aifnet_five_folds.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "293c9b22",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/home/sebastian/experiments/aifnet_replication\r\n"
     ]
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "import zipfile\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras import layers\n",
    "from tensorflow.keras.callbacks import ModelCheckpoint, TensorBoard\n",
    "import nibabel as nib\n",
    "from scipy import ndimage\n",
    "import glob\n",
    "import sys,os\n",
    "from natsort import natsorted\n",
    "import tensorflow_probability as tfp\n",
    "import random\n",
    "from aifnet_utils.preprocess import read_nifti_file, normalize, normalize_aif, process_scan, normalize_zero_one\n",
    "from aifnet_utils.losses import MaxCorrelation\n",
    "from aifnet_utils.data_loaders import read_isles_volumepaths_from_file_otf, read_isles_annotations_from_file, ISLES18DataGen_aifvof_otf\n",
    "from aifnet_utils.data_loaders import delay_sequence_padding, anticipate_sequence_padding, late_bolus, early_bolus\n",
    "from aifnet_utils.results import plot_predictions\n",
    "%matplotlib inline\n",
    "\n",
    "!pwd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "cbf7e6af",
   "metadata": {},
   "outputs": [],
   "source": [
    "keras.backend.set_image_data_format('channels_last')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "f64554da",
   "metadata": {},
   "outputs": [],
   "source": [
    "ROOT_EXP = '/home/sebastian/experiments/aifnet_replication/'\n",
    "root_dir     = '/media/sebastian/data/ASAP/ISLES2018_Training'\n",
    "#At insel: /media/sebastian/data/ASAP/ISLES2018_Training\n",
    "#Local: '/Users/sebastianotalora/work/postdoc/data/ISLES/'\n",
    "aif_annotations_path = ROOT_EXP + 'annotated_aif_vof_complete_revised.csv'\n",
    "min_num_volumes_ctp = 43\n",
    "#ROOT_EXP = '/home/sebastian/experiments/aifnet_replication'\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "c1faa9f3",
   "metadata": {},
   "outputs": [],
   "source": [
    "current_fold = 1\n",
    "#Reading AIFs and VOFs for each of the partitions\n",
    "\n",
    "train_partition_path = ROOT_EXP + '/partitions/fold_'+str(current_fold) +'/train.txt'\n",
    "valid_partition_path = ROOT_EXP + '/partitions/fold_'+str(current_fold) +'/valid.txt'\n",
    "test_partition_path = ROOT_EXP + '/partitions/fold_'+str(current_fold) +'/test.txt'\n",
    "\n",
    "aif_annotations_train, vof_annotations_train = read_isles_annotations_from_file(aif_annotations_path, train_partition_path, \n",
    "                                                root_dir, min_num_volumes_ctp, return_aif_only = False)\n",
    "aif_annotations_valid, vof_annotations_valid = read_isles_annotations_from_file(aif_annotations_path, valid_partition_path, root_dir, \n",
    "                                         min_num_volumes_ctp, return_aif_only = False)\n",
    "aif_annotations_test, vof_annotations_test = read_isles_annotations_from_file(aif_annotations_path,  test_partition_path,\n",
    "                                        root_dir, min_num_volumes_ctp, return_aif_only = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "7481194c",
   "metadata": {},
   "outputs": [],
   "source": [
    "ctp_volumes_train = read_isles_volumepaths_from_file_otf(root_dir, train_partition_path, aif_annotations_path)\n",
    "ctp_volumes_valid = read_isles_volumepaths_from_file_otf(root_dir, valid_partition_path, aif_annotations_path)\n",
    "ctp_volumes_test = read_isles_volumepaths_from_file_otf(root_dir, test_partition_path, aif_annotations_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "7510fc0f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "64 64\n",
      "9 9\n",
      "21 21\n"
     ]
    }
   ],
   "source": [
    "print(len(ctp_volumes_train), len(aif_annotations_train))\n",
    "print(len(ctp_volumes_valid), len(aif_annotations_valid))\n",
    "print(len(ctp_volumes_test), len(aif_annotations_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "d21a88eb",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_model_twoPvols(width=256, height=256, num_channels=43):\n",
    "    \"\"\"Build a 3D convolutional neural network model.\"\"\"\n",
    "    #width and height of the PCT is 256, the number of slices is variable, and the number of channels are\n",
    "    #the number of timepoints in the PCT sequence        \n",
    "    inputs = keras.Input((width, height, None , 43))\n",
    "\n",
    "    x = layers.Conv3D(filters=16, kernel_size=(3,3,1), activation=\"relu\", data_format='channels_last', padding='same')(inputs)\n",
    "    #x = layers.Dropout(0.3)(x)\n",
    "    \n",
    "    x = layers.Conv3D(filters=32, kernel_size=(3,3,3), activation=\"relu\", data_format='channels_last', padding='same')(x)\n",
    "    #x = layers.Dropout(0.3)(x)\n",
    "    \n",
    "    x = layers.Conv3D(filters=64, kernel_size=(3,3,3), activation=\"relu\", data_format='channels_last', padding='same')(x)\n",
    "    #x = layers.Dropout(0.3)(x)\n",
    "    \n",
    "    x = layers.Conv3D(filters=128, kernel_size=(3,3,3), activation=\"relu\", data_format='channels_last', padding='same')(x)\n",
    "\n",
    "    #x = layers.Conv3D(filters=256, kernel_size=(3,3,3), activation=\"relu\", data_format='channels_last', padding='same')(x)\n",
    "    #x = layers.Dropout(0.3)(x)\n",
    "    Lout = layers.Conv3D(filters=1, kernel_size=(3,3,3), activation=\"relu\", data_format='channels_last', padding='same')(x)\n",
    "    Lout2 = layers.Conv3D(filters=1, kernel_size=(3,3,3), activation=\"relu\", data_format='channels_last', padding='same')(x)\n",
    "\n",
    "    P_vol_aif = tf.keras.activations.sigmoid(Lout,name=\"aif\")\n",
    "    P_vol_vof = tf.keras.activations.sigmoid(Lout2,name=\"vof\")\n",
    "    #Voxelwise multiplication of P_vol and each of the CTP time points    \n",
    "    voxelwise_mult_each_ctp_aif = tf.keras.layers.Multiply()([inputs,P_vol_aif])    \n",
    "    voxelwise_mult_each_ctp_vof = tf.keras.layers.Multiply()([inputs,P_vol_vof])        \n",
    "    #The 3D average pooling block averages the volumetric information along the x-y-z axes, \n",
    "    #such that the predicted vascular function y(t) is a 1D vector of length T.\n",
    "    outputs_aif = layers.GlobalAveragePooling3D(data_format='channels_last', name='aif_loss')(voxelwise_mult_each_ctp_aif)\n",
    "    outputs_vof = layers.GlobalAveragePooling3D(data_format='channels_last', name='vof_loss')(voxelwise_mult_each_ctp_vof)    \n",
    "    #outputs_aif = tf.keras.activations.linear\n",
    "\n",
    "    # Define the model.\n",
    "    model = keras.Model(inputs, [outputs_aif,outputs_vof], name=\"aifnet\")\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "af312141",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"aifnet\"\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "input_1 (InputLayer)            [(None, 256, 256, No 0                                            \n",
      "__________________________________________________________________________________________________\n",
      "conv3d (Conv3D)                 (None, 256, 256, Non 6208        input_1[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "conv3d_1 (Conv3D)               (None, 256, 256, Non 13856       conv3d[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv3d_2 (Conv3D)               (None, 256, 256, Non 55360       conv3d_1[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "conv3d_3 (Conv3D)               (None, 256, 256, Non 221312      conv3d_2[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "conv3d_4 (Conv3D)               (None, 256, 256, Non 3457        conv3d_3[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "conv3d_5 (Conv3D)               (None, 256, 256, Non 3457        conv3d_3[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "tf.keras.activations.sigmoid (T (None, 256, 256, Non 0           conv3d_4[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "tf.keras.activations.sigmoid_1  (None, 256, 256, Non 0           conv3d_5[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "multiply (Multiply)             (None, 256, 256, Non 0           input_1[0][0]                    \n",
      "                                                                 tf.keras.activations.sigmoid[0][0\n",
      "__________________________________________________________________________________________________\n",
      "multiply_1 (Multiply)           (None, 256, 256, Non 0           input_1[0][0]                    \n",
      "                                                                 tf.keras.activations.sigmoid_1[0]\n",
      "__________________________________________________________________________________________________\n",
      "aif_loss (GlobalAveragePooling3 (None, 43)           0           multiply[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "vof_loss (GlobalAveragePooling3 (None, 43)           0           multiply_1[0][0]                 \n",
      "==================================================================================================\n",
      "Total params: 303,650\n",
      "Trainable params: 303,650\n",
      "Non-trainable params: 0\n",
      "__________________________________________________________________________________________________\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Build model.\n",
    "model = get_model_twoPvols(width=256, height=256, num_channels=43)\n",
    "model.summary()\n",
    "tf.keras.utils.plot_model(\n",
    "    model,\n",
    "    to_file=\"aifnet.png\",\n",
    "    show_shapes=False,\n",
    "    show_dtype=False,\n",
    "    show_layer_names=True,\n",
    "    rankdir=\"TB\",\n",
    "    expand_nested=False,\n",
    "    dpi=180)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "e101691d",
   "metadata": {},
   "outputs": [],
   "source": [
    "initial_learning_rate = 0.001\n",
    "lr_schedule = keras.optimizers.schedules.ExponentialDecay(\n",
    "    initial_learning_rate, decay_steps=100, decay_rate=0.96, staircase=True\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "1c9e8033",
   "metadata": {},
   "outputs": [],
   "source": [
    "optimizer_aifnet = optimizer = keras.optimizers.SGD(learning_rate=lr_schedule) #keras.optimizers.Adam(learning_rate=initial_learning_rate)\n",
    "model.compile(\n",
    "    loss=[MaxCorrelation,MaxCorrelation],\n",
    "    optimizer=optimizer_aifnet,\n",
    "    metrics=['mae'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "c1e24f4e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define callbacks.\n",
    "early_stopping_cb = keras.callbacks.EarlyStopping(monitor=\"val_aif_loss_mae\", patience=3)\n",
    "path_checkpointer_model = ROOT_EXP +'/results/' \n",
    "path_checkpointer_model += 'aifnet_2Pvols_augment_lr' + str(initial_learning_rate) + '_fold_' + str(current_fold) +'.hdf5'\n",
    "path_tensorboard_log    = ROOT_EXP + '/results/logsTensorBoard/'\n",
    "path_tensorboard_log    += 'aifnet_2Pvols_augment_lr' + str(initial_learning_rate) + '_fold_' + str(current_fold) \n",
    "\n",
    "checkpointer = ModelCheckpoint(filepath=path_checkpointer_model, monitor='val_aif_loss_mae', \n",
    "                               verbose=1, save_best_only=True)\n",
    "tb_callback = TensorBoard(log_dir=path_tensorboard_log, histogram_freq=0, write_graph=True, write_grads=False,\n",
    "                          write_images=False, embeddings_freq=0, embeddings_layer_names=None,\n",
    "                          embeddings_metadata=None)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "2dbab647",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_datagen = ISLES18DataGen_aifvof_otf(ctp_volumes=ctp_volumes_train, annotations_aif=aif_annotations_train,\n",
    "                               annotations_vof=vof_annotations_train,minimum_number_volumes_ctp = 43, batch_size=1,\n",
    "                                         time_arrival_augmentation = True)\n",
    "validation_datagen =  ISLES18DataGen_aifvof_otf(ctp_volumes=ctp_volumes_valid, annotations_aif=aif_annotations_valid,\n",
    "                               annotations_vof=vof_annotations_valid,minimum_number_volumes_ctp = 43, batch_size=1,\n",
    "                                         time_arrival_augmentation = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "7ca80e2b",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/sebastian/anaconda3/envs/aifnet/lib/python3.9/site-packages/tensorflow/python/keras/utils/generic_utils.py:494: CustomMaskWarning: Custom mask layers require a config and must override get_config. When loading, the custom mask layer must be passed to the custom_objects argument.\n",
      "  warnings.warn('Custom mask layers require a config and must override '\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "64/64 [==============================] - 82s 1s/step - loss: -1.6168 - aif_loss_loss: -0.8083 - vof_loss_loss: -0.8085 - aif_loss_mae: 0.3527 - vof_loss_mae: 0.3529 - val_loss: -1.6527 - val_aif_loss_loss: -0.8258 - val_vof_loss_loss: -0.8268 - val_aif_loss_mae: 0.3404 - val_vof_loss_mae: 0.3406\n",
      "\n",
      "Epoch 00001: val_aif_loss_mae improved from inf to 0.34044, saving model to /home/sebastian/experiments/aifnet_replication//results/aifnet_2Pvols_augment_lr0.001_fold_1.hdf5\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x7f821447f1c0>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nb_epochs=1\n",
    "model.fit(train_datagen,batch_size=1,callbacks=[checkpointer,tb_callback,early_stopping_cb],\n",
    "          epochs=nb_epochs, validation_data=validation_datagen)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "0fda3926",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "def plot_predictions(model,x,y,prefix_fig, normalize_preds=True, type_pred='AIF', savefig=True):\n",
    "    pred = model.predict(np.expand_dims(normalize(x), axis=0))    \n",
    "    if type_pred=='AIF':\n",
    "        pred = pred[0][0]\n",
    "        plt.title('AIF Function Predictions')        \n",
    "    else:\n",
    "        pred = pred[1][0]\n",
    "        plt.title('VOF Function Predictions')\n",
    "\n",
    "    if normalize_preds:\n",
    "        #print('Normalizing')\n",
    "        pred = normalize_zero_one(pred)\n",
    "        y = normalize_zero_one(y)\n",
    "    plt.plot(y)\n",
    "    plt.plot(pred)    \n",
    "    plt.legend(['y', 'prediction'])\n",
    "    plt.xlabel('Time (s)')\n",
    "    plt.ylabel('Normalized Density (HU)')\n",
    "    if savefig:\n",
    "        plt.savefig(prefix_fig +str(case_number) + '.png')\n",
    "        plt.figure().clear()\n",
    "        plt.close()\n",
    "        plt.cla()\n",
    "        plt.clf()\n",
    "        return y,pred\n",
    "    plt.show()\n",
    "    return"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "771945eb",
   "metadata": {},
   "outputs": [],
   "source": [
    "random.uniform?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "89df0e73",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[3.153851991828017e-06,\n",
       " 8.919617791760681e-05,\n",
       " 0.00014163336313737364,\n",
       " 0.0067085000154349615,\n",
       " 0.05218042230267252,\n",
       " 0.4218960369730479]"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lrs = [0.000001, 0.00001, 0.0001, 0.001, 0.01, 0.1]\n",
    "random_lrs = [random.uniform(0,lr)*10 for lr in lrs]\n",
    "random_lrs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "90b91b51",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "type_predictions = 'AIF'\n",
    "results_meassures = []\n",
    "prefix_fig = ROOT_EXP+'/results/predictions_aif/'+path_tensorboard_log.split('/')[-1]+'_'\n",
    "for case_number in range(len(ctp_volumes_test)):\n",
    "    case_id = ctp_volumes_test[case_number]['image'].split('.')[-2]\n",
    "    cur_nib = nib.load(ctp_volumes_test[case_number]['image'])\n",
    "    ctp_vals = cur_nib.get_fdata()\n",
    "    x = normalize(ctp_vals[:,:,:,0:min_num_volumes_ctp])\n",
    "    #x =  ctp_volumes_test[case_number]['ctpvals']\n",
    "    if type_predictions == 'AIF':\n",
    "        y = aif_annotations_test[case_id]\n",
    "    if type_predictions == 'VOF':\n",
    "        y = vof_annotations_test[case_id]\n",
    "    results_meassures.append(plot_predictions(model,x,y, prefix_fig, True, type_predictions,True))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "ef85c8ff",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(array([0.09601971, 0.        , 0.09969336, 0.06033708, 0.1602994 ,\n",
       "         0.15101941, 0.23050125, 0.22050815, 0.38713572, 0.55151444,\n",
       "         0.64912307, 0.7779629 , 0.88893823, 0.92315329, 0.93129155,\n",
       "         0.98498206, 0.89316954, 1.        , 0.75896258, 0.64192067,\n",
       "         0.64687791, 0.46862187, 0.4314192 , 0.2159226 , 0.34338038,\n",
       "         0.21488483, 0.2123785 , 0.4175338 , 0.24493687, 0.26847882,\n",
       "         0.36723223, 0.34193261, 0.30349051, 0.37313306, 0.30462053,\n",
       "         0.44617088, 0.31554538, 0.28896311, 0.37174539, 0.31573989,\n",
       "         0.31100047, 0.21206836, 0.34418673]),\n",
       "  array([0.        , 0.0038326 , 0.00646599, 0.01450629, 0.00260341,\n",
       "         0.0208921 , 0.0532353 , 0.1402633 , 0.23332919, 0.37391043,\n",
       "         0.5265577 , 0.64420885, 0.7782669 , 0.845487  , 0.93192935,\n",
       "         0.97899985, 1.        , 0.976105  , 0.9412413 , 0.86446244,\n",
       "         0.7844749 , 0.7008507 , 0.621785  , 0.5313992 , 0.45689875,\n",
       "         0.4292025 , 0.4051262 , 0.4149765 , 0.3860866 , 0.40759298,\n",
       "         0.41619244, 0.43504378, 0.42767978, 0.45211306, 0.44395283,\n",
       "         0.45549455, 0.46103743, 0.46506038, 0.47288734, 0.46034858,\n",
       "         0.48974046, 0.45211306, 0.51686496], dtype=float32)),\n",
       " (array([1.15217419e-01, 4.68409277e-03, 4.64128065e-02, 7.69206363e-02,\n",
       "         7.06771148e-02, 0.00000000e+00, 4.04125185e-04, 4.40314259e-02,\n",
       "         1.32111727e-01, 1.48985179e-01, 2.93048942e-01, 3.63658142e-01,\n",
       "         4.76803180e-01, 6.26641891e-01, 7.57054923e-01, 8.16735635e-01,\n",
       "         9.71322235e-01, 8.52642772e-01, 1.00000000e+00, 8.97122345e-01,\n",
       "         7.61407975e-01, 7.53470691e-01, 6.31260202e-01, 6.06886134e-01,\n",
       "         7.47929216e-01, 6.61398879e-01, 6.41170194e-01, 5.48476381e-01,\n",
       "         6.25890431e-01, 3.53512812e-01, 4.29686713e-01, 4.58293912e-01,\n",
       "         4.05618846e-01, 3.44011819e-01, 3.54338926e-01, 4.17497458e-01,\n",
       "         4.55448698e-01, 2.61232915e-01, 2.64271520e-01, 3.12855475e-01,\n",
       "         3.64486152e-01, 3.91360721e-01, 2.04147089e-01]),\n",
       "  array([0.08628361, 0.14657864, 0.02253495, 0.06236418, 0.03578521,\n",
       "         0.        , 0.03001953, 0.08328413, 0.10762674, 0.12962635,\n",
       "         0.19914344, 0.24799657, 0.3263043 , 0.35240778, 0.45154804,\n",
       "         0.55150026, 0.6782703 , 0.80930585, 0.786976  , 0.9164585 ,\n",
       "         0.91190875, 0.97602963, 0.99400175, 1.        , 0.94107157,\n",
       "         0.81038004, 0.9015981 , 0.8545148 , 0.79045165, 0.8372934 ,\n",
       "         0.82857645, 0.7344958 , 0.79383785, 0.7211472 , 0.715365  ,\n",
       "         0.70038134, 0.6457364 , 0.6391663 , 0.63496196, 0.548794  ,\n",
       "         0.5518168 , 0.50257355, 0.51543087], dtype=float32)),\n",
       " (array([6.35868741e-04, 1.57269786e-01, 0.00000000e+00, 1.49682305e-01,\n",
       "         1.89146763e-01, 1.24299152e-01, 1.19949285e-02, 1.36499317e-01,\n",
       "         1.55683175e-01, 2.16071236e-01, 1.28743467e-01, 3.97257742e-01,\n",
       "         3.86721882e-01, 7.68487441e-01, 8.92013424e-01, 8.34272860e-01,\n",
       "         8.66897000e-01, 1.00000000e+00, 9.00127873e-01, 8.87603814e-01,\n",
       "         7.24256199e-01, 5.98689452e-01, 4.27044384e-01, 5.10822012e-01,\n",
       "         3.25354082e-01, 5.16723923e-01, 3.08009111e-01, 4.68236527e-01,\n",
       "         5.04387471e-01, 4.27118120e-01, 2.92318779e-01, 3.36513521e-01,\n",
       "         3.05692321e-01, 4.92288981e-01, 2.96998571e-01, 3.89268395e-01,\n",
       "         6.42625566e-01, 4.96150497e-01, 3.19590023e-01, 2.49650491e-01,\n",
       "         3.77360919e-01, 1.97179468e-01, 1.88901575e-01]),\n",
       "  array([0.09802078, 0.06139739, 0.0509953 , 0.08641805, 0.05617737,\n",
       "         0.01728129, 0.05560815, 0.        , 0.11911378, 0.1512582 ,\n",
       "         0.2630026 , 0.15580042, 0.20740606, 0.20550758, 0.37505797,\n",
       "         0.52541804, 0.64820653, 0.7647419 , 0.88219965, 0.9059269 ,\n",
       "         0.93605375, 1.        , 0.9245438 , 0.95758843, 0.9500978 ,\n",
       "         0.7585152 , 0.6972011 , 0.66364205, 0.68814725, 0.57463926,\n",
       "         0.59005475, 0.5765082 , 0.42498228, 0.5424021 , 0.4987635 ,\n",
       "         0.5650204 , 0.52372724, 0.5423346 , 0.5679045 , 0.5639547 ,\n",
       "         0.5474081 , 0.61865944, 0.6535583 ], dtype=float32)),\n",
       " (array([0.08640992, 0.01895835, 0.        , 0.01219579, 0.05768382,\n",
       "         0.165084  , 0.08614722, 0.14954697, 0.3305215 , 0.56493153,\n",
       "         0.73518053, 0.89298612, 0.99376098, 1.        , 0.91537347,\n",
       "         0.93278096, 0.80768434, 0.73379592, 0.70642432, 0.6685583 ,\n",
       "         0.61499092, 0.5352501 , 0.45525582, 0.38972109, 0.3473569 ,\n",
       "         0.25759563, 0.20670205, 0.18778719, 0.20064068, 0.25167449,\n",
       "         0.19557363, 0.18096011, 0.20237676, 0.2219995 , 0.13739044,\n",
       "         0.1279807 , 0.12991873, 0.14877871, 0.17372689, 0.10801522,\n",
       "         0.18853767, 0.1885346 , 0.12836305]),\n",
       "  array([0.        , 0.03132579, 0.01227528, 0.02190082, 0.04829331,\n",
       "         0.00943969, 0.02819027, 0.04578158, 0.08232404, 0.13755083,\n",
       "         0.19478333, 0.35938725, 0.47185206, 0.58415675, 0.72881037,\n",
       "         0.8125219 , 0.90332323, 0.95640814, 0.9658626 , 0.9469629 ,\n",
       "         0.9796559 , 1.        , 0.93806237, 0.81501335, 0.72418624,\n",
       "         0.7028135 , 0.7071138 , 0.6590413 , 0.53805137, 0.43362775,\n",
       "         0.50042874, 0.54103416, 0.4696789 , 0.3055479 , 0.2914417 ,\n",
       "         0.37797037, 0.42864293, 0.42410892, 0.3388334 , 0.3068709 ,\n",
       "         0.2931714 , 0.31199005, 0.33894378], dtype=float32)),\n",
       " (array([0.12493935, 0.03553582, 0.        , 0.05311646, 0.14283543,\n",
       "         0.08331088, 0.03414017, 0.0938779 , 0.17416567, 0.24841865,\n",
       "         0.4715241 , 0.60018657, 0.64021144, 0.74985204, 1.        ,\n",
       "         0.90534802, 0.87861122, 0.86698488, 0.80547861, 0.64305735,\n",
       "         0.58293812, 0.623011  , 0.5871457 , 0.50274942, 0.50784425,\n",
       "         0.4272625 , 0.35823221, 0.2752784 , 0.21692682, 0.30991872,\n",
       "         0.25061391, 0.2460963 , 0.22962464, 0.19057624, 0.24804099,\n",
       "         0.22498161, 0.18452587, 0.18242963, 0.25209725, 0.34183314,\n",
       "         0.24276216, 0.1303459 , 0.0981723 ]),\n",
       "  array([0.00319967, 0.        , 0.08884395, 0.1335194 , 0.01582391,\n",
       "         0.02412288, 0.00813681, 0.09397051, 0.18938686, 0.16332209,\n",
       "         0.26392087, 0.42721164, 0.5411351 , 0.67310506, 0.7601611 ,\n",
       "         0.853145  , 0.9507649 , 0.9848438 , 1.        , 0.962115  ,\n",
       "         0.8665324 , 0.9233238 , 0.89547795, 0.7763543 , 0.76274896,\n",
       "         0.6243292 , 0.6365296 , 0.63553756, 0.5136708 , 0.5274384 ,\n",
       "         0.43008015, 0.4241087 , 0.46446705, 0.37971944, 0.3481956 ,\n",
       "         0.34911138, 0.30297917, 0.324472  , 0.37759632, 0.33651438,\n",
       "         0.3443064 , 0.31616214, 0.26849416], dtype=float32)),\n",
       " (array([0.00133695, 0.        , 0.05524775, 0.12786004, 0.17638235,\n",
       "         0.18375315, 0.21390283, 0.33899015, 0.64832204, 0.72018541,\n",
       "         0.87562841, 1.        , 0.9191497 , 0.87742421, 0.82838527,\n",
       "         0.76656255, 0.68624993, 0.69816782, 0.64995109, 0.53879001,\n",
       "         0.3769836 , 0.36313478, 0.33977984, 0.3165159 , 0.34656074,\n",
       "         0.3127269 , 0.26044967, 0.21149451, 0.20461077, 0.20905674,\n",
       "         0.22734042, 0.24301418, 0.20772067, 0.20053689, 0.18141188,\n",
       "         0.15927991, 0.18989721, 0.22497099, 0.25588922, 0.27612454,\n",
       "         0.26658721, 0.25829756, 0.22935547]),\n",
       "  array([0.06451518, 0.04086095, 0.05294639, 0.05174264, 0.        ,\n",
       "         0.0098404 , 0.02642127, 0.06835547, 0.15225051, 0.29503345,\n",
       "         0.4968295 , 0.6910515 , 0.8555579 , 0.91661364, 0.94305617,\n",
       "         0.954348  , 1.        , 0.97082233, 0.9067146 , 0.80638784,\n",
       "         0.7630662 , 0.6849848 , 0.59907377, 0.5188299 , 0.4545411 ,\n",
       "         0.36373514, 0.31540933, 0.31270358, 0.33823   , 0.30267942,\n",
       "         0.30624005, 0.33585978, 0.30546775, 0.28967518, 0.2813501 ,\n",
       "         0.27104366, 0.23297638, 0.19797173, 0.22551419, 0.25682768,\n",
       "         0.2799626 , 0.24435608, 0.2382095 ], dtype=float32)),\n",
       " (array([0.03570145, 0.        , 0.08970818, 0.14632708, 0.00754684,\n",
       "         0.12915172, 0.14739909, 0.13508038, 0.18248075, 0.38774374,\n",
       "         0.62778885, 0.80932491, 0.80365611, 0.88544758, 0.96060455,\n",
       "         1.        , 0.96388766, 0.80130814, 0.60818736, 0.48394369,\n",
       "         0.60083236, 0.58574008, 0.5139982 , 0.43204953, 0.42197523,\n",
       "         0.33894977, 0.22326541, 0.15833675, 0.33689695, 0.13398314,\n",
       "         0.06276724, 0.1525835 , 0.2865653 , 0.23361429, 0.15833471,\n",
       "         0.10551172, 0.12369903, 0.04957249, 0.08751046, 0.21731114,\n",
       "         0.1954269 , 0.15090679, 0.17540619]),\n",
       "  array([0.06102398, 0.00489912, 0.        , 0.0196771 , 0.05603271,\n",
       "         0.07504943, 0.08416041, 0.10360331, 0.17069359, 0.30279702,\n",
       "         0.43795666, 0.58528346, 0.735294  , 0.8956442 , 0.95759726,\n",
       "         0.96537983, 1.        , 0.97169185, 0.90819144, 0.8340481 ,\n",
       "         0.7866848 , 0.71385074, 0.6418191 , 0.56676584, 0.45519572,\n",
       "         0.43033153, 0.42687604, 0.4341825 , 0.4421762 , 0.40237662,\n",
       "         0.37111592, 0.3538807 , 0.33149296, 0.35325488, 0.35550094,\n",
       "         0.33721372, 0.3096005 , 0.32840607, 0.37735116, 0.4186942 ,\n",
       "         0.379221  , 0.38259968, 0.42174655], dtype=float32)),\n",
       " (array([0.1176073 , 0.01124042, 0.        , 0.05008395, 0.12456219,\n",
       "         0.08857993, 0.06059363, 0.06763742, 0.14214072, 0.24013805,\n",
       "         0.34211058, 0.45450381, 0.58859445, 0.73840509, 0.8352444 ,\n",
       "         0.89909869, 0.9672495 , 0.93261468, 0.93985809, 0.97661611,\n",
       "         1.        , 0.93756097, 0.90157119, 0.89598306, 0.9074231 ,\n",
       "         0.97670307, 0.92565171, 0.82507674, 0.91139328, 0.76871162,\n",
       "         0.72603243, 0.76981816, 0.68737635, 0.60249955, 0.56003388,\n",
       "         0.56838172, 0.63077048, 0.60042185, 0.57736648, 0.57636056,\n",
       "         0.56687344, 0.55628667, 0.53246928]),\n",
       "  array([0.050365  , 0.0803921 , 0.03898905, 0.00529256, 0.07570512,\n",
       "         0.        , 0.00633779, 0.03395089, 0.01841334, 0.02121723,\n",
       "         0.1143347 , 0.2114285 , 0.2108644 , 0.35946798, 0.41645283,\n",
       "         0.4426142 , 0.59459686, 0.6140001 , 0.6644453 , 0.7585555 ,\n",
       "         0.9179184 , 0.97865003, 0.9525136 , 0.88616025, 0.8686124 ,\n",
       "         0.9687479 , 1.        , 0.9395559 , 0.8216956 , 0.8781827 ,\n",
       "         0.8995548 , 0.87217677, 0.8592551 , 0.7853307 , 0.7894812 ,\n",
       "         0.8054916 , 0.71151423, 0.76993144, 0.73625153, 0.60953987,\n",
       "         0.5715961 , 0.5923626 , 0.5885079 ], dtype=float32)),\n",
       " (array([0.01533133, 0.0759108 , 0.0430788 , 0.02659837, 0.14036021,\n",
       "         0.11336837, 0.10391426, 0.08798022, 0.03420696, 0.00602502,\n",
       "         0.02623706, 0.04672352, 0.        , 0.17148118, 0.20202787,\n",
       "         0.20102881, 0.37454016, 0.40335506, 0.60240258, 0.85924877,\n",
       "         0.92817787, 0.9570705 , 0.95635363, 0.91768913, 0.82014704,\n",
       "         0.86282204, 0.94322658, 1.        , 0.91800966, 0.93046052,\n",
       "         0.88272442, 0.79367709, 0.77895413, 0.77161921, 0.71228244,\n",
       "         0.64393561, 0.73970182, 0.72583522, 0.60621757, 0.44894215,\n",
       "         0.51482977, 0.40028909, 0.29898524]),\n",
       "  array([0.04087755, 0.14400664, 0.09620608, 0.01040458, 0.        ,\n",
       "         0.01104657, 0.01982765, 0.06407257, 0.15198214, 0.08550519,\n",
       "         0.09571224, 0.14065163, 0.0978018 , 0.37666363, 0.522945  ,\n",
       "         0.55701685, 0.47942233, 0.6486024 , 0.7342496 , 0.7805347 ,\n",
       "         0.8728148 , 0.8864108 , 0.907794  , 0.9269272 , 0.91688377,\n",
       "         0.91680044, 0.9391003 , 0.9634435 , 0.9479648 , 1.        ,\n",
       "         0.9810057 , 0.8839848 , 0.93611264, 0.88294464, 0.8611262 ,\n",
       "         0.8441968 , 0.7450369 , 0.7450091 , 0.72074926, 0.68269277,\n",
       "         0.6814798 , 0.7974765 , 0.8352953 ], dtype=float32)),\n",
       " (array([0.1003281 , 0.        , 0.03641742, 0.11278354, 0.20563732,\n",
       "         0.39082576, 0.48756573, 0.63063993, 0.78528326, 0.91170212,\n",
       "         0.9699392 , 1.        , 0.92606445, 0.82157378, 0.73873611,\n",
       "         0.56223864, 0.50687824, 0.47959655, 0.42891514, 0.32991243,\n",
       "         0.30022279, 0.26736413, 0.26051953, 0.25045027, 0.19207457,\n",
       "         0.22854165, 0.20068211, 0.1704124 , 0.16298317, 0.18042709,\n",
       "         0.18903772, 0.19429353, 0.17239359, 0.13911938, 0.16281569,\n",
       "         0.16076984, 0.1455711 , 0.15004802, 0.17812474, 0.16788466,\n",
       "         0.21791438, 0.24629344, 0.23662734]),\n",
       "  array([0.03580873, 0.01912426, 0.00181203, 0.        , 0.03083456,\n",
       "         0.03659788, 0.11882263, 0.25134456, 0.4082621 , 0.5497173 ,\n",
       "         0.64425784, 0.77649814, 0.87624884, 0.95399696, 1.        ,\n",
       "         0.9607141 , 0.9324017 , 0.8909176 , 0.8096356 , 0.71339875,\n",
       "         0.58970547, 0.5545156 , 0.49558568, 0.43071213, 0.39278528,\n",
       "         0.36262745, 0.33373514, 0.29512376, 0.25813743, 0.23510195,\n",
       "         0.20141022, 0.21219559, 0.22299542, 0.2133064 , 0.18827078,\n",
       "         0.18040493, 0.19523062, 0.20354503, 0.16499488, 0.16641179,\n",
       "         0.19806665, 0.20388006, 0.2069654 ], dtype=float32)),\n",
       " (array([0.1961839 , 0.0424418 , 0.2254287 , 0.24257705, 0.07080527,\n",
       "         0.25738928, 0.12956552, 0.19311675, 0.25707894, 0.23622441,\n",
       "         0.32688919, 0.46247308, 0.50244678, 0.67281217, 1.        ,\n",
       "         0.63730801, 0.89248548, 0.80189105, 0.607874  , 0.66604495,\n",
       "         0.3556923 , 0.25601393, 0.33898307, 0.39320219, 0.17827817,\n",
       "         0.23546278, 0.15556575, 0.10082156, 0.15717695, 0.36325593,\n",
       "         0.14207777, 0.24234563, 0.24690944, 0.1287584 , 0.3029268 ,\n",
       "         0.14857189, 0.07567136, 0.1388964 , 0.24398364, 0.00314594,\n",
       "         0.06444999, 0.0399747 , 0.        ]),\n",
       "  array([0.01461972, 0.        , 0.06530835, 0.07557526, 0.05723245,\n",
       "         0.10268375, 0.17048335, 0.28748658, 0.4325542 , 0.585675  ,\n",
       "         0.73925513, 0.89255446, 0.96949625, 1.        , 0.9917683 ,\n",
       "         0.9728315 , 0.9562836 , 0.89160824, 0.80704826, 0.74422306,\n",
       "         0.6212181 , 0.59350425, 0.5384498 , 0.45190814, 0.4768873 ,\n",
       "         0.42338118, 0.40967193, 0.39126906, 0.3981456 , 0.3558396 ,\n",
       "         0.39586207, 0.40329853, 0.38201326, 0.35308543, 0.37393737,\n",
       "         0.3458843 , 0.34375173, 0.36673138, 0.38928416, 0.41403285,\n",
       "         0.4939301 , 0.46053097, 0.40279055], dtype=float32)),\n",
       " (array([0.11787289, 0.06013277, 0.05591049, 0.03392931, 0.        ,\n",
       "         0.06157828, 0.0759166 , 0.03682859, 0.04012003, 0.13215335,\n",
       "         0.25051605, 0.28252485, 0.33673799, 0.43213504, 0.56510333,\n",
       "         0.68757496, 0.76191237, 0.86027426, 0.95572141, 1.        ,\n",
       "         0.93666985, 0.91231497, 0.82729361, 0.72606927, 0.69486666,\n",
       "         0.54905071, 0.52225683, 0.44981819, 0.35397898, 0.37963829,\n",
       "         0.36480108, 0.31570119, 0.27458405, 0.2689828 , 0.31435011,\n",
       "         0.2576231 , 0.24259055, 0.27051235, 0.3068125 , 0.24667876,\n",
       "         0.25672476, 0.24943807, 0.25759316]),\n",
       "  array([0.06005911, 0.0317092 , 0.0605179 , 0.08219225, 0.07431678,\n",
       "         0.06839888, 0.06205969, 0.01827757, 0.        , 0.03048614,\n",
       "         0.04127644, 0.07910427, 0.13811457, 0.19611576, 0.24131393,\n",
       "         0.3083475 , 0.39401263, 0.5418918 , 0.6936123 , 0.7858317 ,\n",
       "         0.88349473, 0.9457044 , 0.99552023, 1.        , 0.9147231 ,\n",
       "         0.8544445 , 0.7674306 , 0.6957165 , 0.63378036, 0.54383945,\n",
       "         0.49298036, 0.4382061 , 0.39317554, 0.36277544, 0.34394756,\n",
       "         0.2967587 , 0.29770717, 0.2888998 , 0.25965437, 0.30817324,\n",
       "         0.29371485, 0.30493528, 0.32711363], dtype=float32)),\n",
       " (array([0.14862873, 0.25726636, 0.18180732, 0.09005944, 0.07769893,\n",
       "         0.15974207, 0.04953085, 0.        , 0.01756471, 0.07514113,\n",
       "         0.0955849 , 0.1725044 , 0.18450429, 0.20089403, 0.29096356,\n",
       "         0.44309179, 0.51112241, 0.66343195, 0.81454178, 0.84791736,\n",
       "         0.84555916, 0.91862728, 0.91786254, 0.88759076, 0.94174689,\n",
       "         1.        , 0.86085973, 0.70973525, 0.65135238, 0.74643514,\n",
       "         0.6034369 , 0.42242218, 0.31789701, 0.30819771, 0.30227514,\n",
       "         0.39035556, 0.30530272, 0.26487576, 0.33799041, 0.24537459,\n",
       "         0.31279805, 0.32304021, 0.35626481]),\n",
       "  array([0.01362176, 0.01985363, 0.00736702, 0.01216207, 0.03146413,\n",
       "         0.        , 0.0308577 , 0.02432718, 0.00866368, 0.00480266,\n",
       "         0.0179216 , 0.0526068 , 0.10248925, 0.16733125, 0.24018326,\n",
       "         0.3059989 , 0.43181273, 0.5507273 , 0.6462594 , 0.7163932 ,\n",
       "         0.7967616 , 0.867456  , 0.9440807 , 1.        , 0.9947052 ,\n",
       "         0.9618316 , 0.9324839 , 0.86128205, 0.77770483, 0.75937945,\n",
       "         0.7252565 , 0.6577465 , 0.57625973, 0.50692135, 0.49881992,\n",
       "         0.43493477, 0.42396882, 0.3938425 , 0.3419716 , 0.42011997,\n",
       "         0.35409254, 0.34668592, 0.3418497 ], dtype=float32)),\n",
       " (array([0.        , 0.00873262, 0.0969077 , 0.23479579, 0.39139223,\n",
       "         0.55008733, 0.75427905, 0.92290376, 0.95056907, 1.        ,\n",
       "         0.93466232, 0.79040209, 0.6197421 , 0.53051352, 0.42302108,\n",
       "         0.32559552, 0.29475061, 0.2621349 , 0.19722167, 0.14574144,\n",
       "         0.19945383, 0.1916601 , 0.17304726, 0.16069424, 0.18029045,\n",
       "         0.21952996, 0.20529374, 0.1829441 , 0.29158424, 0.26153764,\n",
       "         0.25545626, 0.27592198, 0.2574516 , 0.24852579, 0.2482414 ,\n",
       "         0.25762983, 0.28116562, 0.26655126, 0.21614202, 0.16287417,\n",
       "         0.23438208, 0.23229606, 0.21689309]),\n",
       "  array([0.        , 0.01349172, 0.0165676 , 0.03356536, 0.08966206,\n",
       "         0.18821181, 0.31792942, 0.4550682 , 0.64330107, 0.72455525,\n",
       "         0.8367336 , 0.950201  , 0.99953675, 1.        , 0.9436175 ,\n",
       "         0.86085474, 0.80057514, 0.7058368 , 0.61294544, 0.5287733 ,\n",
       "         0.4691781 , 0.38968974, 0.32938007, 0.2830314 , 0.23048589,\n",
       "         0.2137047 , 0.20086125, 0.19450796, 0.2095865 , 0.20969479,\n",
       "         0.20986024, 0.2108048 , 0.21062131, 0.23547797, 0.24690157,\n",
       "         0.2492645 , 0.28654197, 0.2754959 , 0.2714649 , 0.28778586,\n",
       "         0.31250864, 0.35254464, 0.3619783 ], dtype=float32)),\n",
       " (array([0.        , 0.04218194, 0.07750446, 0.10346738, 0.11706264,\n",
       "         0.15836894, 0.30060577, 0.48043093, 0.61969964, 0.82312245,\n",
       "         0.94107475, 0.99045321, 1.        , 0.97503065, 0.91529551,\n",
       "         0.81847915, 0.67809571, 0.60344452, 0.49027025, 0.37137757,\n",
       "         0.32312875, 0.31855809, 0.31087549, 0.28734286, 0.22951764,\n",
       "         0.19397939, 0.15405295, 0.1435501 , 0.26116117, 0.24911124,\n",
       "         0.23809665, 0.24120499, 0.23612254, 0.23979127, 0.24895579,\n",
       "         0.25863419, 0.25414507, 0.24797643, 0.26155942, 0.28224528,\n",
       "         0.2384822 , 0.27806496, 0.29640909]),\n",
       "  array([0.01497751, 0.        , 0.00773208, 0.02122894, 0.02893284,\n",
       "         0.14089681, 0.2589882 , 0.40461257, 0.60613906, 0.7468705 ,\n",
       "         0.86971766, 0.96223396, 1.        , 0.95942676, 0.9069275 ,\n",
       "         0.83438116, 0.7191797 , 0.6266728 , 0.55855525, 0.50097895,\n",
       "         0.4282034 , 0.38861665, 0.35297012, 0.32286105, 0.30448443,\n",
       "         0.3142759 , 0.29909733, 0.27479246, 0.31790423, 0.29973996,\n",
       "         0.2967749 , 0.31777084, 0.32285354, 0.33591446, 0.36833847,\n",
       "         0.39676023, 0.39185792, 0.38765085, 0.39408642, 0.4026302 ,\n",
       "         0.39511234, 0.38838366, 0.39218298], dtype=float32)),\n",
       " (array([0.11548659, 0.09607243, 0.07827782, 0.07166123, 0.08631343,\n",
       "         0.03709963, 0.06077413, 0.07954761, 0.        , 0.1168727 ,\n",
       "         0.11326463, 0.06974797, 0.10599154, 0.12310499, 0.17043108,\n",
       "         0.21169391, 0.18550174, 0.30779628, 0.39158859, 0.46864522,\n",
       "         0.61859952, 0.77436026, 0.87319056, 0.9255505 , 0.96361841,\n",
       "         1.        , 0.98233353, 0.94209233, 0.96782958, 0.96264249,\n",
       "         0.89302514, 0.79534928, 0.78368758, 0.74410743, 0.70318479,\n",
       "         0.67292827, 0.67767543, 0.59135611, 0.55008937, 0.55564159,\n",
       "         0.51101208, 0.48681287, 0.4697854 ]),\n",
       "  array([0.0919653 , 0.03130896, 0.06045711, 0.08218987, 0.        ,\n",
       "         0.0670526 , 0.05085983, 0.02741833, 0.09366062, 0.01649161,\n",
       "         0.01959492, 0.05078512, 0.03144497, 0.10127523, 0.09521611,\n",
       "         0.06466956, 0.11383598, 0.16508277, 0.1699925 , 0.17777187,\n",
       "         0.28190136, 0.3098561 , 0.37299314, 0.46315774, 0.54347414,\n",
       "         0.6457129 , 0.68616706, 0.69112664, 0.73932374, 0.7942715 ,\n",
       "         0.89856195, 1.        , 0.9263347 , 0.85120577, 0.7879174 ,\n",
       "         0.75388443, 0.77763814, 0.7201522 , 0.68419206, 0.67883027,\n",
       "         0.66877896, 0.6383397 , 0.6155304 ], dtype=float32)),\n",
       " (array([0.14696369, 0.11471699, 0.12210459, 0.16412501, 0.23600492,\n",
       "         0.25898173, 0.20757741, 0.11032485, 0.        , 0.02399931,\n",
       "         0.07286146, 0.10617408, 0.07194528, 0.15312182, 0.19962808,\n",
       "         0.2182877 , 0.23338037, 0.30256052, 0.3992652 , 0.49425666,\n",
       "         0.53720375, 0.54523704, 0.61820789, 0.75151349, 0.91783863,\n",
       "         0.94168455, 0.96967405, 1.        , 0.98242062, 0.97734687,\n",
       "         0.94433733, 0.91495529, 0.99600772, 0.95812842, 0.9066418 ,\n",
       "         0.84613248, 0.72834653, 0.79212484, 0.83503971, 0.80077251,\n",
       "         0.63985569, 0.64651755, 0.69083763]),\n",
       "  array([0.08550246, 0.23985106, 0.18659256, 0.09550738, 0.11968062,\n",
       "         0.07160132, 0.17689377, 0.21848309, 0.00854941, 0.19103147,\n",
       "         0.1688314 , 0.10502806, 0.28726244, 0.10039157, 0.1178828 ,\n",
       "         0.19974229, 0.        , 0.15363614, 0.14689012, 0.09063433,\n",
       "         0.25386074, 0.27974552, 0.33281755, 0.36968133, 0.4768718 ,\n",
       "         0.57663715, 0.70472807, 0.8220821 , 0.8364619 , 0.9557835 ,\n",
       "         1.        , 0.97254556, 0.8804724 , 0.92971504, 0.977299  ,\n",
       "         0.9818325 , 0.95375746, 0.93300456, 0.94491863, 0.9349388 ,\n",
       "         0.87122726, 0.8469789 , 0.81595945], dtype=float32)),\n",
       " (array([0.        , 0.00970449, 0.01940899, 0.02911348, 0.03940386,\n",
       "         0.06064613, 0.08285077, 0.11375454, 0.158957  , 0.28493757,\n",
       "         0.41091819, 0.54837924, 0.69073308, 0.8009023 , 0.90034319,\n",
       "         0.95566018, 1.        , 0.90637267, 0.78832357, 0.73565114,\n",
       "         0.69455111, 0.55244429, 0.39916318, 0.29521395, 0.28303846,\n",
       "         0.25622475, 0.20748944, 0.16467063, 0.12904892, 0.11136496,\n",
       "         0.11161875, 0.11187255, 0.11212635, 0.11187636, 0.11121547,\n",
       "         0.1105546 , 0.10989373, 0.10923284, 0.11903243, 0.13445491,\n",
       "         0.14987739, 0.16529987, 0.18072235]),\n",
       "  array([0.07492509, 0.07059561, 0.06651073, 0.06287225, 0.06020302,\n",
       "         0.05947533, 0.05035009, 0.02402312, 0.        , 0.06371002,\n",
       "         0.12182933, 0.20824772, 0.31112182, 0.4419663 , 0.57971627,\n",
       "         0.73573965, 0.8855715 , 0.96033907, 1.        , 0.9969379 ,\n",
       "         0.98533297, 0.9446019 , 0.8980661 , 0.84401333, 0.77469885,\n",
       "         0.70463216, 0.64181036, 0.5956292 , 0.57249284, 0.5475066 ,\n",
       "         0.5153336 , 0.48326302, 0.4513606 , 0.43657738, 0.43578854,\n",
       "         0.4353941 , 0.43562648, 0.4367104 , 0.4394897 , 0.4430716 ,\n",
       "         0.4473384 , 0.45212194, 0.4575292 ], dtype=float32)),\n",
       " (array([0.09078003, 0.06652885, 0.04227766, 0.01802647, 0.        ,\n",
       "         0.03703827, 0.06679247, 0.06726746, 0.07712896, 0.11514996,\n",
       "         0.15317095, 0.2589808 , 0.38165526, 0.51237928, 0.64450639,\n",
       "         0.81464888, 0.98908871, 0.99894681, 1.        , 0.96623853,\n",
       "         0.92841792, 0.81472226, 0.70102665, 0.62018678, 0.57220277,\n",
       "         0.51086509, 0.43863606, 0.36683461, 0.29531643, 0.26209036,\n",
       "         0.24962507, 0.23715979, 0.22469451, 0.21398363, 0.20402747,\n",
       "         0.19407131, 0.18411515, 0.17415896, 0.17079602, 0.1696308 ,\n",
       "         0.16846557, 0.16730037, 0.16613515]),\n",
       "  array([1.4740463e-02, 9.9205952e-03, 5.1186355e-03, 3.6662660e-04,\n",
       "         0.0000000e+00, 4.6860348e-02, 7.3097244e-02, 3.9840721e-02,\n",
       "         2.0721473e-02, 7.3289506e-02, 1.1904432e-01, 1.4178176e-01,\n",
       "         1.4852148e-01, 2.7276832e-01, 4.0685374e-01, 5.7880163e-01,\n",
       "         6.9304776e-01, 7.7780163e-01, 8.3866543e-01, 9.0443110e-01,\n",
       "         9.6264553e-01, 9.8320681e-01, 1.0000000e+00, 9.7902215e-01,\n",
       "         9.1626775e-01, 8.6937535e-01, 8.3685774e-01, 7.8979951e-01,\n",
       "         7.3268682e-01, 6.8775475e-01, 6.4950967e-01, 6.1091399e-01,\n",
       "         5.7187623e-01, 5.5910462e-01, 5.5985016e-01, 5.5815369e-01,\n",
       "         5.5428761e-01, 5.4819441e-01, 5.3887045e-01, 5.2775097e-01,\n",
       "         5.1481259e-01, 4.9990433e-01, 4.8320067e-01], dtype=float32)),\n",
       " (array([0.03366059, 0.02482444, 0.01598829, 0.00715214, 0.        ,\n",
       "         0.02708969, 0.05045728, 0.03943279, 0.03051642, 0.03369403,\n",
       "         0.03687163, 0.09137038, 0.16794584, 0.27426232, 0.39049245,\n",
       "         0.55695494, 0.74016164, 0.86216083, 0.96855811, 0.99175549,\n",
       "         1.        , 0.95257321, 0.8988532 , 0.81177085, 0.67431789,\n",
       "         0.55367594, 0.45364565, 0.38013872, 0.33315503, 0.28902969,\n",
       "         0.24727276, 0.20551578, 0.16375881, 0.14753922, 0.15247914,\n",
       "         0.15741906, 0.16235901, 0.16729893, 0.16827748, 0.16658924,\n",
       "         0.164901  , 0.16321276, 0.16152452]),\n",
       "  array([0.05251657, 0.05337547, 0.05427564, 0.05522642, 0.0554022 ,\n",
       "         0.0387569 , 0.02313696, 0.01923264, 0.01495413, 0.00731328,\n",
       "         0.        , 0.00901775, 0.02523427, 0.06069141, 0.1010649 ,\n",
       "         0.21014884, 0.34094623, 0.47723788, 0.6087597 , 0.7345821 ,\n",
       "         0.8596096 , 0.9210161 , 0.9630328 , 0.98843086, 1.        ,\n",
       "         0.9672767 , 0.8823497 , 0.8165356 , 0.76945335, 0.74190074,\n",
       "         0.72684276, 0.7005898 , 0.6630393 , 0.62119573, 0.5769126 ,\n",
       "         0.52502054, 0.4652799 , 0.39901567, 0.37418622, 0.3779454 ,\n",
       "         0.3780546 , 0.37487733, 0.36871856], dtype=float32)),\n",
       " (array([0.        , 0.00938009, 0.01876018, 0.0281403 , 0.03695209,\n",
       "         0.03514123, 0.03367898, 0.03543789, 0.03699428, 0.03738891,\n",
       "         0.03778356, 0.03441579, 0.02942952, 0.06994921, 0.12563752,\n",
       "         0.20705232, 0.29486737, 0.45032587, 0.61794141, 0.71521018,\n",
       "         0.79983591, 0.8709491 , 0.93963377, 0.99051293, 1.        ,\n",
       "         0.96976797, 0.86500682, 0.76321203, 0.66589607, 0.58394432,\n",
       "         0.52083045, 0.45771665, 0.39460279, 0.34223972, 0.30305802,\n",
       "         0.26387633, 0.22469468, 0.18551296, 0.17125438, 0.17732314,\n",
       "         0.18339191, 0.18946068, 0.19552947]),\n",
       "  array([0.02160654, 0.05087413, 0.07319889, 0.08527584, 0.08765095,\n",
       "         0.08491965, 0.07967312, 0.05347842, 0.02172333, 0.0143411 ,\n",
       "         0.        , 0.00860116, 0.02686913, 0.0299537 , 0.03001939,\n",
       "         0.06192046, 0.10255991, 0.20118712, 0.3137715 , 0.41424826,\n",
       "         0.5108712 , 0.63093704, 0.75200576, 0.84342885, 0.8730847 ,\n",
       "         0.9064792 , 0.9437888 , 0.97290313, 1.        , 0.9879172 ,\n",
       "         0.92475593, 0.8626326 , 0.8015456 , 0.74862486, 0.7045461 ,\n",
       "         0.6610849 , 0.6174265 , 0.5730996 , 0.5401402 , 0.5162403 ,\n",
       "         0.491593  , 0.46659824, 0.442548  ], dtype=float32))]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results_meassures"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "id": "043082cf",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "prediction_ids = []\n",
    "prediction_meassures = []\n",
    "\n",
    "type_predictions = 'AIF'\n",
    "results_meassures = []\n",
    "prefix_fig = '/home/sebastian/experiments/aifnet_replication/results/predictions_aif/'+path_tensorboard_log.split('/')[-1]+'_'\n",
    "for case_number in range(len(ctp_volumes_test)):\n",
    "    case_id = ctp_volumes_test[case_number]['image'].split('.')[-2]\n",
    "    prediction_ids.append(case_id)\n",
    "    x =  ctp_volumes_test[case_number]['ctpvals']\n",
    "    if type_predictions == 'AIF':\n",
    "        y = aif_annotations_test[case_id]\n",
    "    if type_predictions == 'VOF':\n",
    "        y = vof_annotations_test[case_id]\n",
    "    results_meassures.append(plot_predictions(model,x,y, prefix_fig, True, type_predictions,True))\n",
    "\n",
    "preds_fold = tfp.stats.correlation(np.array(results_meassures)[:,1,:],np.array(results_meassures)[:,0,:], sample_axis=0, event_axis=None)\n",
    "preds_fold = preds_fold.numpy()\n",
    "prediction_meassures.append([preds_fold.mean(),preds_fold.std(),preds_fold.var()])\n",
    "\n",
    "savetxt('results/pearson_fold_'+str(current_fold)+'.csv', prediction_meassures, delimiter=',',fmt='%1.5f')\n",
    "savetxt('results/allpreds_fold_'+str(current_fold)+'.csv', np.array(results_meassures)[:,1,:], delimiter=',',fmt='%1.5f')\n",
    "\n",
    "\n",
    "test_ids_file=open('results/pred_ids_fold_'+str(current_fold)+'.csv','w')\n",
    "for element in prediction_ids:        \n",
    "    test_ids_file.write(element+'\\n')\n",
    "test_ids_file.close()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "id": "43529e3f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['345624',\n",
       " '339335',\n",
       " '339343',\n",
       " '345766',\n",
       " '345773',\n",
       " '345794',\n",
       " '345801',\n",
       " '345871',\n",
       " '345878',\n",
       " '345885',\n",
       " '345892',\n",
       " '345927',\n",
       " '345934',\n",
       " '345976',\n",
       " '345983',\n",
       " '346018',\n",
       " '346025',\n",
       " '346116',\n",
       " '346123',\n",
       " '346158',\n",
       " '346165']"
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prediction_ids"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a7aabc53",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.9.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}