_indx, sample_num)\n",
    "    p_wave_label.append(p_labels)  \n",
    "    s_wave_label.append(s_labels)\n",
    "p_wave_label = np.array(p_wave_label)\n",
    "s_wave_label = np.array(s_wave_label)\n",
    "phase_label = np.concatenate([np.reshape(p_wave_label, [p_wave_label.shape[0], p_wave_label.shape[1], 1]), \\\n",
    "                              np.reshape(s_wave_label, [s_wave_label.shape[0], s_wave_label.shape[1], 1])], axis=-1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "f5995de3-128e-4184-becc-18943d7961eb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train_phase shape (8500, 6000, 3) X_test_phase shape (1500, 6000, 3)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# Data splitting for EQ detection\n",
    "X_train_phase, X_test_phase, y_train_phase, y_test_phase = train_test_split(normalized_phase_data, phase_label, test_size=0.15, random_state=202404)\n",
    "print('X_train_phase shape', X_train_phase.shape, 'X_test_phase shape', X_test_phase.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fba59d14-84fc-4a7d-bd9e-961d3c700eee",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "3ead5213",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "32/32 [==============================] - 148s 4s/step - loss: 0.1064 - accuracy: 0.4628 - val_loss: 0.3593 - val_accuracy: 0.0425\n",
      "Epoch 2/20\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/eqp/lib/python3.11/site-packages/keras/src/engine/training.py:3079: UserWarning: You are saving your model as an HDF5 file via `model.save()`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')`.\n",
      "  saving_api.save_model(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "32/32 [==============================] - 501s 16s/step - loss: 0.0205 - accuracy: 0.6132 - val_loss: 0.2107 - val_accuracy: 0.0427\n",
      "Epoch 3/20\n",
      "32/32 [==============================] - 919s 29s/step - loss: 0.0188 - accuracy: 0.6839 - val_loss: 0.1341 - val_accuracy: 0.4502\n",
      "Epoch 4/20\n",
      "32/32 [==============================] - 118s 4s/step - loss: 0.0163 - accuracy: 0.6665 - val_loss: 0.0899 - val_accuracy: 0.7919\n",
      "Epoch 5/20\n",
      "32/32 [==============================] - 191s 6s/step - loss: 0.0147 - accuracy: 0.7056 - val_loss: 0.0655 - val_accuracy: 0.8257\n",
      "Epoch 6/20\n",
      "32/32 [==============================] - 121s 4s/step - loss: 0.0138 - accuracy: 0.7239 - val_loss: 0.0487 - val_accuracy: 0.8107\n",
      "Epoch 7/20\n",
      "32/32 [==============================] - 303s 10s/step - loss: 0.0133 - accuracy: 0.7340 - val_loss: 0.0378 - val_accuracy: 0.7951\n",
      "Epoch 8/20\n",
      "32/32 [==============================] - 608s 19s/step - loss: 0.0128 - accuracy: 0.7473 - val_loss: 0.0301 - val_accuracy: 0.7780\n",
      "Epoch 9/20\n",
      "32/32 [==============================] - 141s 4s/step - loss: 0.0126 - accuracy: 0.7580 - val_loss: 0.0261 - val_accuracy: 0.7769\n",
      "Epoch 10/20\n",
      "32/32 [==============================] - 587s 19s/step - loss: 0.0123 - accuracy: 0.7689 - val_loss: 0.0238 - val_accuracy: 0.7912\n",
      "Epoch 11/20\n",
      "32/32 [==============================] - 120s 4s/step - loss: 0.0121 - accuracy: 0.7813 - val_loss: 0.0223 - val_accuracy: 0.8016\n",
      "Epoch 12/20\n",
      "32/32 [==============================] - 1126s 36s/step - loss: 0.0119 - accuracy: 0.7877 - val_loss: 0.0217 - val_accuracy: 0.8177\n",
      "Epoch 13/20\n",
      "32/32 [==============================] - 120s 4s/step - loss: 0.0118 - accuracy: 0.7916 - val_loss: 0.0207 - val_accuracy: 0.8258\n",
      "Epoch 14/20\n",
      "32/32 [==============================] - 935s 30s/step - loss: 0.0116 - accuracy: 0.7935 - val_loss: 0.0198 - val_accuracy: 0.8177\n",
      "Epoch 15/20\n",
      "32/32 [==============================] - 118s 4s/step - loss: 0.0115 - accuracy: 0.7996 - val_loss: 0.0194 - val_accuracy: 0.8452\n",
      "Epoch 16/20\n",
      "32/32 [==============================] - 494s 16s/step - loss: 0.0113 - accuracy: 0.8035 - val_loss: 0.0184 - val_accuracy: 0.8603\n",
      "Epoch 17/20\n",
      "32/32 [==============================] - 451s 14s/step - loss: 0.0113 - accuracy: 0.8079 - val_loss: 0.0171 - val_accuracy: 0.8631\n",
      "Epoch 18/20\n",
      "32/32 [==============================] - 804s 26s/step - loss: 0.0112 - accuracy: 0.8102 - val_loss: 0.0157 - val_accuracy: 0.8522\n",
      "Epoch 19/20\n",
      "32/32 [==============================] - 431s 14s/step - loss: 0.0111 - accuracy: 0.8090 - val_loss: 0.0144 - val_accuracy: 0.8547\n",
      "Epoch 20/20\n",
      "32/32 [==============================] - 119s 4s/step - loss: 0.0110 - accuracy: 0.8109 - val_loss: 0.0135 - val_accuracy: 0.8450\n",
      "time cost od picking model: 8353.6608 sec\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "## Start the training process\n",
    "log_path = './training_model/logs/'\n",
    "save_model_path = './training_model/'\n",
    "fig_path = './figs/'\n",
    "\n",
    "os.makedirs(save_model_path, exist_ok=True)\n",
    "os.makedirs(log_path , exist_ok=True)\n",
    "os.makedirs(fig_path , exist_ok=True)\n",
    "\n",
    "val_splitting_rate = 0.05 # ratio of validation data\n",
    "learning_rate = 0.001 # Specify the learning rate\n",
    "phase_batch = 256 # batch size for picking\n",
    "phase_epoch = 20 #number of iterations\n",
    "\n",
    "optimizer = Adam(learning_rate=learning_rate)\n",
    "\n",
    "# train the picking model\n",
    "picking_model.compile(loss='binary_crossentropy', optimizer= optimizer, metrics=['accuracy'])\n",
    "\n",
    "csv_logger = CSVLogger(f'{log_path}EQ_picking_logs.csv')\n",
    "\n",
    "\n",
    "# Addig stopping condition\n",
    "es=EarlyStopping(monitor='loss', mode='min', verbose=1, patience=5)\n",
    "mc=ModelCheckpoint(f'{save_model_path}EQ_picking_best_model_new.h5', monitor='val_loss', mode='min', save_best_only=True)\n",
    "\n",
    "\n",
    "# Tic-Toc\n",
    "import time\n",
    "start=time.time()\n",
    "history_phase = picking_model.fit(X_train_phase, y_train_phase, verbose=1, epochs=phase_epoch,\n",
    "                                  validation_split=val_splitting_rate, shuffle=True, batch_size=phase_batch, callbacks=[csv_logger, es, mc])\n",
    "end = time.time()\n",
    "running_time = end - start\n",
    "print('time cost od picking model: %.4f sec' % running_time)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "4f3df984-ae9a-48c8-b615-d30e5489c146",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Plot training & validation accuracy values\n",
    "plt.figure(figsize=(8, 4))\n",
    "plt.subplot(121)\n",
    "plt.plot(history_phase.history['accuracy'], linewidth=2)\n",
    "plt.plot(history_phase.history['val_accuracy'], linewidth=2)\n",
    "plt.title('Model accuracy', fontsize=16)\n",
    "plt.xlabel('Epoch', fontsize=14)\n",
    "plt.ylabel('Accuracy', fontsize=14)\n",
    "plt.legend(['Train', 'Validation'], loc='lower right',fontsize='large', title_fontsize='large')\n",
    "# plt.show()\n",
    "\n",
    "# Plot training & validation loss values\n",
    "plt.subplot(122)\n",
    "plt.plot(history_phase.history['loss'], linewidth=2)\n",
    "plt.plot(history_phase.history['val_loss'], linewidth=2)\n",
    "plt.title('Model loss', fontsize=16)\n",
    "plt.xlabel('Epoch', fontsize=14)\n",
    "plt.ylabel('Accuracy', fontsize=14)\n",
    "plt.legend(['Train', 'Validation'], loc='upper right',fontsize='large', title_fontsize='large')\n",
    "plt.tight_layout()\n",
    "plt.savefig(f'{fig_path}Acc_and_loss_curves_detection.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "3aae6808-3961-4776-a62c-9b4ac6a4f4fb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "47/47 [==============================] - 7s 146ms/step\n"
     ]
    }
   ],
   "source": [
    "from keras.models import load_model\n",
    "\n",
    "# picking_model = load_model(f'{save_model_path}EQ_picking_best_model_new.h5')\n",
    "test_phase_out = picking_model.predict(X_test_phase)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "75bcc62d-2ef9-400a-8c9d-0679b7de0fb4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MAE and $sigma$ of p: 2.9011s\t5.7200s\n",
      "MAE and $sigma$ of s: 1.4900s\t5.2832s\n"
     ]
    }
   ],
   "source": [
    "def to_numpy(array):\n",
    "    array = np.array(array)\n",
    "    return array\n",
    "\n",
    "\n",
    "pre_indx_p = []\n",
    "pre_indx_s = []\n",
    "lab_indx_p = []\n",
    "lab_indx_s = []\n",
    "for i in range (test_phase_out.shape[0]):\n",
    "    # print(y_test_phase[i, :, :0].shape)\n",
    "    pre_ind_p = np.argmax(test_phase_out[i, :, 0])\n",
    "    pre_ind_s = np.argmax(test_phase_out[i, :, 1])\n",
    "    lab_ind_p = np.argmax(y_test_phase[i, :, 0])\n",
    "    lab_ind_s = np.argmax(y_test_phase[i, :, 1])\n",
    "    pre_indx_p.append(pre_ind_p)\n",
    "    pre_indx_s.append(pre_ind_s)\n",
    "    lab_indx_p.append(lab_ind_p)\n",
    "    lab_indx_s.append(lab_ind_s) \n",
    "    \n",
    "pre_indx_p = to_numpy(pre_indx_p)\n",
    "pre_indx_s = to_numpy(pre_indx_s)\n",
    "lab_indx_p = to_numpy(lab_indx_p)\n",
    "lab_indx_s = to_numpy(lab_indx_s)\n",
    "\n",
    "err_indx_p = lab_indx_p - pre_indx_p\n",
    "err_indx_p = err_indx_p/100\n",
    "err_indx_s = lab_indx_s - pre_indx_s\n",
    "err_indx_s = err_indx_s/100\n",
    "\n",
    "# xx = 2\n",
    "# err_indx_p[(err_indx_p > xx) | (err_indx_p < -xx)] = 0\n",
    "# err_indx_s[(err_indx_s > xx) | (err_indx_s < -xx)] = 0\n",
    "\n",
    "mae_p = np.mean(np.abs(err_indx_p - np.mean(err_indx_p)))\n",
    "sigma_p = np.std(err_indx_p)\n",
    "mae_s = np.mean(np.abs(err_indx_s - np.mean(err_indx_s)))\n",
    "sigma_s = np.std(err_indx_s)\n",
    "\n",
    "print(f'MAE and $sigma$ of p: {mae_p:.4f}s\\t{sigma_p:.4f}s\\nMAE and $sigma$ of s: {mae_s:.4f}s\\t{sigma_s:.4f}s')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d7466e53-b55a-4b6e-a6b4-2ea1af5b3422",
   "metadata": {},
   "outputs": [],
   "source": [
    "err_indx_s/100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c0538081-22f8-4e17-ba67-070eceaf8abd",
   "metadata": {},
   "outputs": [],
   "source": [
    "np.mean(np.abs(err_indx_s - np.mean(err_indx_s)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "d41e5af6-8fab-402c-a8f1-e323ffa9de6d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 3))  # Set figure size\n",
    "ax = plt.subplot(121)\n",
    "plt.hist(err_indx_p, bins=30, color='palegoldenrod', edgecolor='black', alpha=0.7)  # Adjust bins, color, edgecolor, and transparency\n",
    "plt.xlabel('Picking error (s)', fontsize=12, fontweight='bold')  # Add x-axis label with custom font size\n",
    "plt.ylabel('Count', fontsize=12, fontweight='bold')  # Add y-axis label with custom font size\n",
    "plt.xticks(fontsize=10, fontweight='bold')  # Set font size for x-axis ticks\n",
    "plt.yticks(fontsize=10, fontweight='bold')  # Set font size for y-axis ticks\n",
    "plt.text(0.2, 0.8, f'MAE={mae_p:.2f} s\\n$\\sigma$={sigma_p:.2f} s', fontsize=10, fontweight='bold', ha='center', va='center', transform=ax.transAxes)\n",
    "plt.grid(axis='y', linestyle='--', alpha=0.5)  # Add grid lines with custom style and transparency\n",
    "plt.legend(['Pwave'], prop={'weight': 'bold'})\n",
    "\n",
    "ax = plt.subplot(122)\n",
    "plt.hist(err_indx_s, bins=30, color='skyblue', edgecolor='black', alpha=0.7)  # Adjust bins, color, edgecolor, and transparency\n",
    "plt.xlabel('Picking error (s)', fontsize=12, fontweight='bold')  # Add x-axis label with custom font size\n",
    "plt.ylabel('Count', fontsize=12, fontweight='bold')  # Add y-axis label with custom font size\n",
    "plt.xticks(fontsize=10, fontweight='bold')  # Set font size for x-axis ticks\n",
    "plt.yticks(fontsize=10, fontweight='bold')  # Set font size for y-axis ticks\n",
    "plt.text(0.2, 0.8, f'MAE={mae_s:.2f} s\\n$\\sigma$={sigma_s:.2f} s', fontsize=10, fontweight='bold', ha='center', va='center', transform=ax.transAxes)\n",
    "plt.grid(axis='y', linestyle='--', alpha=0.5)  # Add grid lines with custom style and transparency\n",
    "plt.legend(['Swave'], prop={'weight': 'bold'})\n",
    "\n",
    "plt.tight_layout()\n",
    "\n",
    "plt.savefig(f'{fig_path}Arrival_picking_testing_data.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "857cd431-3371-4967-ac25-d4e0ccf22ff6",
   "metadata": {},
   "outputs": [],
   "source": [
    "## Define fancy visualization functions\n",
    "\n",
    "def plot_labels_signals(signal, p_predict, s_predict, p_labels, s_labels, sample_num):\n",
    "    data = signal\n",
    "    p_predict_indx = np.argmax(p_predict)\n",
    "    s_predict_indx = np.argmax(s_predict)\n",
    "    s_labels_indx = np.argmax(s_labels)\n",
    "    p_labels_indx = np.argmax(p_labels)\n",
    "    \n",
    "    time = np.linspace(0, sample_num, sample_num)\n",
    "    fig = plt.figure(figsize=(6, 6))\n",
    "    ax1 = fig.add_subplot(511)\n",
    "    plt.plot(time, data[:,0], 'k',label='Z')\n",
    "    ymin,ymax = ax1.get_ylim()\n",
    "    # plt.vlines(spt,ymin,yma,color='r',linewidth=2)\n",
    "    # plt.vlines(sst,ymin,yma,color='b',linewidth=2)\n",
    "    legend_properties = {'weight':'bold'}\n",
    "    # ymin, ymax = ax1.get_ylim()\n",
    "    print('ymin, ymax:', ymin, ymax)\n",
    "    plt.vlines(p_labels_indx,ymin+0.2,ymax-0.2,color='deepskyblue',linewidth=2)\n",
    "    plt.vlines(s_labels_indx,ymin+0.2,ymax-0.2,color='tomato',linewidth=2)\n",
    "    plt.vlines(p_predict_indx,ymin,ymax,color='tomato',linewidth=2, linestyle='--')\n",
    "    plt.vlines(s_predict_indx,ymin,ymax,color='deepskyblue',linewidth=2, linestyle='--')\n",
    "    plt.legend(loc = 'upper right', borderaxespad=0., prop=legend_properties)\n",
    "    plt.ylabel('Amplitude', fontsize=12)\n",
    "    plt.xlim([0, 6000])\n",
    "    ax1.set_xticklabels([])\n",
    "    # ax1.set_xticks([])\n",
    "    \n",
    "    ax = fig.add_subplot(512) \n",
    "    plt.plot(time, data[:,1], 'k',label='N')\n",
    "    # ymin,ymax = ax1.get_ylim()\n",
    "    plt.vlines(p_labels_indx,ymin+0.2,ymax-0.2,color='deepskyblue',linewidth=2)\n",
    "    plt.vlines(s_labels_indx,ymin+0.2,ymax-0.2,color='tomato',linewidth=2)\n",
    "    plt.vlines(p_predict_indx,ymin,ymax,color='tomato',linewidth=2, linestyle='--')\n",
    "    plt.vlines(s_predict_indx,ymin,ymax,color='deepskyblue',linewidth=2, linestyle='--')\n",
    "    legend_properties = {'weight':'bold'}\n",
    "    plt.legend(loc = 'upper right', borderaxespad=0., prop=legend_properties)\n",
    "    plt.ylabel('Amplitude', fontsize=12) \n",
    "    plt.xlim([0, 6000])\n",
    "    ax.set_xticklabels([])\n",
    "    \n",
    "    ax = fig.add_subplot(513) \n",
    "    plt.plot(time, data[:,2], 'k',label='E')\n",
    "    # ymin,yma = ax1.get_ylim()\n",
    "    # plt.vlines(spt,ymin,yma,color='r',linewidth=2)\n",
    "    # plt.vlines(sst,ymin,yma,color='b',linewidth=2)\n",
    "    legend_properties = {'weight':'bold'}\n",
    "    # ymin, ymax = ax.get_ylim()\n",
    "    plt.vlines(p_labels_indx,ymin+0.2,ymax-0.2,color='deepskyblue',linewidth=2)\n",
    "    plt.vlines(s_labels_indx,ymin+0.2,ymax-0.2,color='tomato',linewidth=2)\n",
    "    plt.vlines(p_predict_indx,ymin,ymax,color='tomato',linewidth=2, linestyle='--')\n",
    "    plt.vlines(s_predict_indx,ymin,ymax,color='deepskyblue',linewidth=2, linestyle='--')\n",
    "    plt.legend(loc = 'upper right', borderaxespad=0., prop=legend_properties)\n",
    "    plt.ylabel('Amplitude', fontsize=12) \n",
    "    plt.xlim([0, 6000])\n",
    "    # plt.ylim([-1, 1])\n",
    "    ymin,yma = ax.get_ylim()\n",
    "    ax.set_xticklabels([])\n",
    "    # ax.set_xticks([])\n",
    "    \n",
    "    ax = fig.add_subplot(514) \n",
    "    plt.plot(time, p_predict, label='P-wave predicted', linestyle='--')\n",
    "    plt.plot(time, s_predict, label='S-wave predicted', linestyle='--')\n",
    "    plt.legend(loc = 'upper right', borderaxespad=0., prop=legend_properties)\n",
    "    plt.ylabel('Probability', fontsize=12) \n",
    "    # plt.xlabel('Sample', fontsize=12) \n",
    "    plt.xlim([0, 6000])\n",
    "    # plt.ylim([-1, 1])\n",
    "    ax.set_xticklabels([])\n",
    "    \n",
    "    ax = fig.add_subplot(515) \n",
    "    plt.plot(time, p_labels, label='P-wave label', linestyle='--')\n",
    "    plt.plot(time, s_labels, label='S-wave label', linestyle='--')\n",
    "    plt.legend(loc = 'upper right', borderaxespad=0., prop=legend_properties)\n",
    "    plt.ylabel('Probability', fontsize=12) \n",
    "    plt.xlabel('Sample', fontsize=12) \n",
    "    plt.xlim([0, 6000])\n",
    "    plt.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "3b670cae-7554-475e-a25c-120b8adfb6a5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1500, 6000, 2)\n",
      "ymin, ymax: -1.4062203047810837 1.1145819192752897\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x600 with 5 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(test_phase_out.shape)\n",
    "\n",
    "m = 10\n",
    "\n",
    "# def plot_labels_signals(signal, p_predict, s_predict, p_labels, s_labels, sample_num):\n",
    "plot_labels_signals(X_test_phase[m, :, :], p_predict=test_phase_out[m, :, 0], s_predict=test_phase_out[m, :, 1], p_labels=y_test_phase[m, :, 0], s_labels=y_test_phase[m, :, 1], sample_num=6000)\n",
    "# plt.savefig(f'./figs/phase_picking/p_and_s_wave/p_and_swave_picking_testing_num_{m}.pdf', bbox_inches=\"tight\", dpi=300)\n",
    "# plt.savefig(f'./figs/phase_picking/p_and_s_wave/p_and_swave_picking_testing_num_{m}.png', bbox_inches=\"tight\", dpi=300)\n",
    "plt.show()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "38a6d2c0-e0bf-42ac-98bb-83acb7320ce7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "266/266 [==============================] - 39s 145ms/step\n",
      "(8500, 6000, 2)\n"
     ]
    }
   ],
   "source": [
    "train_phase_out = picking_model.predict(X_train_phase)\n",
    "print(train_phase_out.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "2a1fb77d-1f28-478c-9169-7a8158df7fcc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MAE and $sigma$ of p: 2.7359s\t4.8377s\n",
      "MAE and $sigma$ of s: 1.4759s\t5.1832s\n"
     ]
    }
   ],
   "source": [
    "pre_indx_p = []\n",
    "pre_indx_s = []\n",
    "lab_indx_p = []\n",
    "lab_indx_s = []\n",
    "for i in range (test_phase_out.shape[0]):\n",
    "    # print(y_test_phase[i, :, :0].shape)\n",
    "    pre_ind_p = np.argmax(train_phase_out[i, :, 0])\n",
    "    pre_ind_s = np.argmax(train_phase_out[i, :, 1])\n",
    "    lab_ind_p = np.argmax(y_train_phase[i, :, 0])\n",
    "    lab_ind_s = np.argmax(y_train_phase[i, :, 1])\n",
    "    pre_indx_p.append(pre_ind_p)\n",
    "    pre_indx_s.append(pre_ind_s)\n",
    "    lab_indx_p.append(lab_ind_p)\n",
    "    lab_indx_s.append(lab_ind_s) \n",
    "    \n",
    "pre_indx_p = to_numpy(pre_indx_p)\n",
    "pre_indx_s = to_numpy(pre_indx_s)\n",
    "lab_indx_p = to_numpy(lab_indx_p)\n",
    "lab_indx_s = to_numpy(lab_indx_s)\n",
    "\n",
    "pre_indx_p = to_numpy(pre_indx_p)\n",
    "pre_indx_s = to_numpy(pre_indx_s)\n",
    "lab_indx_p = to_numpy(lab_indx_p)\n",
    "lab_indx_s = to_numpy(lab_indx_s)\n",
    "\n",
    "err_indx_p = lab_indx_p - pre_indx_p\n",
    "err_indx_p_train = err_indx_p/100\n",
    "err_indx_s = lab_indx_s - pre_indx_s\n",
    "err_indx_s_train = err_indx_s/100\n",
    "\n",
    "\n",
    "# xx = 2\n",
    "# err_indx_p_train[(err_indx_p_train > xx) | (err_indx_p_train < -xx)] = 0\n",
    "# err_indx_s_train[(err_indx_s_train > xx) | (err_indx_s_train < -xx)] = 0\n",
    "mae_p = np.mean(np.abs(err_indx_p_train - np.mean(err_indx_p_train)))\n",
    "sigma_p = np.std(err_indx_p_train)\n",
    "mae_s = np.mean(np.abs(err_indx_s_train - np.mean(err_indx_s_train)))\n",
    "sigma_s = np.std(err_indx_s_train)\n",
    "\n",
    "\n",
    "print(f'MAE and $sigma$ of p: {mae_p:.4f}s\\t{sigma_p:.4f}s\\nMAE and $sigma$ of s: {mae_s:.4f}s\\t{sigma_s:.4f}s')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "a4027cb5-5fe1-4894-9e97-52973f9ad9f8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 3))  # Set figure size\n",
    "ax = plt.subplot(121)\n",
    "plt.hist(err_indx_p_train, bins=30, color='palegoldenrod', edgecolor='black', alpha=0.7)  # Adjust bins, color, edgecolor, and transparency\n",
    "plt.xlabel('Picking error (s)', fontsize=12, fontweight='bold')  # Add x-axis label with custom font size\n",
    "plt.ylabel('Count', fontsize=12, fontweight='bold')  # Add y-axis label with custom font size\n",
    "plt.xticks(fontsize=10, fontweight='bold')  # Set font size for x-axis ticks\n",
    "plt.yticks(fontsize=10, fontweight='bold')  # Set font size for y-axis ticks\n",
    "plt.text(0.2, 0.8, f'MAE={mae_p:.2f} s\\n$\\sigma$={sigma_p:.2f} s', fontsize=10, fontweight='bold', ha='center', va='center', transform=ax.transAxes)\n",
    "plt.grid(axis='y', linestyle='--', alpha=0.5)  # Add grid lines with custom style and transparency\n",
    "plt.legend(['Pwave'], prop={'weight': 'bold'})\n",
    "\n",
    "ax = plt.subplot(122)\n",
    "plt.hist(err_indx_s_train, bins=30, color='skyblue', edgecolor='black', alpha=0.7)  # Adjust bins, color, edgecolor, and transparency\n",
    "plt.xlabel('Picking error (s)', fontsize=12, fontweight='bold')  # Add x-axis label with custom font size\n",
    "plt.ylabel('Count', fontsize=12, fontweight='bold')  # Add y-axis label with custom font size\n",
    "plt.xticks(fontsize=10, fontweight='bold')  # Set font size for x-axis ticks\n",
    "plt.yticks(fontsize=10, fontweight='bold')  # Set font size for y-axis ticks\n",
    "plt.text(0.2, 0.8, f'MAE={mae_s:.2f} s\\n$\\sigma$={sigma_s:.2f} s', fontsize=10, fontweight='bold', ha='center', va='center', transform=ax.transAxes)\n",
    "plt.grid(axis='y', linestyle='--', alpha=0.5)  # Add grid lines with custom style and transparency\n",
    "plt.legend(['Swave'], prop={'weight': 'bold'})\n",
    "\n",
    "plt.tight_layout()\n",
    "\n",
    "plt.savefig(f'{fig_path}Arrival_picking_training_data.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "a2b2a06d-aadf-4279-b649-751f6e3c0387",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ymin, ymax: -1.1730294162789618 1.1034775912513792\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x600 with 5 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "m = 600\n",
    "\n",
    "plot_labels_signals(X_train_phase[m, :, :], p_predict=train_phase_out[m, :, 0], s_predict=train_phase_out[m, :, 1], p_labels=y_train_phase[m, :, 0], s_labels=y_train_phase[m, :, 1], sample_num=6000)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "3519d7fa-f563-419a-9268-ff9ae76562a6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x304686a50>]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure()\n",
    "plt.plot(test_phase_out[20, :, 0])\n",
    "\n",
    "plt.figure()\n",
    "plt.plot(y_test_phase[20, :, 0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "551c42d2-4b08-4c41-80c8-d29c1a9d801d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x305489a90>]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure()\n",
    "plt.plot(test_phase_out[20, :, 1])\n",
    "\n",
    "plt.figure()\n",
    "plt.plot(y_test_phase[20, :, 1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2ca995fb-f793-49dd-8563-e5635d9587d3",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}