Skip to content

GitLab

Commit 05c9a892 authored by Jonathan Lee's avatar Jonathan Lee
Browse files

changed normalization and updated annotations

parent bc9b39a1
Hide whitespace changes
Inline Side-by-side
Showing with 148 additions and 130 deletions
.ipynb_checkpoints/dexa-score-checkpoint.ipynb
View file @ 05c9a892
......@@ -2,9 +2,18 @@
"cells": [
{
"cell_type": "code",
"execution_count": 384,
"execution_count": 1,
"metadata": {},
"outputs": [],
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/ifshome/jolee/miniconda/lib/python3.8/site-packages/nilearn/datasets/__init__.py:87: FutureWarning: Fetchers from the nilearn.datasets module will be updated in version 0.9 to return python strings instead of bytes and Pandas dataframes instead of Numpy arrays.\n",
" warn(\"Fetchers from the nilearn.datasets module will be \"\n"
]
}
],
"source": [
"# import necessary packages\n",
"from sklearn.preprocessing import LabelBinarizer\n",
......@@ -30,7 +39,7 @@
},
{
"cell_type": "code",
"execution_count": 385,
"execution_count": 2,
"metadata": {},
"outputs": [
{
......@@ -139,7 +148,7 @@
},
{
"cell_type": "code",
"execution_count": 389,
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
......@@ -177,7 +186,7 @@
},
{
"cell_type": "code",
"execution_count": 390,
"execution_count": 4,
"metadata": {},
"outputs": [],
"source": [
......@@ -195,7 +204,7 @@
},
{
"cell_type": "code",
"execution_count": 391,
"execution_count": 5,
"metadata": {},
"outputs": [
{
......@@ -204,7 +213,7 @@
"(128, 184, 1)"
]
},
"execution_count": 391,
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
......@@ -219,7 +228,7 @@
},
{
"cell_type": "code",
"execution_count": 421,
"execution_count": 6,
"metadata": {},
"outputs": [],
"source": [
......@@ -237,14 +246,13 @@
"for i in imgs_avg:\n",
" imgs_avg_nib.append(nib.load(i).get_fdata())\n",
"\n",
"# divide by 255 to normalize pixel values\n",
"for i in range(len(imgs_max)): \n",
" imgs_merged.append(np.concatenate((imgs_max_nib[i] / 255, imgs_avg_nib[i] / 255), axis=2))"
" imgs_merged.append(np.concatenate((imgs_max_nib[i], imgs_avg_nib[i]), axis=2))"
]
},
{
"cell_type": "code",
"execution_count": 422,
"execution_count": 7,
"metadata": {},
"outputs": [
{
......@@ -253,7 +261,7 @@
"(128, 184, 2)"
]
},
"execution_count": 422,
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
......@@ -266,14 +274,14 @@
},
{
"cell_type": "code",
"execution_count": 423,
"execution_count": 8,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"<ipython-input-423-acee21f7a78c>:8: MatplotlibDeprecationWarning: Passing non-integers as three-element position specification is deprecated since 3.3 and will be removed two minor releases later.\n",
"<ipython-input-8-acee21f7a78c>:8: MatplotlibDeprecationWarning: Passing non-integers as three-element position specification is deprecated since 3.3 and will be removed two minor releases later.\n",
" plt.subplot(5 / columns + 1, columns, i + 1)\n"
]
},
......@@ -304,14 +312,14 @@
},
{
"cell_type": "code",
"execution_count": 424,
"execution_count": 9,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"<ipython-input-424-5117e302ece2>:8: MatplotlibDeprecationWarning: Passing non-integers as three-element position specification is deprecated since 3.3 and will be removed two minor releases later.\n",
"<ipython-input-9-5117e302ece2>:8: MatplotlibDeprecationWarning: Passing non-integers as three-element position specification is deprecated since 3.3 and will be removed two minor releases later.\n",
" plt.subplot(5 / columns + 1, columns, i + 1)\n"
]
},
......@@ -342,7 +350,7 @@
},
{
"cell_type": "code",
"execution_count": 445,
"execution_count": 10,
"metadata": {},
"outputs": [],
"source": [
......@@ -356,16 +364,16 @@
},
{
"cell_type": "code",
"execution_count": 446,
"execution_count": 11,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.image.AxesImage at 0x7fe71c447ac0>"
"<matplotlib.image.AxesImage at 0x7f297d00d850>"
]
},
"execution_count": 446,
"execution_count": 11,
"metadata": {},
"output_type": "execute_result"
},
......@@ -391,7 +399,7 @@
},
{
"cell_type": "code",
"execution_count": 458,
"execution_count": 12,
"metadata": {},
"outputs": [],
"source": [
......@@ -419,7 +427,7 @@
},
{
"cell_type": "code",
"execution_count": 451,
"execution_count": 13,
"metadata": {},
"outputs": [
{
......@@ -444,7 +452,7 @@
},
{
"cell_type": "code",
"execution_count": 452,
"execution_count": 14,
"metadata": {},
"outputs": [],
"source": [
......@@ -462,6 +470,7 @@
"from tensorflow.keras.layers import Input\n",
"from tensorflow.keras.models import Model\n",
"\n",
"# double check with regress is set to false here \n",
"def create_mlp(dim, regress=False):\n",
" \n",
" # define our MLP network\n",
......@@ -497,7 +506,7 @@
},
{
"cell_type": "code",
"execution_count": 453,
"execution_count": 15,
"metadata": {
"scrolled": false
},
......@@ -516,88 +525,88 @@
"text": [
"[INFO] training model...\n",
"Epoch 1/40\n",
"23/23 [==============================] - 2s 40ms/step - loss: 5907202.5000 - val_loss: 85.3816\n",
"23/23 [==============================] - 2s 43ms/step - loss: 739162.7500 - val_loss: 80.7780\n",
"Epoch 2/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 9870156.0000 - val_loss: 82.9214\n",
"23/23 [==============================] - 1s 33ms/step - loss: 108194.5469 - val_loss: 77.7209\n",
"Epoch 3/40\n",
"23/23 [==============================] - 1s 32ms/step - loss: 10330770.0000 - val_loss: 89.8360\n",
"23/23 [==============================] - 1s 33ms/step - loss: 4658813.0000 - val_loss: 175.1640\n",
"Epoch 4/40\n",
"23/23 [==============================] - 1s 32ms/step - loss: 1725934.7500 - val_loss: 94.6955\n",
"23/23 [==============================] - 1s 34ms/step - loss: 9749125.0000 - val_loss: 248.0767\n",
"Epoch 5/40\n",
"23/23 [==============================] - 1s 32ms/step - loss: 709866.3750 - val_loss: 94.1317\n",
"23/23 [==============================] - 1s 33ms/step - loss: 4491801.0000 - val_loss: 274.3863\n",
"Epoch 6/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 1453704.2500 - val_loss: 82.3401\n",
"23/23 [==============================] - 1s 33ms/step - loss: 254635.0469 - val_loss: 380.1227\n",
"Epoch 7/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 1854183.1250 - val_loss: 110.8519\n",
"23/23 [==============================] - 1s 33ms/step - loss: 82372.6016 - val_loss: 377.8203\n",
"Epoch 8/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 3644127.0000 - val_loss: 111.8308\n",
"23/23 [==============================] - 1s 33ms/step - loss: 5235496.5000 - val_loss: 294.4142\n",
"Epoch 9/40\n",
"23/23 [==============================] - 1s 36ms/step - loss: 1630926.7500 - val_loss: 112.9966\n",
"23/23 [==============================] - 1s 33ms/step - loss: 499993.4688 - val_loss: 302.4116\n",
"Epoch 10/40\n",
"23/23 [==============================] - 1s 35ms/step - loss: 315047.1875 - val_loss: 81.3186\n",
"23/23 [==============================] - 1s 33ms/step - loss: 3674149.7500 - val_loss: 301.1278\n",
"Epoch 11/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 799771.1250 - val_loss: 75.0370\n",
"23/23 [==============================] - 1s 33ms/step - loss: 236464.9531 - val_loss: 246.8560\n",
"Epoch 12/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 1064065.3750 - val_loss: 77.3539\n",
"23/23 [==============================] - 1s 33ms/step - loss: 19876822.0000 - val_loss: 69.6813\n",
"Epoch 13/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 1303266.5000 - val_loss: 95.6228\n",
"23/23 [==============================] - 1s 33ms/step - loss: 2988911.0000 - val_loss: 80.7738\n",
"Epoch 14/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 2436036.5000 - val_loss: 98.8011\n",
"23/23 [==============================] - 1s 33ms/step - loss: 9260542.0000 - val_loss: 85.6476\n",
"Epoch 15/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 9746630.0000 - val_loss: 77.0218\n",
"23/23 [==============================] - 1s 33ms/step - loss: 1247110.8750 - val_loss: 88.1314\n",
"Epoch 16/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 21851116.0000 - val_loss: 80.2659\n",
"23/23 [==============================] - 1s 33ms/step - loss: 2866541.7500 - val_loss: 90.6115\n",
"Epoch 17/40\n",
"23/23 [==============================] - 1s 35ms/step - loss: 10444255.0000 - val_loss: 73.1865\n",
"23/23 [==============================] - 1s 36ms/step - loss: 158247.6875 - val_loss: 88.9790\n",
"Epoch 18/40\n",
"23/23 [==============================] - 1s 35ms/step - loss: 1842511.8750 - val_loss: 130.2612\n",
"23/23 [==============================] - 1s 34ms/step - loss: 410635.1250 - val_loss: 90.1120\n",
"Epoch 19/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 846315.1875 - val_loss: 81.5081\n",
"23/23 [==============================] - 1s 34ms/step - loss: 8200141.0000 - val_loss: 90.3297\n",
"Epoch 20/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 7538889.0000 - val_loss: 79.4730\n",
"23/23 [==============================] - 1s 34ms/step - loss: 2934102.5000 - val_loss: 90.0181\n",
"Epoch 21/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 3532151.0000 - val_loss: 69.8874\n",
"23/23 [==============================] - 1s 33ms/step - loss: 4385727.5000 - val_loss: 91.2359\n",
"Epoch 22/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 1252445.0000 - val_loss: 114.9249\n",
"23/23 [==============================] - 1s 35ms/step - loss: 7016348.5000 - val_loss: 92.0402\n",
"Epoch 23/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 2978219.5000 - val_loss: 108.4571\n",
"23/23 [==============================] - 1s 37ms/step - loss: 1570839.8750 - val_loss: 92.5150\n",
"Epoch 24/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 922540.4375 - val_loss: 81.8545\n",
"23/23 [==============================] - 1s 35ms/step - loss: 578274.3125 - val_loss: 87.3443\n",
"Epoch 25/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 3633383.0000 - val_loss: 69.3617\n",
"23/23 [==============================] - 1s 35ms/step - loss: 561783.6875 - val_loss: 87.6562\n",
"Epoch 26/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 1673403.2500 - val_loss: 68.7792\n",
"23/23 [==============================] - 1s 41ms/step - loss: 9761388.0000 - val_loss: 88.0854\n",
"Epoch 27/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 3747849.2500 - val_loss: 66.1937\n",
"23/23 [==============================] - 1s 34ms/step - loss: 671722.1250 - val_loss: 87.9868\n",
"Epoch 28/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 5839762.0000 - val_loss: 61.7886\n",
"23/23 [==============================] - 1s 35ms/step - loss: 291089.4688 - val_loss: 88.8749\n",
"Epoch 29/40\n",
"23/23 [==============================] - 1s 35ms/step - loss: 6155945.0000 - val_loss: 57.3682\n",
"23/23 [==============================] - 1s 34ms/step - loss: 6084078.0000 - val_loss: 90.0092\n",
"Epoch 30/40\n",
"23/23 [==============================] - 1s 36ms/step - loss: 9463727.0000 - val_loss: 61.3495\n",
"23/23 [==============================] - 1s 34ms/step - loss: 674623.6250 - val_loss: 91.0310\n",
"Epoch 31/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 267309.3750 - val_loss: 77.1904\n",
"23/23 [==============================] - 1s 33ms/step - loss: 3171207.5000 - val_loss: 92.0963\n",
"Epoch 32/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 5791735.5000 - val_loss: 76.2954\n",
"23/23 [==============================] - 1s 33ms/step - loss: 1611184.8750 - val_loss: 94.2985\n",
"Epoch 33/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 1950277.5000 - val_loss: 143.1821\n",
"23/23 [==============================] - 1s 32ms/step - loss: 3172797.7500 - val_loss: 98.5941\n",
"Epoch 34/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 288345.4375 - val_loss: 111.6001\n",
"23/23 [==============================] - 1s 34ms/step - loss: 423511.1875 - val_loss: 104.0902\n",
"Epoch 35/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 15917235.0000 - val_loss: 93.8450\n",
"23/23 [==============================] - 1s 35ms/step - loss: 126338.9531 - val_loss: 102.0765\n",
"Epoch 36/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 6554179.5000 - val_loss: 89.9294\n",
"23/23 [==============================] - 1s 34ms/step - loss: 520707.8750 - val_loss: 101.0857\n",
"Epoch 37/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 916703.4375 - val_loss: 86.5181\n",
"23/23 [==============================] - 1s 34ms/step - loss: 1672894.8750 - val_loss: 100.0803\n",
"Epoch 38/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 1636418.8750 - val_loss: 81.6224\n",
"23/23 [==============================] - 1s 34ms/step - loss: 3788303.0000 - val_loss: 100.6476\n",
"Epoch 39/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 3871.7839 - val_loss: 72.8129\n",
"23/23 [==============================] - 1s 35ms/step - loss: 2577893.5000 - val_loss: 98.6984\n",
"Epoch 40/40\n",
"23/23 [==============================] - 1s 35ms/step - loss: 740887.7500 - val_loss: 73.8067\n",
"23/23 [==============================] - 1s 34ms/step - loss: 1781253.5000 - val_loss: 97.7365\n",
"[INFO] predicting t-score...\n",
"[INFO] average t-score: -1.47828, std t-score: 1.406043864383835\n",
"[INFO] mean absolute percentage error: 73.81%, std: 54.82%\n"
"[INFO] mean absolute percentage error: 97.74%, std: 18.17%\n"
]
}
],
......
dexa-score.ipynb
View file @ 05c9a892
......@@ -2,9 +2,18 @@
"cells": [
{
"cell_type": "code",
"execution_count": 384,
"execution_count": 1,
"metadata": {},
"outputs": [],
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/ifshome/jolee/miniconda/lib/python3.8/site-packages/nilearn/datasets/__init__.py:87: FutureWarning: Fetchers from the nilearn.datasets module will be updated in version 0.9 to return python strings instead of bytes and Pandas dataframes instead of Numpy arrays.\n",
" warn(\"Fetchers from the nilearn.datasets module will be \"\n"
]
}
],
"source": [
"# import necessary packages\n",
"from sklearn.preprocessing import LabelBinarizer\n",
......@@ -30,7 +39,7 @@
},
{
"cell_type": "code",
"execution_count": 385,
"execution_count": 2,
"metadata": {},
"outputs": [
{
......@@ -139,7 +148,7 @@
},
{
"cell_type": "code",
"execution_count": 389,
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
......@@ -177,7 +186,7 @@
},
{
"cell_type": "code",
"execution_count": 390,
"execution_count": 4,
"metadata": {},
"outputs": [],
"source": [
......@@ -195,7 +204,7 @@
},
{
"cell_type": "code",
"execution_count": 391,
"execution_count": 5,
"metadata": {},
"outputs": [
{
......@@ -204,7 +213,7 @@
"(128, 184, 1)"
]
},
"execution_count": 391,
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
......@@ -219,7 +228,7 @@
},
{
"cell_type": "code",
"execution_count": 421,
"execution_count": 6,
"metadata": {},
"outputs": [],
"source": [
......@@ -237,14 +246,13 @@
"for i in imgs_avg:\n",
" imgs_avg_nib.append(nib.load(i).get_fdata())\n",
"\n",
"# divide by 255 to normalize pixel values\n",
"for i in range(len(imgs_max)): \n",
" imgs_merged.append(np.concatenate((imgs_max_nib[i] / 255, imgs_avg_nib[i] / 255), axis=2))"
" imgs_merged.append(np.concatenate((imgs_max_nib[i], imgs_avg_nib[i]), axis=2))"
]
},
{
"cell_type": "code",
"execution_count": 422,
"execution_count": 7,
"metadata": {},
"outputs": [
{
......@@ -253,7 +261,7 @@
"(128, 184, 2)"
]
},
"execution_count": 422,
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
......@@ -266,14 +274,14 @@
},
{
"cell_type": "code",
"execution_count": 423,
"execution_count": 8,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"<ipython-input-423-acee21f7a78c>:8: MatplotlibDeprecationWarning: Passing non-integers as three-element position specification is deprecated since 3.3 and will be removed two minor releases later.\n",
"<ipython-input-8-acee21f7a78c>:8: MatplotlibDeprecationWarning: Passing non-integers as three-element position specification is deprecated since 3.3 and will be removed two minor releases later.\n",
" plt.subplot(5 / columns + 1, columns, i + 1)\n"
]
},
......@@ -304,14 +312,14 @@
},
{
"cell_type": "code",
"execution_count": 424,
"execution_count": 9,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"<ipython-input-424-5117e302ece2>:8: MatplotlibDeprecationWarning: Passing non-integers as three-element position specification is deprecated since 3.3 and will be removed two minor releases later.\n",
"<ipython-input-9-5117e302ece2>:8: MatplotlibDeprecationWarning: Passing non-integers as three-element position specification is deprecated since 3.3 and will be removed two minor releases later.\n",
" plt.subplot(5 / columns + 1, columns, i + 1)\n"
]
},
......@@ -342,7 +350,7 @@
},
{
"cell_type": "code",
"execution_count": 445,
"execution_count": 10,
"metadata": {},
"outputs": [],
"source": [
......@@ -356,16 +364,16 @@
},
{
"cell_type": "code",
"execution_count": 446,
"execution_count": 11,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.image.AxesImage at 0x7fe71c447ac0>"
"<matplotlib.image.AxesImage at 0x7f297d00d850>"
]
},
"execution_count": 446,
"execution_count": 11,
"metadata": {},
"output_type": "execute_result"
},
......@@ -391,7 +399,7 @@
},
{
"cell_type": "code",
"execution_count": 458,
"execution_count": 12,
"metadata": {},
"outputs": [],
"source": [
......@@ -419,7 +427,7 @@
},
{
"cell_type": "code",
"execution_count": 451,
"execution_count": 13,
"metadata": {},
"outputs": [
{
......@@ -444,7 +452,7 @@
},
{
"cell_type": "code",
"execution_count": 452,
"execution_count": 14,
"metadata": {},
"outputs": [],
"source": [
......@@ -462,6 +470,7 @@
"from tensorflow.keras.layers import Input\n",
"from tensorflow.keras.models import Model\n",
"\n",
"# double check with regress is set to false here \n",
"def create_mlp(dim, regress=False):\n",
" \n",
" # define our MLP network\n",
......@@ -497,7 +506,7 @@
},
{
"cell_type": "code",
"execution_count": 453,
"execution_count": 15,
"metadata": {
"scrolled": false
},
......@@ -516,88 +525,88 @@
"text": [
"[INFO] training model...\n",
"Epoch 1/40\n",
"23/23 [==============================] - 2s 40ms/step - loss: 5907202.5000 - val_loss: 85.3816\n",
"23/23 [==============================] - 2s 43ms/step - loss: 739162.7500 - val_loss: 80.7780\n",
"Epoch 2/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 9870156.0000 - val_loss: 82.9214\n",
"23/23 [==============================] - 1s 33ms/step - loss: 108194.5469 - val_loss: 77.7209\n",
"Epoch 3/40\n",
"23/23 [==============================] - 1s 32ms/step - loss: 10330770.0000 - val_loss: 89.8360\n",
"23/23 [==============================] - 1s 33ms/step - loss: 4658813.0000 - val_loss: 175.1640\n",
"Epoch 4/40\n",
"23/23 [==============================] - 1s 32ms/step - loss: 1725934.7500 - val_loss: 94.6955\n",
"23/23 [==============================] - 1s 34ms/step - loss: 9749125.0000 - val_loss: 248.0767\n",
"Epoch 5/40\n",
"23/23 [==============================] - 1s 32ms/step - loss: 709866.3750 - val_loss: 94.1317\n",
"23/23 [==============================] - 1s 33ms/step - loss: 4491801.0000 - val_loss: 274.3863\n",
"Epoch 6/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 1453704.2500 - val_loss: 82.3401\n",
"23/23 [==============================] - 1s 33ms/step - loss: 254635.0469 - val_loss: 380.1227\n",
"Epoch 7/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 1854183.1250 - val_loss: 110.8519\n",
"23/23 [==============================] - 1s 33ms/step - loss: 82372.6016 - val_loss: 377.8203\n",
"Epoch 8/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 3644127.0000 - val_loss: 111.8308\n",
"23/23 [==============================] - 1s 33ms/step - loss: 5235496.5000 - val_loss: 294.4142\n",
"Epoch 9/40\n",
"23/23 [==============================] - 1s 36ms/step - loss: 1630926.7500 - val_loss: 112.9966\n",
"23/23 [==============================] - 1s 33ms/step - loss: 499993.4688 - val_loss: 302.4116\n",
"Epoch 10/40\n",
"23/23 [==============================] - 1s 35ms/step - loss: 315047.1875 - val_loss: 81.3186\n",
"23/23 [==============================] - 1s 33ms/step - loss: 3674149.7500 - val_loss: 301.1278\n",
"Epoch 11/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 799771.1250 - val_loss: 75.0370\n",
"23/23 [==============================] - 1s 33ms/step - loss: 236464.9531 - val_loss: 246.8560\n",
"Epoch 12/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 1064065.3750 - val_loss: 77.3539\n",
"23/23 [==============================] - 1s 33ms/step - loss: 19876822.0000 - val_loss: 69.6813\n",
"Epoch 13/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 1303266.5000 - val_loss: 95.6228\n",
"23/23 [==============================] - 1s 33ms/step - loss: 2988911.0000 - val_loss: 80.7738\n",
"Epoch 14/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 2436036.5000 - val_loss: 98.8011\n",
"23/23 [==============================] - 1s 33ms/step - loss: 9260542.0000 - val_loss: 85.6476\n",
"Epoch 15/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 9746630.0000 - val_loss: 77.0218\n",
"23/23 [==============================] - 1s 33ms/step - loss: 1247110.8750 - val_loss: 88.1314\n",
"Epoch 16/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 21851116.0000 - val_loss: 80.2659\n",
"23/23 [==============================] - 1s 33ms/step - loss: 2866541.7500 - val_loss: 90.6115\n",
"Epoch 17/40\n",
"23/23 [==============================] - 1s 35ms/step - loss: 10444255.0000 - val_loss: 73.1865\n",
"23/23 [==============================] - 1s 36ms/step - loss: 158247.6875 - val_loss: 88.9790\n",
"Epoch 18/40\n",
"23/23 [==============================] - 1s 35ms/step - loss: 1842511.8750 - val_loss: 130.2612\n",
"23/23 [==============================] - 1s 34ms/step - loss: 410635.1250 - val_loss: 90.1120\n",
"Epoch 19/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 846315.1875 - val_loss: 81.5081\n",
"23/23 [==============================] - 1s 34ms/step - loss: 8200141.0000 - val_loss: 90.3297\n",
"Epoch 20/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 7538889.0000 - val_loss: 79.4730\n",
"23/23 [==============================] - 1s 34ms/step - loss: 2934102.5000 - val_loss: 90.0181\n",
"Epoch 21/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 3532151.0000 - val_loss: 69.8874\n",
"23/23 [==============================] - 1s 33ms/step - loss: 4385727.5000 - val_loss: 91.2359\n",
"Epoch 22/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 1252445.0000 - val_loss: 114.9249\n",
"23/23 [==============================] - 1s 35ms/step - loss: 7016348.5000 - val_loss: 92.0402\n",
"Epoch 23/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 2978219.5000 - val_loss: 108.4571\n",
"23/23 [==============================] - 1s 37ms/step - loss: 1570839.8750 - val_loss: 92.5150\n",
"Epoch 24/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 922540.4375 - val_loss: 81.8545\n",
"23/23 [==============================] - 1s 35ms/step - loss: 578274.3125 - val_loss: 87.3443\n",
"Epoch 25/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 3633383.0000 - val_loss: 69.3617\n",
"23/23 [==============================] - 1s 35ms/step - loss: 561783.6875 - val_loss: 87.6562\n",
"Epoch 26/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 1673403.2500 - val_loss: 68.7792\n",
"23/23 [==============================] - 1s 41ms/step - loss: 9761388.0000 - val_loss: 88.0854\n",
"Epoch 27/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 3747849.2500 - val_loss: 66.1937\n",
"23/23 [==============================] - 1s 34ms/step - loss: 671722.1250 - val_loss: 87.9868\n",
"Epoch 28/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 5839762.0000 - val_loss: 61.7886\n",
"23/23 [==============================] - 1s 35ms/step - loss: 291089.4688 - val_loss: 88.8749\n",
"Epoch 29/40\n",
"23/23 [==============================] - 1s 35ms/step - loss: 6155945.0000 - val_loss: 57.3682\n",
"23/23 [==============================] - 1s 34ms/step - loss: 6084078.0000 - val_loss: 90.0092\n",
"Epoch 30/40\n",
"23/23 [==============================] - 1s 36ms/step - loss: 9463727.0000 - val_loss: 61.3495\n",
"23/23 [==============================] - 1s 34ms/step - loss: 674623.6250 - val_loss: 91.0310\n",
"Epoch 31/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 267309.3750 - val_loss: 77.1904\n",
"23/23 [==============================] - 1s 33ms/step - loss: 3171207.5000 - val_loss: 92.0963\n",
"Epoch 32/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 5791735.5000 - val_loss: 76.2954\n",
"23/23 [==============================] - 1s 33ms/step - loss: 1611184.8750 - val_loss: 94.2985\n",
"Epoch 33/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 1950277.5000 - val_loss: 143.1821\n",
"23/23 [==============================] - 1s 32ms/step - loss: 3172797.7500 - val_loss: 98.5941\n",
"Epoch 34/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 288345.4375 - val_loss: 111.6001\n",
"23/23 [==============================] - 1s 34ms/step - loss: 423511.1875 - val_loss: 104.0902\n",
"Epoch 35/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 15917235.0000 - val_loss: 93.8450\n",
"23/23 [==============================] - 1s 35ms/step - loss: 126338.9531 - val_loss: 102.0765\n",
"Epoch 36/40\n",
"23/23 [==============================] - 1s 33ms/step - loss: 6554179.5000 - val_loss: 89.9294\n",
"23/23 [==============================] - 1s 34ms/step - loss: 520707.8750 - val_loss: 101.0857\n",
"Epoch 37/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 916703.4375 - val_loss: 86.5181\n",
"23/23 [==============================] - 1s 34ms/step - loss: 1672894.8750 - val_loss: 100.0803\n",
"Epoch 38/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 1636418.8750 - val_loss: 81.6224\n",
"23/23 [==============================] - 1s 34ms/step - loss: 3788303.0000 - val_loss: 100.6476\n",
"Epoch 39/40\n",
"23/23 [==============================] - 1s 34ms/step - loss: 3871.7839 - val_loss: 72.8129\n",
"23/23 [==============================] - 1s 35ms/step - loss: 2577893.5000 - val_loss: 98.6984\n",
"Epoch 40/40\n",
"23/23 [==============================] - 1s 35ms/step - loss: 740887.7500 - val_loss: 73.8067\n",
"23/23 [==============================] - 1s 34ms/step - loss: 1781253.5000 - val_loss: 97.7365\n",
"[INFO] predicting t-score...\n",
"[INFO] average t-score: -1.47828, std t-score: 1.406043864383835\n",
"[INFO] mean absolute percentage error: 73.81%, std: 54.82%\n"
"[INFO] mean absolute percentage error: 97.74%, std: 18.17%\n"
]
}
],
......
model_keras.dexa/saved_model.pb
View file @ 05c9a892
No preview for this file type
model_keras.dexa/variables/variables.data-00000-of-00001
View file @ 05c9a892
No preview for this file type
model_keras.dexa/variables/variables.index
View file @ 05c9a892
No preview for this file type
model_weights.dexa.data-00000-of-00001
View file @ 05c9a892
No preview for this file type
model_weights.dexa.index
View file @ 05c9a892
No preview for this file type
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment