225 lines (224 with data), 14.5 kB

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "from options.test_options import TestOptions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+ RESULTS_DIR=./results/sampledata\n",
      "+ CARSON_PATH=./pretrained_models/carson_Jan2021.h5\n",
      "+ CARMEN_PATH=./pretrained_models/carmen_Jan2021.h5\n",
      "+ CLASS=edges2shoes\n",
      "+ FORMAT=NIFTI\n",
      "+ GPU_ID=0\n",
      "+ NUM_TEST=10\n",
      "+ NUM_SAMPLES=10\n",
      "+ CUDA_VISIBLE_DEVICES=0\n",
      "+ python ./test.py --dataformat NIFTI --dataroot ./datasets/edges2shoes --results_dir ./results/sampledata\n",
      "2021-02-02 02:29:35.232225: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library libcudart.so.10.1\n",
      "NIFTI\n"
     ]
    }
   ],
   "source": [
    "!bash ./scripts/test_CarSON.sh"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from models import deep_strain\n",
    "from data import base_dataset\n",
    "from data import nifti_dataset\n",
    "from options import base_options"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "class BaseOptions():\n",
    "    \n",
    "    def __init__(self):\n",
    "        self.name = 'for_test_only'\n",
    "        self.dataroot = '../sample_data/'\n",
    "        self.max_dataset_size = float(\"inf\")\n",
    "        self.preprocess = 'reshape_to_carson_crop_zscore'\n",
    "        self.image_shape  = (128,128,1)\n",
    "        self.volume_shape = (128,128,16,1)\n",
    "        self.criterion_netS = (128,128,16,1)\n",
    "        self.pretrained_models_netS  = './pretrained_models/carson_Jan2021.h5'\n",
    "        self.pretrained_models_netME = './pretrained_models/carmen_Jan2021.h5'\n",
    "        self.nlabels = 4\n",
    "        self.netS_lr = 5e-4\n",
    "        \n",
    "        \n",
    "        self.lambda_i = 0.01\n",
    "        self.lambda_a = 0.5\n",
    "        self.lambda_s = 0.1\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "opt = BaseOptions()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = nifti_dataset.NiftiDataset(opt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.optimizers import Adam"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = dataset.__loaddata__(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "ds = deep_strain.DeepStrain(Adam, opt)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "netS = ds.test_segmentations()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = netS.predict(x, batch_size=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = np.argmax(y,-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f93e532f4a8>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(y[10])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pylab as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}