From a65d74dd23c12705f07db7b04a9b2975d128bab0 Mon Sep 17 00:00:00 2001 From: Panos Achlioptas Date: Tue, 14 Aug 2018 16:58:24 -0700 Subject: [PATCH] updating with GAN-training notebooks --- notebooks/train_latent_gan.ipynb | 857 ++++++-------------------- notebooks/train_raw_gan.ipynb | 343 +++++++++++ notebooks/train_single_class_ae.ipynb | 311 +++------- src/autoencoder.py | 58 +- src/general_utils.py | 40 ++ src/generators_discriminators.py | 21 +- src/neural_net.py | 14 + src/vanilla_gan.py | 6 +- src/w_gan_gp.py | 2 +- 9 files changed, 695 insertions(+), 957 deletions(-) create mode 100755 notebooks/train_raw_gan.ipynb diff --git a/notebooks/train_latent_gan.ipynb b/notebooks/train_latent_gan.ipynb index 99e9b0a..e4c76a1 100755 --- a/notebooks/train_latent_gan.ipynb +++ b/notebooks/train_latent_gan.ipynb @@ -11,7 +11,7 @@ }, { "cell_type": "code", - "execution_count": 2, + "execution_count": 1, "metadata": { "collapsed": false, "deletable": true, @@ -21,25 +21,27 @@ "source": [ "import numpy as np\n", "import os.path as osp\n", + "import matplotlib.pylab as plt\n", "\n", - "from latent_3d_points.src.autoencoder import Configuration as Conf\n", "from latent_3d_points.src.point_net_ae import PointNetAutoEncoder\n", + "from latent_3d_points.src.autoencoder import Configuration as Conf\n", + "from latent_3d_points.src.neural_net import MODEL_SAVER_ID\n", "\n", "from latent_3d_points.src.in_out import snc_category_to_synth_id, create_dir, PointCloudDataSet, \\\n", " load_all_point_clouds_under_folder\n", "\n", + "from latent_3d_points.src.general_utils import plot_3d_point_cloud\n", "from latent_3d_points.src.tf_utils import reset_tf_graph\n", "\n", "from latent_3d_points.src.vanilla_gan import Vanilla_GAN\n", "from latent_3d_points.src.w_gan_gp import W_GAN_GP\n", - "\n", "from latent_3d_points.src.generators_discriminators import latent_code_discriminator_two_layers,\\\n", "latent_code_generator_two_layers" ] }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 2, "metadata": { "collapsed": false, "deletable": true, @@ -52,9 +54,30 @@ "%matplotlib inline" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Specify where the raw point-clouds and the pre-trained AE are." + ] + }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 9, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "# Top-dir of where point-clouds are stored.\n", + "top_in_dir = '../data/shape_net_core_uniform_samples_2048/' \n", + "\n", + "ae_configuration = '../data/single_class_ae/configuration'" + ] + }, + { + "cell_type": "code", + "execution_count": 3, "metadata": { "collapsed": false }, @@ -68,62 +91,20 @@ } ], "source": [ - "# Set DATA/AE parameters\n", + "# Where to save GANs check-points etc.\n", + "top_out_dir = '../data/'\n", + "experiment_name = 'latent_gan_with_chamfer_ae'\n", "\n", - "top_out_dir = '../data/' # Use to save Neural-Net check-points etc.\n", - "top_in_dir = '../data/shape_net_core_uniform_samples_2048/' # Top-dir of where point-clouds are stored.\n", - "\n", - "ae_configuration = '../data/single_class_ae/configuration' # AE model-description. You can alternatively, \n", - " # use your own way to load a pre-trained AE.\n", - "\n", - "ae_epoch = 100 # Model/epoch of AE to load.\n", - "bneck_size = 128 # Bottleneck-AE size\n", - "\n", - "experiment_name = 'latent_gan'\n", - "n_pc_points = 2048 # Number of points per model.\n", + "ae_epoch = 500 # Epoch of AE to load.\n", + "bneck_size = 128 # Bottleneck-size of the AE\n", + "n_pc_points = 2048 # Number of points per model.\n", "\n", "class_name = raw_input('Give me the class name (e.g. \"chair\"): ').lower()" ] }, { "cell_type": "code", - "execution_count": null, - "metadata": { - "collapsed": true - }, - "outputs": [], - "source": [ - "# Set GAN, training parameters.\n", - "\n", - "# save_model = False\n", - "# class_name = raw_input('Give me the class type.\\n').lower()\n", - "# syn_id = snc_category_to_synth_id()[class_name]\n", - "# ae_loss = 'emd'\n", - "use_wgan = False\n", - "max_epochs = 500\n", - "plot_train_curve = True\n", - "saver_step = np.hstack([np.array([1, 5, 10]), np.arange(50, max_epochs+1, 50)])\n", - "save_synthetic_samples = True\n", - "experiment_tag = 'test'\n", - "\n", - "init_lr = 0.0001\n", - "batch_size = 50\n", - "noise_params = {'mu':0, 'sigma': 0.2}\n", - "noise_dim = bneck_size\n", - "beta = 0.5\n", - "n_syn_samples = train_data.num_examples # How many samples to produce in each save step.\n", - "n_out = [bneck_size]\n", - "accum_syn_data = []\n", - "train_stats = []\n", - "\n", - "if save_synthetic_samples:\n", - " synthetic_data_out_dir = osp.join(top_out_dir, 'OUT/synthetic_samples/', experiment_tag)\n", - " create_dir(synthetic_data_out_dir)" - ] - }, - { - "cell_type": "code", - "execution_count": 5, + "execution_count": 6, "metadata": { "collapsed": false }, @@ -132,20 +113,22 @@ "name": "stdout", "output_type": "stream", "text": [ - "6778 pclouds were loaded. They belong in 1 shape-classes.\n" + "6778 pclouds were loaded. They belong in 1 shape-classes.\n", + "Shape of DATA = (6778, 2048, 3)\n" ] } ], "source": [ - "# Load point-clouds\n", + "# Load point-clouds.\n", "syn_id = snc_category_to_synth_id()[class_name]\n", "class_dir = osp.join(top_in_dir , syn_id)\n", - "all_pc_data = load_all_point_clouds_under_folder(class_dir, n_threads=8, file_ending='.ply', verbose=True)" + "all_pc_data = load_all_point_clouds_under_folder(class_dir, n_threads=8, file_ending='.ply', verbose=True)\n", + "print 'Shape of DATA =', all_pc_data.point_clouds.shape" ] }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 10, "metadata": { "collapsed": false }, @@ -154,40 +137,8 @@ "name": "stdout", "output_type": "stream", "text": [ - "Building Encoder\n", - "encoder_conv_layer_0 conv params = 256 bnorm params = 128\n", - "Tensor(\"single_class_ae_2/Relu:0\", shape=(?, 2048, 64), dtype=float32)\n", - "output size: 131072 \n", - "\n", - "encoder_conv_layer_1 conv params = 8320 bnorm params = 256\n", - "Tensor(\"single_class_ae_2/Relu_1:0\", shape=(?, 2048, 128), dtype=float32)\n", - "output size: 262144 \n", - "\n", - "encoder_conv_layer_2 conv params = 16512 bnorm params = 256\n", - "Tensor(\"single_class_ae_2/Relu_2:0\", shape=(?, 2048, 128), dtype=float32)\n", - "output size: 262144 \n", - "\n", - "encoder_conv_layer_3 conv params = 33024 bnorm params = 512\n", - "Tensor(\"single_class_ae_2/Relu_3:0\", shape=(?, 2048, 256), dtype=float32)\n", - "output size: 524288 \n", - "\n", - "encoder_conv_layer_4 conv params = 32896 bnorm params = 256\n", - "Tensor(\"single_class_ae_2/Relu_4:0\", shape=(?, 2048, 128), dtype=float32)\n", - "output size: 262144 \n", - "\n", - "Tensor(\"single_class_ae_2/Max:0\", shape=(?, 128), dtype=float32)\n", - "Building Decoder\n", - "decoder_fc_0 FC params = 33024 Tensor(\"single_class_ae_2/Relu_5:0\", shape=(?, 256), dtype=float32)\n", - "output size: 256 \n", - "\n", - "decoder_fc_1 FC params = 65792 Tensor(\"single_class_ae_2/Relu_6:0\", shape=(?, 256), dtype=float32)\n", - "output size: 256 \n", - "\n", - "decoder_fc_2 FC params = 1579008 Tensor(\"single_class_ae_2/decoder_fc_2/BiasAdd:0\", shape=(?, 6144), dtype=float32)\n", - "output size: 6144 \n", - "\n", - "INFO:tensorflow:Restoring parameters from ../data/single_class_ae/models.ckpt-100\n", - "Model restored in epoch 100.\n" + "INFO:tensorflow:Restoring parameters from ../data/single_class_ae/models.ckpt-500\n", + "Model restored in epoch 500.\n" ] } ], @@ -195,35 +146,114 @@ "# Load pre-trained AE\n", "reset_tf_graph()\n", "ae_conf = Conf.load(ae_configuration)\n", + "ae_conf.encoder_args['verbose'] = False\n", + "ae_conf.decoder_args['verbose'] = False\n", "ae = PointNetAutoEncoder(ae_conf.experiment_name, ae_conf)\n", "ae.restore_model(ae_conf.train_dir, ae_epoch, verbose=True)" ] }, { "cell_type": "code", - "execution_count": 19, + "execution_count": 11, "metadata": { "collapsed": false }, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Shape of DATA = (6778, 128)\n" + ] + } + ], "source": [ - "# Convert raw-data to latent codes.\n", + "# Use AE to convert raw pointclouds to latent codes.\n", "latent_codes = ae.get_latent_codes(all_pc_data.point_clouds)\n", - "train_data = PointCloudDataSet(latent_codes)" + "latent_data = PointCloudDataSet(latent_codes)\n", + "print 'Shape of DATA =', latent_data.point_clouds.shape" ] }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 12, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAASwAAAElCAYAAABect+9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nLBYWlqKBFSwWFpaigRUsFhaWouH/AxHRsevNNy+9AAAAAElFTkSuQmCC\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAASwAAAElCAYAAABect+9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCgZWsFhYWAqG/x9QU4KQVPB+nAAAAABJRU5ErkJggg==\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# Check the decoded AE latent-codes look descent.\n", + "L = ae.decode(latent_codes)\n", + "i = 0\n", + "plot_3d_point_cloud(L[i][:, 0], L[i][:, 1], L[i][:, 2], in_u_sphere=True);\n", + "i = 20\n", + "plot_3d_point_cloud(L[i][:, 0], L[i][:, 1], L[i][:, 2], in_u_sphere=True);" + ] + }, + { + "cell_type": "code", + "execution_count": 13, "metadata": { "collapsed": false }, "outputs": [], - "source": [] + "source": [ + "# Set GAN parameters.\n", + "\n", + "use_wgan = True # Wasserstein with gradient penalty, or not?\n", + "n_epochs = 1 # Epochs to train.\n", + "\n", + "plot_train_curve = True\n", + "save_gan_model = False\n", + "saver_step = np.hstack([np.array([1, 5, 10]), np.arange(50, n_epochs + 1, 50)])\n", + "\n", + "# If true, every 'saver_step' epochs we produce & save synthetic pointclouds.\n", + "save_synthetic_samples = True\n", + "# How many synthetic samples to produce at each save step.\n", + "n_syn_samples = latent_data.num_examples\n", + "\n", + "# Optimization parameters\n", + "init_lr = 0.0001\n", + "batch_size = 50\n", + "noise_params = {'mu':0, 'sigma': 0.2}\n", + "noise_dim = bneck_size\n", + "beta = 0.5 # ADAM's momentum.\n", + "\n", + "n_out = [bneck_size] # Dimensionality of generated samples.\n", + "\n", + "if save_synthetic_samples:\n", + " synthetic_data_out_dir = osp.join(top_out_dir, 'OUT/synthetic_samples/', experiment_name)\n", + " create_dir(synthetic_data_out_dir)\n", + "\n", + "if save_gan_model:\n", + " train_dir = osp.join(top_out_dir, 'OUT/latent_gan', experiment_name)\n", + " create_dir(train_dir)" + ] }, { "cell_type": "code", - "execution_count": 19, + "execution_count": 14, "metadata": { "collapsed": false }, @@ -232,22 +262,32 @@ "reset_tf_graph()\n", "\n", "if use_wgan:\n", - " lam = 10\n", - " gan = W_GAN_GP(experiment_tag, init_lr, lam, n_out, noise_dim, \\\n", + " lam = 10 # lambda of W-GAN-GP\n", + " gan = W_GAN_GP(experiment_name, init_lr, lam, n_out, noise_dim, \\\n", " latent_code_discriminator_two_layers, \n", " latent_code_generator_two_layers,\\\n", - " beta=beta\n", - " )\n", + " beta=beta)\n", "else: \n", - " gan = Vanilla_GAN(experiment_tag, init_lr, n_out, noise_dim,\n", + " gan = Vanilla_GAN(experiment_name, init_lr, n_out, noise_dim,\n", " latent_code_discriminator_two_layers, latent_code_generator_two_layers,\n", - " beta=beta\n", - " )" + " beta=beta)" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [ + "accum_syn_data = []\n", + "train_stats = []" ] }, { "cell_type": "code", - "execution_count": 19, + "execution_count": 16, "metadata": { "collapsed": false, "deletable": true, @@ -258,517 +298,48 @@ "name": "stdout", "output_type": "stream", "text": [ - "1 (0.41234670566475912, 1.9826304430546968)\n", - "2 (0.092744595168725311, 3.7401423350624414)\n", - "3 (0.086751907124467523, 4.9079365854677945)\n", - "4 (0.14127983541592307, 5.5437571193860924)\n", - "5 (0.11072289509617764, 6.0356144200200621)\n", - "6 (0.057010742665632913, 6.0249628647514015)\n", - "7 (0.034189508660979896, 6.2511722398840863)\n", - "8 (0.02702255496836227, 6.6130070810732633)\n", - "9 (0.028049864069275236, 6.6683759316154152)\n", - "10 (0.039642552173008092, 6.2399310899817424)\n", - "11 (0.067302263495714762, 5.3281164915665338)\n", - "12 (0.095872920589602506, 4.6584717170051908)\n", - "13 (0.11670367102260175, 4.179448787025783)\n", - "14 (0.13206631431113119, 3.9878793301789659)\n", - "15 (0.16565291175375815, 3.8262052411618441)\n", - "16 (0.2338887218547904, 3.6161000645678976)\n", - "17 (0.31642804664114249, 3.3541466132454252)\n", - "18 (0.4007429537565812, 3.000937961495441)\n", - "19 (0.50492612304894824, 2.6533821873042895)\n", - "20 (0.60389162457507595, 2.2986403237218442)\n", - "21 (0.69011772767357205, 2.038876848635466)\n", - "22 (0.76728998733603437, 1.8087486132331516)\n", - "23 (0.84428684193155035, 1.6176354708878891)\n", - "24 (0.92259861023529721, 1.4711052713186845)\n", - "25 (0.98413353536439974, 1.3383844899094623)\n", - "26 (1.058455739332282, 1.2266646571781323)\n", - "27 (1.0933632700339608, 1.1370870299961255)\n", - "28 (1.1494868034901826, 1.0696246064227559)\n", - "29 (1.1742720966753752, 1.0035977975181911)\n", - "30 (1.1838847149973331, 0.96809209015058439)\n", - "31 (1.1922432749167733, 0.93998518819394317)\n", - "32 (1.1877336014872011, 0.92872822388358733)\n", - "33 (1.1868084233740102, 0.91059028687684429)\n", - "34 (1.1850983339807262, 0.89915602621824842)\n", - "35 (1.2114195180975873, 0.89582783813061917)\n", - "36 (1.2173548066097757, 0.88707890044087945)\n", - "37 (1.2133679514345916, 0.87999030507129172)\n", - "38 (1.198970634004344, 0.89574599628863127)\n", - "39 (1.2071005385854969, 0.88056844680205637)\n", - "40 (1.2186715789463209, 0.86747113051621805)\n", - "41 (1.2213859060536261, 0.86541393528813904)\n", - "42 (1.2314226160878721, 0.85650893190632693)\n", - "43 (1.2326037386189337, 0.85444202060284824)\n", - "44 (1.2325126844903698, 0.85386169008586721)\n", - "45 (1.2365720323894336, 0.84808804626050205)\n", - "46 (1.2394928206568179, 0.84139824224554971)\n", - "47 (1.2442763452944547, 0.83001712664313942)\n", - "48 (1.2364845638689788, 0.8281611535860145)\n", - "49 (1.2246730410534403, 0.83234623463257496)\n", - "50 (1.2170278870541116, 0.83547034885572347)\n", - "51 (1.2177783084952314, 0.83441622049912167)\n", - "52 (1.2080399171165799, 0.83851466178894041)\n", - "53 (1.2104164289391559, 0.8362542484117591)\n", - "54 (1.2086933135986329, 0.84081916446271154)\n", - "55 (1.2005858390227608, 0.84906800207884414)\n", - "56 (1.1888435229011205, 0.85882545554119605)\n", - "57 (1.1837518443231998, 0.8666842486547387)\n", - "58 (1.1763970292132833, 0.87175706935965491)\n", - "59 (1.1706018105797145, 0.88083527295485786)\n", - "60 (1.1620517865471218, 0.89237378732017847)\n", - "61 (1.1564908359361732, 0.89998021955075469)\n", - "62 (1.1506333662116008, 0.90722646557766462)\n", - "63 (1.1477497536203136, 0.91215687212736707)\n", - "64 (1.1434479029282281, 0.91675022322198618)\n", - "65 (1.1312506592792013, 0.93126815505649729)\n", - "66 (1.1285485609717991, 0.94138839193012402)\n", - "67 (1.1246995044791179, 0.94805420999941614)\n", - "68 (1.1245239506597104, 0.95029682957607764)\n", - "69 (1.1173864260963771, 0.96180713176727295)\n", - "70 (1.1130206429440042, 0.96728296901868738)\n", - "71 (1.1074646234512329, 0.97435969528944599)\n", - "72 (1.1116392156352168, 0.98440828167873884)\n", - "73 (1.1043307490970777, 0.9885701920675195)\n", - "74 (1.1059170401614644, 0.98864997988161829)\n", - "75 (1.1091293739235919, 0.98809874990712043)\n", - "76 (1.0934065694394319, 1.0077143301134524)\n", - "77 (1.0928748576537421, 1.0033487444338591)\n", - "78 (1.1046450148458067, 1.007246548714845)\n", - "79 (1.1028323339379351, 1.0021378159523011)\n", - "80 (1.1005484601725704, 1.0051725460135419)\n", - "81 (1.0959360843119414, 1.0098319027734839)\n", - "82 (1.0976150657819665, 1.0204707446305648)\n", - "83 (1.0849669166233229, 1.0259531207706618)\n", - "84 (1.0864333634791168, 1.0258639475573663)\n", - "85 (1.0888100401214931, 1.0318965367648913)\n", - "86 (1.0972434297851894, 1.0288777921510779)\n", - "87 (1.0880611269370368, 1.0355861026307811)\n", - "88 (1.0888737569684568, 1.027182708615842)\n", - "89 (1.0991977137068043, 1.0300792170607525)\n", - "90 (1.083988266405852, 1.0367559883905493)\n", - "91 (1.0820493200550909, 1.0490973368934964)\n", - "92 (1.0886567737745203, 1.044057239656863)\n", - "93 (1.0856366976447727, 1.0433787812357362)\n", - "94 (1.0845431436663089, 1.053264059709466)\n", - "95 (1.085616317520971, 1.0447780692059061)\n", - "96 (1.0936182239781256, 1.0452802352283312)\n", - "97 (1.0946472955786664, 1.0533039725345115)\n", - "98 (1.0863736961198889, 1.0424712616464367)\n", - "99 (1.0931022840997446, 1.0425160998883454)\n", - "100 (1.0822759167007778, 1.0619962246521659)\n", - "101 (1.0924004026081251, 1.0445017130478569)\n", - "102 (1.0889965181765349, 1.0494524458180303)\n", - "103 (1.0878925898800726, 1.0601960208105004)\n", - "104 (1.0923226480898649, 1.0554197586101035)\n", - "105 (1.0795898655186529, 1.0619218587875365)\n", - "106 (1.0880165680595066, 1.0662745615710383)\n", - "107 (1.08158983717794, 1.0582538692847543)\n", - "108 (1.0933125335237255, 1.0540366691091787)\n", - "109 (1.0810984834380772, 1.0612267006998477)\n", - "110 (1.083308660465738, 1.0659396435903465)\n", - "111 (1.0922857160153596, 1.0592926294907279)\n", - "112 (1.0834299631740736, 1.068205714225769)\n", - "113 (1.0790481204571931, 1.0633363982905513)\n", - "114 (1.0994532839111659, 1.0618501797966335)\n", - "115 (1.0808369527692381, 1.0715536060540571)\n", - "116 (1.0907261698142341, 1.0695569644803586)\n", - "117 (1.0898898549701856, 1.0635280567666758)\n", - "118 (1.0775508585183518, 1.0656558990478515)\n", - "119 (1.0828532809796541, 1.0680917066076527)\n", - "120 (1.0944272564805073, 1.0573546658391537)\n", - "121 (1.0994597388350444, 1.0599324096804079)\n", - "122 (1.0833238612050595, 1.0579029083251954)\n", - "123 (1.093057918548584, 1.0559155987656634)\n", - "124 (1.0980987849442856, 1.0584892739420351)\n", - "125 (1.0797851536584937, 1.067490825964057)\n", - "126 (1.104499105785204, 1.0691997662834498)\n", - "127 (1.0869002430335335, 1.0554412961006165)\n", - "128 (1.082266310505245, 1.0587739975556083)\n", - "129 (1.0944828515467437, 1.0637096809304278)\n", - "130 (1.0919911161712978, 1.0740552513495736)\n", - "131 (1.1006181374840114, 1.0687074790830198)\n", - "132 (1.1055420269136844, 1.0614448106807211)\n", - "133 (1.0858643365942913, 1.0582544632579969)\n", - "134 (1.1010037157846533, 1.0613328047420667)\n", - "135 (1.0843141260354414, 1.0631380490634752)\n", - "136 (1.0828684951948082, 1.0674977053766666)\n", - "137 (1.0911453615064206, 1.0533388137817383)\n", - "138 (1.101715824396714, 1.0603238909140877)\n", - "139 (1.0921493302220884, 1.0620634488437486)\n", - "140 (1.0983918065610139, 1.0671820391779361)\n", - "141 (1.1014487618985382, 1.0681644268657851)\n", - "142 (1.0858702462652454, 1.0562026220819225)\n", - "143 (1.1029990108116814, 1.0618750598119653)\n", - "144 (1.0909447530041569, 1.0529683216758396)\n", - "145 (1.1244222444036733, 1.0721162314000336)\n", - "146 (1.0970736482869023, 1.0498763540516729)\n", - "147 (1.0940958795340165, 1.0602823433668718)\n", - "148 (1.0947580923204836, 1.0655735259470733)\n", - "149 (1.0922399500141973, 1.0624245436295219)\n", - "150 (1.0968767347543136, 1.0592749300210371)\n", - "151 (1.0979481800742772, 1.0549658510995947)\n", - "152 (1.0937964734823808, 1.0641959221466728)\n", - "153 (1.1025822167811186, 1.0675089737643366)\n", - "154 (1.0921534512353981, 1.0497446661410124)\n", - "155 (1.1071245685867641, 1.0697445449621781)\n", - "156 (1.1027637087780497, 1.0734816323155942)\n", - "157 (1.1014230961384981, 1.0805889482083528)\n", - "158 (1.0977142567219942, 1.0630795178206072)\n", - "159 (1.0854606079018634, 1.0735903657000998)\n", - "160 (1.1181007965751317, 1.0664543991503508)\n", - "161 (1.1004351211630781, 1.0702304041903952)\n", - "162 (1.1038450463958409, 1.0641243343767912)\n", - "163 (1.0902274121408877, 1.0623204236445221)\n", - "164 (1.090473440937374, 1.0758818590122721)\n", - "165 (1.0873744264892911, 1.0622744995614757)\n", - "166 (1.0931511827137159, 1.0634818248126818)\n", - "167 (1.1438659491746321, 1.0817479232083196)\n", - "168 (1.0957175731658935, 1.0621172148248423)\n", - "169 (1.0884431906368421, 1.0597897716190503)\n", - "170 (1.1040473642556563, 1.0609183119690937)\n", - "171 (1.0830436592516692, 1.0606860212657763)\n", - "172 (1.0832457671994749, 1.07373358944188)\n", - "173 (1.0993376648944357, 1.0695622537447058)\n", - "174 (1.0934916636218195, 1.0811071758684905)\n", - "175 (1.0924969616143601, 1.0562890109808549)\n", - "176 (1.0988444831060327, 1.0848697439483974)\n", - "177 (1.0878931937010392, 1.0615405404049418)\n", - "178 (1.0808032891024715, 1.0593567361002383)\n", - "179 (1.1109555244445801, 1.0814189320025236)\n", - "180 (1.0934983725133149, 1.0725876507551775)\n", - "181 (1.088009848801986, 1.0704948300900667)\n", - "182 (1.093365170126376, 1.0604619176491448)\n", - "183 (1.0872117778529291, 1.0751653381015944)\n", - "184 (1.1001784127691518, 1.0776064147119937)\n", - "185 (1.0969955547996189, 1.079428805993951)\n", - "186 (1.0971569755802983, 1.0752022763957148)\n", - "187 (1.0846536029940066, 1.0825046321620111)\n", - "188 (1.1004284744677337, 1.0807601643645246)\n", - "189 (1.099483160868935, 1.0819614752479221)\n", - "190 (1.0945242337558581, 1.0819563269615173)\n", - "191 (1.0781862699467204, 1.0710381471592447)\n", - "192 (1.080699430341306, 1.0719538025234057)\n", - "193 (1.079686543215876, 1.0864965547686039)\n", - "194 (1.0756645218185756, 1.0827673803205076)\n", - "195 (1.085701260359391, 1.0545230917308641)\n", - "196 (1.0849277377128601, 1.0761291156644406)\n", - "197 (1.077771966872008, 1.0688308124956878)\n", - "198 (1.1290029085200766, 1.1071248671282892)\n", - "199 (1.0791631226954252, 1.0808049233063408)\n", - "200 (1.0844230651855469, 1.0833020630090133)\n", - "201 (1.1045408938242041, 1.0964778822401295)\n", - "202 (1.0831859008125637, 1.0848259832548059)\n", - "203 (1.0771517484084419, 1.0847623446713324)\n", - "204 (1.078889012336731, 1.0897384752397952)\n", - "205 (1.0986618430718131, 1.0783246097357377)\n", - "206 (1.0935321818227353, 1.0884758646073549)\n", - "207 (1.0827571760053221, 1.0998788237571717)\n", - "208 (1.0781103927156199, 1.0874165638633397)\n", - "209 (1.0809282390967658, 1.0898178831390712)\n", - "210 (1.0855537108753039, 1.0791064257207124)\n", - "211 (1.0937402922174204, 1.0940179171769515)\n", - "212 (1.0836133905079053, 1.0913811289745829)\n", - "213 (1.0971437231354091, 1.0884505235630533)\n", - "214 (1.0851517573646876, 1.1084770529166512)\n", - "215 (1.0722814746524976, 1.075094178966854)\n", - "216 (1.0803877877152483, 1.087859873149706)\n", - "217 (1.0901770161545794, 1.090301841756572)\n", - "218 (1.076680382438328, 1.0817292907963629)\n", - "219 (1.0720723261003908, 1.1070210695266725)\n", - "220 (1.09796716493109, 1.1114240247270335)\n", - "221 (1.0769786995390187, 1.0816185904585798)\n", - "222 (1.0736316546149876, 1.0858277844346087)\n", - "223 (1.0654935225196507, 1.0892683630404265)\n", - "224 (1.0872417076774266, 1.1247962378937266)\n", - "225 (1.0715149646219999, 1.1033577436986177)\n", - "226 (1.0777521817580513, 1.098313477764959)\n", - "227 (1.0720502516497736, 1.1023028622502866)\n", - "228 (1.0798908845238064, 1.1105305148207623)\n", - "229 (1.0859618233597796, 1.0873072753781858)\n", - "230 (1.078237163502237, 1.1009573184925576)\n", - "231 (1.0685710544171541, 1.1003237729487212)\n", - "232 (1.0702748531880586, 1.0838427559189174)\n", - "233 (1.0936279364254164, 1.1186339896658193)\n", - "234 (1.0789416468661763, 1.104107473725858)\n", - "235 (1.0829502686210302, 1.0935399755187656)\n", - "236 (1.0843820763670879, 1.1096118429432744)\n", - "237 (1.0756791068160017, 1.1043383225150731)\n", - "238 (1.0896834674088851, 1.1053425555643828)\n", - "239 (1.0797132958536562, 1.1008561823679053)\n", - "240 (1.0796409238939699, 1.0945076553717903)\n", - "241 (1.0879441997279291, 1.0973274655964063)\n", - "242 (1.0782027742137079, 1.0737029288126074)\n", - "243 (1.0836969116459723, 1.1062331147815869)\n", - "244 (1.0792640867440597, 1.0976996437363002)\n", - "245 (1.0704491848530977, 1.0995744041774584)\n", - "246 (1.077424929452979, 1.0980342512545378)\n", - "247 (1.0792140126228333, 1.1057461769684502)\n", - "248 (1.0818266334741011, 1.1042159567708554)\n", - "249 (1.0789558514304782, 1.0922253774560016)\n", - "250 (1.0821485436480978, 1.1078594777895057)\n", - "251 (1.0986064273378124, 1.1071850292060685)\n", - "252 (1.0764238103576329, 1.0836266517639159)\n", - "253 (1.0710044829741767, 1.0940350827963456)\n", - "254 (1.0762397071589593, 1.0976879524148029)\n", - "255 (1.0963872722957446, 1.1063155734020731)\n", - "256 (1.0799085741457732, 1.0816329893858536)\n", - "257 (1.1018067981885826, 1.1150849798451299)\n", - "258 (1.0709388453027477, 1.0883600924326027)\n", - "259 (1.0774241509644882, 1.10421797192615)\n", - "260 (1.0793337713117186, 1.0913003356560418)\n", - "261 (1.0783874299215235, 1.1238983278689176)\n", - "262 (1.0877888280412424, 1.1200138677721438)\n", - "263 (1.0881402238555578, 1.1012280775153118)\n", - "264 (1.0862244357233461, 1.0869347593058711)\n", - "265 (1.089232543758724, 1.1024264662162118)\n", - "266 (1.0780294553093288, 1.0950782516728277)\n", - "267 (1.0767166982526364, 1.0954291685767796)\n", - "268 (1.0658336976300116, 1.1005429464837779)\n", - "269 (1.0879795639411263, 1.1013139657352282)\n", - "270 (1.0839057797970979, 1.0975371329680732)\n", - "271 (1.0844539383183356, 1.1087700615758482)\n", - "272 (1.0684055561604706, 1.1049769515576571)\n", - "273 (1.0763044331384741, 1.0983816784361133)\n", - "274 (1.0924348509829978, 1.1001919233280679)\n", - "275 (1.0666166061940401, 1.0843881586323614)\n", - "276 (1.0764738601187001, 1.1012485151705536)\n", - "277 (1.0711921883665998, 1.1201193871705428)\n", - "278 (1.0726214626561041, 1.1154373168945313)\n", - "279 (1.0627256580021069, 1.0829286907030189)\n", - "280 (1.0759893204854882, 1.113433745633001)\n", - "281 (1.0823959298755812, 1.1091961591140083)\n", - "282 (1.0770195028056269, 1.0892869581346927)\n", - "283 (1.0700465134952379, 1.1061340658561043)\n", - "284 (1.0748619976251022, 1.1179745083269865)\n", - "285 (1.0672801536062488, 1.1097009451492974)\n", - "286 (1.0738005746965822, 1.1055828996326613)\n", - "287 (1.0766839944798015, 1.1073296230772267)\n", - "288 (1.0754690045895783, 1.1103822407515154)\n", - "289 (1.0608378384424293, 1.1068059377048327)\n", - "290 (1.073976299037104, 1.1098686368569084)\n", - "291 (1.0685776358065397, 1.1057679036389227)\n", - "292 (1.0650276588357013, 1.1117810529211294)\n", - "293 (1.0661615537560505, 1.1185731276221897)\n", - "294 (1.0623828602873762, 1.1236476271048836)\n", - "295 (1.0631220952324245, 1.1128026568371316)\n", - "296 (1.0716917442238849, 1.1114328907883686)\n", - "297 (1.0767520583194234, 1.1192943469337795)\n", - "298 (1.064658705047939, 1.1129804212114085)\n", - "299 (1.0580332009688667, 1.1027672145677649)\n", - "300 (1.0616067368051281, 1.1225489035896632)\n", - "301 (1.0614260481751483, 1.1288470200870349)\n", - "302 (1.056058941198432, 1.1326043232627536)\n", - "303 (1.0605222448058751, 1.1212668600289717)\n", - "304 (1.0540918422781902, 1.1296033470526985)\n", - "305 (1.0546653405479762, 1.1317274627478227)\n", - "306 (1.05226672835972, 1.1332802388979042)\n", - "307 (1.0495982740236365, 1.1258043957793193)\n", - "308 (1.0578014270119045, 1.1260743628377501)\n", - "309 (1.0528204005697499, 1.1232508519421454)\n", - "310 (1.0581569847853287, 1.1221160738364511)\n", - "311 (1.0585332600966744, 1.1217583013617474)\n", - "312 (1.0564358804536902, 1.122343029146609)\n", - "313 (1.0506887467011161, 1.1187775824380957)\n", - "314 (1.0546258496201557, 1.1469619999761167)\n", - "315 (1.0984885360883629, 1.1326617790305096)\n", - "316 (1.0433142309603485, 1.1427765981010769)\n", - "317 (1.0654204373774321, 1.1524633936260058)\n", - "318 (1.0482395348341569, 1.1333403649537459)\n", - "319 (1.0470010384269384, 1.1271858365639396)\n", - "320 (1.0396336959755939, 1.1317492085954417)\n", - "321 (1.0459459413652834, 1.1385792478271153)\n", - "322 (1.0519470790158147, 1.1399862320526786)\n", - "323 (1.0572510258011196, 1.1393570656361787)\n", - "324 (1.044840739602628, 1.1348146008408588)\n", - "325 (1.0527639425319173, 1.141965171046879)\n", - "326 (1.0453737901604694, 1.1313047300214354)\n", - "327 (1.0420962069345556, 1.1686837134153947)\n", - "328 (1.0390396009320799, 1.1294684461925342)\n", - "329 (1.0425608178843622, 1.1443643855011982)\n", - "330 (1.0471658431965372, 1.1331670693729234)\n", - "331 (1.0355068268983261, 1.1349258516145788)\n", - "332 (1.0409401701844256, 1.153739849899126)\n", - "333 (1.038916750057884, 1.1571318895920464)\n", - "334 (1.0328026978865914, 1.1517153952432715)\n", - "335 (1.0345994700556216, 1.1339586771052816)\n", - "336 (1.0356478349022242, 1.1387823161871538)\n", - "337 (1.0363468942434892, 1.141047185400258)\n", - "338 (1.0330405162728351, 1.1531915856444317)\n", - "339 (1.0342830771985261, 1.1591535749642745)\n", - "340 (1.0273139383481897, 1.1444447895754939)\n", - "341 (1.0246166125587794, 1.1464537532433219)\n", - "342 (1.0271252471467722, 1.1525707965311798)\n", - "343 (1.0310020021770312, 1.1657600319903829)\n", - "344 (1.036887876883797, 1.1722449841706648)\n", - "345 (1.0263645379439643, 1.1614459685657335)\n", - "346 (1.0204919162003889, 1.1578345480172529)\n", - "347 (1.0334540429322616, 1.1721395067546678)\n", - "348 (1.0263580695442531, 1.1597896990568741)\n", - "349 (1.028287635160529, 1.1585095908330834)\n", - "350 (1.0312383086785026, 1.1790412379347759)\n", - "351 (1.0375405591467153, 1.1677709263304006)\n", - "352 (1.0318952487862629, 1.1582336643467779)\n", - "353 (1.0291378259658814, 1.1559235147807909)\n", - "354 (1.0150156974792481, 1.1696013709773188)\n", - "355 (1.0302156038906263, 1.1772369540256002)\n", - "356 (1.0131931641827459, 1.1670813026635543)\n", - "357 (1.0288627147674561, 1.1678081714588664)\n", - "358 (1.031393591735674, 1.1638143010761426)\n", - "359 (1.0247207014457038, 1.1799902423568394)\n", - "360 (1.0255438338155332, 1.1767726893010346)\n", - "361 (1.0168931121411531, 1.1648034095764159)\n", - "362 (1.0123746939327405, 1.1835304892581442)\n", - "363 (1.0120658760485441, 1.1709125327027363)\n", - "364 (1.0117184312447258, 1.183259698100712)\n", - "365 (1.0184814126595207, 1.1734674935755522)\n", - "366 (1.0176250841306604, 1.1791203535121419)\n", - "367 (1.0278602465339328, 1.1738930463790893)\n", - "368 (1.0108408461446348, 1.191278466452723)\n", - "369 (1.015065652391185, 1.1885712566583053)\n", - "370 (1.0153814383175062, 1.1748331303181856)\n", - "371 (1.0050021508465643, 1.1915775983229928)\n", - "372 (1.002527846460757, 1.1925031900405885)\n", - "373 (1.0056702515353326, 1.1836440552835878)\n", - "374 (1.0160857516786326, 1.1639224643292634)\n", - "375 (1.0189541339874268, 1.1926031921220863)\n", - "376 (1.014732669747394, 1.1852628878925158)\n", - "377 (1.0004124615503394, 1.1688532196957133)\n", - "378 (0.99688930096833606, 1.1884836632272471)\n", - "379 (1.0070124610610631, 1.1886068893515545)\n", - "380 (1.0019825950912806, 1.2141083390816398)\n", - "381 (1.0022483322931373, 1.1888621522032696)\n", - "382 (1.0065551141034002, 1.1757550856341485)\n", - "383 (1.0095080681469129, 1.2161582812019016)\n", - "384 (1.0024910382602525, 1.1942758482435476)\n", - "385 (0.99947130783744476, 1.2087061420730922)\n", - "386 (1.0027282326117806, 1.190985924264659)\n", - "387 (1.0007191538810729, 1.1921328570531762)\n", - "388 (1.0020456350368001, 1.1987044267032458)\n", - "389 (1.0019353306811789, 1.2165064542189887)\n", - "390 (0.98688011946885479, 1.1908985091292339)\n", - "391 (1.0104571233624997, 1.2145118910333386)\n", - "392 (0.99680847033210418, 1.2036457756291266)\n", - "393 (0.98757092641747513, 1.2058897992838984)\n", - "394 (0.99415124032808388, 1.1996125682540562)\n", - "395 (0.99816022904022883, 1.2311710373215052)\n", - "396 (0.98751724751099301, 1.210432431490525)\n", - "397 (0.99114588861880093, 1.2063761042511982)\n", - "398 (0.99109065636344573, 1.2037215828895569)\n", - "399 (0.99364452258400293, 1.2026037957357323)\n", - "400 (0.99339977917463884, 1.2180465071097664)\n", - "401 (1.0019065981325896, 1.2364886112835096)\n", - "402 (0.98928052238796071, 1.2158684403999993)\n", - "403 (0.9960503458976746, 1.2197149178256159)\n", - "404 (0.99228734659112017, 1.2188512568888457)\n", - "405 (0.99730241816976795, 1.2186866853548133)\n", - "406 (0.98448995144470874, 1.2118749255719392)\n", - "407 (0.98383546912151831, 1.2099614609842715)\n", - "408 (0.98782284311626267, 1.2193438276000645)\n", - "409 (0.98838030825490542, 1.2159527846004652)\n", - "410 (0.98566196223963864, 1.2458368534627169)\n", - "411 (0.98516834559647937, 1.2198979621348174)\n", - "412 (0.98212570625802742, 1.2272355105565942)\n", - "413 (0.9947710928709611, 1.2173905693966409)\n", - "414 (0.9788789489994878, 1.2305930458981058)\n", - "415 (0.98816393199174302, 1.2115918081739674)\n", - "416 (0.98580364869988479, 1.2336656192074651)\n", - "417 (0.99024714386981472, 1.2382556433263032)\n", - "418 (0.9807121458260909, 1.2283255618551503)\n", - "419 (0.96887932393861853, 1.2158918147501738)\n", - "420 (0.98571897175001066, 1.2383272642674654)\n", - "421 (0.9733897846678029, 1.2279995773149572)\n", - "422 (0.98846726676692132, 1.2449409484863281)\n", - "423 (0.98318220480628638, 1.2452604247176129)\n", - "424 (0.98450899124145508, 1.2359824781832487)\n", - "425 (0.975953733921051, 1.2163982702338176)\n", - "426 (0.97490316836730295, 1.2488578081130981)\n", - "427 (0.98331378542858616, 1.2457415321598881)\n", - "428 (0.97135990028795993, 1.2493930262068043)\n", - "429 (0.9744650877040365, 1.254887416570083)\n", - "430 (0.97534642789674841, 1.2410343885421753)\n", - "431 (0.96969083806742795, 1.2450923665710119)\n", - "432 (0.97273606943047564, 1.2418002004208772)\n", - "433 (0.96399082826531457, 1.2469403681547746)\n", - "434 (0.96959618381831958, 1.2456323675487353)\n", - "435 (0.97014619163844895, 1.2281716004661891)\n", - "436 (0.96981305505918425, 1.2628395360449085)\n", - "437 (0.96645745712777842, 1.2572846086128899)\n", - "438 (0.96132570245991578, 1.2337554195652838)\n", - "439 (0.96194527719331824, 1.2371098502822544)\n", - "440 (0.97035705058471011, 1.2615821719169618)\n", - "441 (0.96378851558851164, 1.2389796028966489)\n", - "442 (0.9789252944614576, 1.2465264662452367)\n", - "443 (0.96373109713844629, 1.2463390417720961)\n", - "444 (0.96440560403077502, 1.2521775706954625)\n", - "445 (0.96883975733881411, 1.2564371347427368)\n", - "446 (0.96713639134946072, 1.2498031543648762)\n", - "447 (0.95972156576488332, 1.251833458568739)\n", - "448 (0.96024649557859998, 1.2428638033244921)\n", - "449 (0.95348995457524843, 1.2484563749769459)\n", - "450 (0.95760008770486582, 1.2543177770531695)\n", - "451 (0.95481168446333509, 1.2484611889590387)\n", - "452 (0.94951753460842625, 1.2875053198441215)\n", - "453 (0.95669987668161804, 1.2561484886252361)\n", - "454 (0.96034746066383692, 1.2788585740586986)\n", - "455 (0.95631661259609724, 1.2677563371865646)\n", - "456 (0.94761000716167942, 1.2834296013997948)\n", - "457 (0.95097091716268789, 1.2720823194669642)\n", - "458 (0.95669341968453447, 1.2786609789599543)\n", - "459 (0.95645303881686672, 1.2758742674537327)\n", - "460 (0.95646172139955599, 1.2619709528010825)\n", - "461 (0.94576526310132902, 1.2790498733520508)\n", - "462 (0.94589370696440989, 1.271905907858973)\n", - "463 (0.94648432109666902, 1.2782192468643188)\n", - "464 (0.95490319832511572, 1.286858516154082)\n", - "465 (0.94545441399449892, 1.2726128661114238)\n", - "466 (0.94413801949957143, 1.2614867687225342)\n", - "467 (0.95714525502660996, 1.2621244529019231)\n", - "468 (0.94477774112120916, 1.2796207785606384)\n", - "469 (0.95459047141282449, 1.2767061249069545)\n", - "470 (0.9480211672575577, 1.2616589836452319)\n", - "471 (0.93982599973678593, 1.2896254757176275)\n", - "472 (0.9430409488470658, 1.2912061965983848)\n", - "473 (0.95392808655033945, 1.3098617175351019)\n", - "474 (0.93599950075149541, 1.2677693071572678)\n", - "475 (0.941600647698278, 1.2909015365268872)\n", - "476 (0.93767955614172893, 1.2861960706503495)\n", - "477 (0.9342052273128344, 1.28046163165051)\n", - "478 (0.93467909875123401, 1.3030612872994465)\n", - "479 (0.92869796027307927, 1.282718827413476)\n", - "480 (0.93508614249851396, 1.2959820420845696)\n", - "481 (0.93490090214687849, 1.2881660632465197)\n", - "482 (0.93849119207133413, 1.2950395822525025)\n", - "483 (0.93482867272003833, 1.2972442611404087)\n", - "484 (0.9391955997632897, 1.2823354057643725)\n", - "485 (0.93475209733714226, 1.291724665786909)\n", - "486 (0.93975991425306904, 1.2996574049410612)\n", - "487 (0.92332864947940996, 1.296057152748108)\n", - "488 (0.93256369259046468, 1.3051859980044158)\n", - "489 (0.92304702945377515, 1.2929749063823535)\n", - "490 (0.92362498148627903, 1.3005575205968773)\n", - "491 (0.93033546675806456, 1.3172764855882395)\n", - "492 (0.92276190882143772, 1.297333232216213)\n", - "493 (0.9231779342112334, 1.298029913591302)\n", - "494 (0.92410729605218633, 1.3216223794481028)\n", - "495 (0.91360711584920473, 1.3081747366034466)\n", - "496 (0.91642054215721458, 1.3187138687009397)\n", - "497 (0.92071057402569312, 1.3147896709649458)\n", - "498 (0.9257079139999721, 1.3276651983675749)\n", - "499 (0.92306890850481782, 1.3125522836394932)\n", - "500 (0.92381941339244011, 1.3220700466114541)\n" + "1 (-1.4845550561392749, -0.1630677901622322)\n" ] + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAASwAAAElCAYAAABect+9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nBYuFhaVg+P/sFI2g9SeJYQAAAABJRU5ErkJggg==\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAASwAAAElCAYAAABect+9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\njIEJFoPByBiYYDEYjIyBCRaDwcgY/j+ubEyz9wt8QQAAAABJRU5ErkJggg==\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAASwAAAElCAYAAABect+9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" } ], "source": [ "# Train the GAN.\n", - "for _ in range(max_epochs):\n", - " loss, duration = gan._single_epoch_train(train_data, batch_size, noise_params)\n", - " epoch = int(gan.sess.run(gan.epoch.assign_add(tf.constant(1.0))))\n", + "for _ in range(n_epochs):\n", + " loss, duration = gan._single_epoch_train(latent_data, batch_size, noise_params)\n", + " epoch = int(gan.sess.run(gan.increment_epoch))\n", " print epoch, loss\n", "\n", - " if save_model and (epoch % saver_step == 0 or epoch <= 5):\n", + " if save_gan_model and epoch in saver_step:\n", " checkpoint_path = osp.join(train_dir, MODEL_SAVER_ID)\n", " gan.saver.save(gan.sess, checkpoint_path, global_step=gan.epoch)\n", "\n", @@ -776,22 +347,34 @@ " syn_latent_data = gan.generate(n_syn_samples, noise_params)\n", " syn_data = ae.decode(syn_latent_data)\n", " np.savez(osp.join(synthetic_data_out_dir, 'epoch_' + str(epoch)), syn_data)\n", - " for k in range(3):\n", - " Point_Cloud(syn_data[k]).plot()\n", + " for k in range(3): # plot three (syntetic) random examples.\n", + " plot_3d_point_cloud(syn_data[k][:, 0], syn_data[k][:, 1], syn_data[k][:, 2],\n", + " in_u_sphere=True)\n", "\n", - " train_stats.append((epoch,) + loss)" + " train_stats.append((epoch, ) + loss)" ] }, { "cell_type": "code", - "execution_count": 20, + "execution_count": 54, "metadata": { "collapsed": false, "deletable": true, "editable": true, "scrolled": true }, - "outputs": [], + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZQAAAEWCAYAAABBvWFzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nWSYcKGZmlon/D828+HCOvPqIAAAAAElFTkSuQmCC\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], "source": [ "if plot_train_curve:\n", " x = range(len(train_stats))\n", @@ -799,7 +382,7 @@ " g_loss = [t[2] for t in train_stats]\n", " plt.plot(x, d_loss, '--')\n", " plt.plot(x, g_loss)\n", - " plt.title('Latent GAN training. (%s, %s)' %(class_name, ae_loss))\n", + " plt.title('Latent GAN training. (%s)' %(class_name))\n", " plt.legend(['Discriminator', 'Generator'], loc=0)\n", " \n", " plt.tick_params(axis='x', which='both', bottom='off', top='off')\n", @@ -808,46 +391,6 @@ " plt.xlabel('Epochs.') \n", " plt.ylabel('Loss.')" ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "collapsed": true - }, - "outputs": [], - "source": [ - "* Εχεις κρατησει τα txts για του περισσοτερους AEs?\n", - "* You can manually read them and write them to the appropriate functions and then also \n", - "add some reading/creation function from raw txt for the configuration.\n", - "\n", - "\n", - "\n", - "\n", - "\n", - "# # TEMP\n", - "# TO load work with pre-trained SN models.\n", - "# import sys\n", - "# sys.path.append(\"../../../../Git_Repos/\")\n", - "# from research.iclr.helper import load_multiple_version_of_pcs\n", - "# # in_data = load_multiple_version_of_pcs('uniform_one', syn_id, n_classes=1)\n", - "# # train_data = in_data['train']\n", - "# !cat /orions4-zfs/projects/optas/DATA/OUT/iclr/nn_models/ae_chair_mlp_with_split_1pc_usampled_bnorm_on_encoder_only_2048_pts_128_bneck_emd/configuration.txt\n", - "\n", - "# ae_configuration = '/orions4-zfs/projects/optas/DATA/OUT/iclr/nn_models/ae_chair_mlp_with_split_1pc_usampled_bnorm_on_encoder_only_2048_pts_128_bneck_emd/configuration.txt\n", - "\n", - "ae_conf = Conf.load(ae_configuration)\n", - "\n", - "# saved_epochs = read_saved_epochs(ae_conf.train_dir)\n", - "# _, best_epoch = find_best_validation_epoch_from_train_stats(osp.join(ae_train_dir, 'train_stats.txt'))\n", - "# if best_epoch % ae_conf.saver_step != 0: # Model was not saved at that epoch.\n", - "# best_epoch += best_epoch % ae_conf.saver_step\n", - "# ae_conf.encoder_args['verbose'] = False\n", - "# ae_conf.decoder_args['verbose'] = False\n", - "# reset_tf_graph()\n", - "# ae = PointNetAutoEncoder(ae_conf.experiment_name, ae_conf) \n", - "# ae.restore_model(ae_conf.train_dir, best_epoch, verbose=True)" - ] } ], "metadata": { diff --git a/notebooks/train_raw_gan.ipynb b/notebooks/train_raw_gan.ipynb new file mode 100755 index 0000000..12eb17e --- /dev/null +++ b/notebooks/train_raw_gan.ipynb @@ -0,0 +1,343 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## This notebook will help you train a raw Point-Cloud GAN.\n", + "\n", + "(Assumes latent_3d_points is in the PYTHONPATH and that a trained AE model exists)" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": { + "collapsed": false, + "deletable": true, + "editable": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Picking GPU 0\n" + ] + } + ], + "source": [ + "from general_tools.notebook.gpu_utils import setup_one_gpu\n", + "GPU = 0\n", + "setup_one_gpu(GPU)" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "import numpy as np\n", + "import os.path as osp\n", + "import matplotlib.pylab as plt\n", + "\n", + "from latent_3d_points.src.autoencoder import Configuration as Conf\n", + "from latent_3d_points.src.neural_net import MODEL_SAVER_ID\n", + "\n", + "from latent_3d_points.src.in_out import snc_category_to_synth_id, create_dir, PointCloudDataSet, \\\n", + " load_all_point_clouds_under_folder\n", + "\n", + "from latent_3d_points.src.general_utils import plot_3d_point_cloud\n", + "from latent_3d_points.src.tf_utils import reset_tf_graph\n", + "\n", + "from latent_3d_points.src.vanilla_gan import Vanilla_GAN\n", + "from latent_3d_points.src.w_gan_gp import W_GAN_GP\n", + "from latent_3d_points.src.generators_discriminators import point_cloud_generator,\\\n", + "mlp_discriminator, leaky_relu" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [ + "%load_ext autoreload\n", + "%autoreload 2\n", + "%matplotlib inline" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": { + "collapsed": false, + "deletable": true, + "editable": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Give me the class name (e.g. \"chair\"): chair\n" + ] + } + ], + "source": [ + "# Use to save Neural-Net check-points etc.\n", + "top_out_dir = '../data/' \n", + "\n", + "# Top-dir of where point-clouds are stored.\n", + "top_in_dir = '../data/shape_net_core_uniform_samples_2048/'\n", + "\n", + "experiment_name = 'raw_gan_with_w_gan_loss'\n", + "\n", + "n_pc_points = 2048 # Number of points per model.\n", + "class_name = raw_input('Give me the class name (e.g. \"chair\"): ').lower()" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "6778 pclouds were loaded. They belong in 1 shape-classes.\n", + "Shape of DATA = (6778, 2048, 3)\n" + ] + } + ], + "source": [ + "# Load point-clouds.\n", + "syn_id = snc_category_to_synth_id()[class_name]\n", + "class_dir = osp.join(top_in_dir , syn_id)\n", + "all_pc_data = load_all_point_clouds_under_folder(class_dir, n_threads=8, file_ending='.ply', verbose=True)\n", + "print 'Shape of DATA =', all_pc_data.point_clouds.shape" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Set GAN parameters." + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [ + "use_wgan = True # Wasserstein with gradient penalty, or not?\n", + "n_epochs = 10 # Epochs to train.\n", + "\n", + "plot_train_curve = True\n", + "save_gan_model = True\n", + "saver_step = np.hstack([np.array([1, 5, 10]), np.arange(50, n_epochs + 1, 50)])\n", + "\n", + "# If true, every 'saver_step' epochs we produce & save synthetic pointclouds.\n", + "save_synthetic_samples = True\n", + "# How many synthetic samples to produce at each save step.\n", + "n_syn_samples = all_pc_data.num_examples\n", + "\n", + "# Optimization parameters\n", + "init_lr = 0.0001\n", + "batch_size = 50\n", + "noise_params = {'mu':0, 'sigma': 0.2}\n", + "noise_dim = 128\n", + "beta = 0.5 # ADAM's momentum.\n", + "\n", + "n_out = [n_pc_points, 3] # Dimensionality of generated samples.\n", + "\n", + "\n", + "discriminator = mlp_discriminator\n", + "generator = point_cloud_generator\n", + "\n", + "if save_synthetic_samples:\n", + " synthetic_data_out_dir = osp.join(top_out_dir, 'OUT/synthetic_samples/', experiment_name)\n", + " create_dir(synthetic_data_out_dir)\n", + "\n", + "if save_gan_model:\n", + " train_dir = osp.join(top_out_dir, 'OUT/raw_gan', experiment_name)\n", + " create_dir(train_dir)" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "metadata": { + "collapsed": false, + "deletable": true, + "editable": true + }, + "outputs": [], + "source": [ + "reset_tf_graph()\n", + "\n", + "if use_wgan:\n", + " lam = 10\n", + " disc_kwargs = {'b_norm': False}\n", + " gan = W_GAN_GP(experiment_name, init_lr, lam, n_out, noise_dim,\n", + " discriminator, generator,\n", + " disc_kwargs=disc_kwargs, beta=beta)\n", + " \n", + "else: \n", + " leak = 0.2\n", + " disc_kwargs = {'non_linearity': leaky_relu(leak), 'b_norm': False}\n", + " gan = Vanilla_GAN(experiment_name, init_lr, n_out, noise_dim,\n", + " discriminator, generator, beta=beta, \n", + " gen_kwargs=gen_kwargs, disc_kwargs=disc_kwargs)\n", + "\n", + "accum_syn_data = []\n", + "train_stats = []" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "1 (0.07408812570351142, -0.1206897652225086)\n", + "INFO:tensorflow:../data/OUT/raw_gan/raw_gan_with_vanilla_gan_loss/models.ckpt-1 is not in all_model_checkpoint_paths. Manually adding it.\n" + ] + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAASwAAAElCAYAAABect+9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAASwAAAElCAYAAABect+9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngkUQhGD4H6WP5w09KIE7AAAAAElFTkSuQmCC\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAASwAAAElCAYAAABect+9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJzsvXl4XPV59/09s++LRrNpRosly8Y2\nBgw20JYQB0gAk8dJGwIkIUCxgadN+pIrTYtbAk3pw1tzJW/6JHGANEDshBDshhCTNoGwr4ltgY13\nS7a1LyNpNJp9P+f9Q/kdzoxmpNlnjvz7XJcubTNnzplzznfu+/7dC8NxHCgUCkUMSOq9AxQKhVIo\nVLAoFIpooIJFoVBEAxUsCoUiGqhgUSgU0UAFi0KhiAYqWBQKRTRQwaJQKKKBChaFQhENVLAoFIpo\nkBX5eFrHQ6FQqgFTyIOohUWhUEQDFSwKhSIaqGBRKBTRQAWLQqGIBipYFApFNFDBolAoooEKFoVC\nEQ1UsCgUimiggkWhUEQDFSwKhSIaqGBRKBTRQAWLQqGIBipYFApFNFDBolAoooEKFoVCEQ1UsCgU\nimiggkWhUEQDFSwKhSIaqGBRKBTRQAWLQqGIBipYFApFNFDBolAoooEKFoVCEQ1UsCgUimiggkWh\nUEQDFSwKhSIaqGBRKBTRQAWLQqGIBipYFApFNFDBolAoooEKFoVCEQ1UsCgUimiggkWhUEQDFSwK\nhSIaqGBRKBTRQAWLQqGIBipYFApFNFDBolAoooEKFoVCEQ1UsCgUimiggkWhUEQDFSwKhSIaZPXe\nAQolHxzHgeO4jJ+zv1iWnfe7VCqFUqkEwzCQSCRgGAYMw9T5aCiVgAoWpWrkEhvh7yzLAgAvOuR7\nIpGA3++HxWJZcNtEhIRilE6nAQCpVCrj8ULxkkgkkEgk855LaXyoYFEWJJ+FQ/4mtHCyrR3yuIW2\nnUt0UqkUJicnYbVaixYUIlhEkITHQPZvamoKAGCz2TIEjFhiVMQaFypY5wD5LJyFXKt8osNxHE6e\nPIlVq1ZlCA6ADPEp5+bPJWLlkL2P8Xic/53juJzWmFQqpS5lA0IFSwQU6lrF43GwLAuZTFa2lZNP\ndDiOQywWg1Qqregx1hrhcWUfH0BdykaFClaNWMi6AebHcUpxrTweD5LJJFpbWwHkvykrdSxLkYVE\njGVZ+P1+eDwedHd38+8vdSlrBxWsIijVtYrFYggGg3mDyJVyrYQWACU/HMcV9R5li1c6nYZEIuHP\nL3Upa8c5JViFBJCB/NbOYtvO51olk0l4vV7YbLZqHBalTlCXsvaIWrA8Hg+MRiOkUmlJrtXY2Bh0\nOh30ev28C6dSAWThtiiF08jv2UL7tphLefr0aZhMJlgsFupSloCoBWvLli145JFH4HK5ABTvWiUS\nCT5IXU0YhqlZ3GcpxJeqfQzlbL+U5wqvOZZleZGiLmXxiFqwiNCUumJVSyGpBfSiLpxGeK+oS1k8\nohcskijYyFALq7ZU8ybOtuIr/fzFXMrp6WkEg0F0dHScky6lqAVLKpXO+yQqBmphNeZrLGVKEbxs\n8WJZ9px1KUUtWBKJhF/ZK4VaCdZSE0axU46VVK6FVS7C6+hcdClFLVhicQkpFEK9XUqO4zA6Ogqt\nVguz2Sw6l/KcFqylaGFRS666NML7W45LyTAMYrEYVCoVgMVrKcn3RkHUgkVjWJk00oVVa1iWRSqV\nQjQahVKphFKpzPvYcq2calpI1X4+2QZQuEtZ7bSfYmicPSkBqVRKLawlBsuySCaTSKVS8Pv9SCaT\n/O/kZ+HvQuRyOf8hlkqloFarYTQaYTQaYTAYoFAoKrKP5QpGuYJVLguJXraINdqH4DktWLWilie9\nUYSRWDzZYlOI8EgkEsTjcUxMTEAul0Mul0Mmk0GtVvM/k+/ZNYHJZBIymQwKhQLRaBR+vx/T09M4\nc+YMUqkUtFotUqkUjEYjUqlU0dZDue9vJZ5fCQur0YSoUEQtWDKZTBSrhEBthKQaF2E+4RkeHi7I\n4iHiUozwhMNhDA4OYuXKlSXvN8Mw0Gg00Gg0cDqdAObOQTgcRl9fH/x+P3p6esCyLHQ6HW+J6fX6\nqrfOqbdYFCNY9d7XbEQtWGJyCetNORZPtvBwHFewxVMK1TonDMNAp9PBYDBAq9XC4XCAZVmEQiH4\n/X6MjIwgGAwCAPR6PS9iOp2OP65KxKDKgVpYIqbcoHstqdRNuJDw+P1+RKNRRCKRilo82Xi9Xt5q\naTSKfZ8lEgkMBgMMBgPfRyydTiMQCCAQCGBgYAChUAgSiQRGo5F/jVJv+kYIugON8SFaCqIXrHIt\nrHJcynKopMVDftdoNJBIJGhvb6+oxSMmKnEjSqVSmM1mmM1m/m/kA2F0dBSBQADvvfce5HJ5RlBf\nrVZXXQiohSViGsElLER4EokEAoEADhw4kPHcSlk8BK/XC47joNVqyzomsbPYzVjKDSuTyWCxWJBM\nJqHVatHV1cVP9wkEAhgbG+NTKoQiRvKdynntbCqV1iBGRC1YlUwcJcKTz8LJ/lnIYsIDzLWyWbdu\n3Tln8SxFiGAoFApYrVZYrVb+f7FYDH6/H7OzsxgaGkI8Hs9IryjXoq92WkOjI2rBymVhFSM8kUgE\nHMfB4/EAqLzFQ0ilUnwJRLUR86dnLSmnlnAhVCoVVCoV7HY7/3hheoXP58PBgwczgvoGg6Hg9Arq\nEtaAF198Effeey/S6TS2bt2Kbdu25Xzcc889hxtvvBEHDhzA+vXr825v586d6Onpwdtvvw2GYXDs\n2DFcddVV/IkoVHhmZmYQj8fR3t5erUOvKWK9CGtNuaJezPucnV4Ri8XQ3d0NjuPg9/sxMTGB3t7e\notIrqGBVkXQ6ja985St4+eWX4Xa7sWHDBmzevBmrV6/OeFwwGMT3vvc9XHbZZYtuc+3atVi5ciUU\nCgW6urpwyy23lJTFTFp0VBua6b50qFRaglarhU6n47vlFppeQV3CKrN//34sX74cnZ2dAIBbbrkF\ne/funSdYDzzwAO677z58+9vfXnSbl1xyCQDgnXfegVqtLrnkQqwnjSJucl13+dIrgsEg/H4/n16R\nSCR44TIajdBqtSVdx2K99qsuWKOjo/wJAAC32419+/ZlPOaDDz7A8PAwbrjhhoIEi1CJPKylZGEt\nJUuuUbuG1jJxVCqVwmQywWQy8X/r6+sDMLeIc/r0aYTD4aLTK8R8jdQ96M6yLL7+9a9j586dRT9X\nKpWKpjSHQgHKFzyJRMJn6ROKTa+gLuECuFwuDA8P87+PjIzwfjswF7s6evQoNm7cCACYmJjA5s2b\n8cILLywYeAcaIw+rkV5nqdDI71W99y2X2ORLrwgEAvD7/fPSK+LxeMGeSaMJW9UFa8OGDejr60N/\nfz9cLheeffZZPPPMM/z/jUYjpqen+d83btyI73znO4uKFTCXhhCPx6uy32Kl3jeUGKjlKmGu167F\n80l6BRneK0yvSCQSOHbsGNLpNLRabUnpFfWi6nsnk8mwY8cOXHvttUin07jzzjuxZs0aPPjgg1i/\nfj02b95c8rbFZGHVgkb7NGxk6tXAr17PF6ZXjIyM4MILL4RcLkc4HF4wvaK5ubnq3SuKoSZyumnT\nJmzatCnjbw899FDOx77xxhsFb5d2HKXUmkboh1UuxEoj3SsWSq8wmUznnmBVC7E08KsV5JOX5Tic\nmAihzayGXiXqU1w16mmN1rtbw0LbyE6vqFSX1koh6sI2MTXwqxUcx+HVU9P42i+P4/7fnKr37jQk\n5Y6qr3eL43O5vYyoBUssMaxaY1TJIWGAZq283rtCyaIRBI+mNdSJSsSwliKXdpjw3F2XQKNonNhD\nI1GuhVVuEXsjuIRiRfSC1QjtOhoFocWoU4r61Faden1Y0RbJ5SFql1Asg1QpxdOoN1QjtDimMSyR\nQmNYSxN6TvJzrr8357RgLTWoABfGuT75uRgazRI7pwWL3uCUYmmE66XRRKSWiDoySwXr3IHjOKTT\naSQSiYzhHsLvyWQS8XgcSqUSFosFZrMZRqOx4froi8nCajRELVjlBt2XImIRYI7jkEqlcopOKBRC\nMBjEwYMHM9JWpFIp5HI5FAoF/12lUkGv1/N/A+bKS8LhMMbGxnDixImMsV2kt1S9+mGVi1jOb7UQ\ntWBRCyuTet5IZPhHtgBlW0TC8yWTyXihIQKk0Wj46dIrV66EXC4v6riSySTfC6qlpQXAXL+o2dlZ\nzMzM4MyZM4hEIojH43A4HDCbzfNGcTU61MISKbT4uXqwLJshOsMzEfzrK6NoM0rxOXcUhw8fRjKZ\nRDqd5m8gofVDfhZaPwqFAlKpdNEbLhAIIBgMVqyOTaFQwGaz8a1Wjhw5AqPRiEgkgtHRUcTjcej1\net4KW6jtMLWw6ovoBavcWsJzARL/yY795LKEyA1Bpg8RsRnxpRGIszjjA2TL5Oju7uYFqBpU89ww\nDMO3TwHm3p9gMAifz4fTp08jFApBo9HwAmYwGBomDlZLwWzE+0PUglWJGJYYP7FI/CdbdEKhEAKB\nAI4dO8b/jxyfVCqdZ/0oFArodLoMYcp3Y3ZyHJqafXAYVPD2H4Nara7lIVcVhmH4DgXt7e18szuf\nz4eRkREEAgHIZDKYzWZEo9G6Hnu9Lbx6I2rBWioxLI7jFlz5Ij+nUil+f3PFf1QqFZRKJdrb26FQ\nKIoe+LoQEobBn3c2AQC8/RXZZFWoVDcE0uyO9IlKJBLw+Xzwer04c+YMBgcHYTKZ+EB+LeNgVLBE\nSqPGsFiWzSlAZ8+enSdAZD+EAkS+E+uH/E0mky14sYbDYQSDQeh0upL3PRxP4cfvDmOlXYvr19hK\n3s5SQ6FQwG63IxgMwmAwoKmpCX6/Hz6fD8PDw0gkEtDr9WhqaoLJZCp5/NZiNMIHbD0RvWDVIq2B\nxH8Wsn5IABqYE6Bs64e4HcK/FRKALpZyL+j3hwP49eEJqOVSXLfaWpdP82rflOWO+QLmLFyLxQKL\nxQJg7kOKxMH6+voQDoeh1WphMpnQ1NQEvV5fEWuXuoQippQGfsIAdDQaRSKRwNjYWE4BItuWSCTz\nrB+5XA6tVjtPgPIxOTmJ5ubmso53MSpxIV/casD1q204z1EdC2GpIpFI+EB+R0cHOI5DJBKBz+fD\n0NAQAoEAFAoFYrEYpqenYTKZSh74cC6fF1ELFnEJZ2dnIZfL88aBEolERvyHBKBlMhlSqRTS6TSU\nSiX0en1BAehG5PRUGAouWfZ2dEoZ7vtUVwX2qD4UejNXO3GUjKPXarVwu90AgHg8jj/+8Y+Ympri\nB6KSOJjZbIZSqazY6y9VaiJYL774Iu69916k02ls3boV27Zty/j/d7/7XTzxxBOQyWSwWq146qmn\n0N7ennd7r7zyCu677z4kEgnMzMzgb//2b/HP//zPGdaPRqOB0WjMEKDsE82yLD744IOMydRi5Ph4\nEN/41Qmo5QweuGzxi/5cp15xIKVSCblcjlWrVgEA/2FLrLBkMgmDwcALmEajmXfN0hhWlUmn0/jK\nV76Cl19+GW63Gxs2bMDmzZuxevVq/jHr1q1DT08PNBoNHnvsMfzjP/4jdu/enXebn/jEJ9DT04Px\n8XFs2bIlY87huYhWKYNMysCkluMc/vCtCZW0cGQyGZqbm/lQAcuyCAQC8Pl8OHXqFCKRCLRaLS9g\ner2+IVos15OqC9b+/fuxfPlydHZ2AgBuueUW7N27N0OwPvGJT/A/X3755Xj66acX3CaJFdEGfnO0\nN6nx7J0XI52IYaD/bNnb6/dGcGQ0iE+uaoZaTtss1wqJRAKTyQSTyYRly5aB4ziEw2H4fD4MDg4i\nGAwiGo2iv78fFoul4UZw1YKqC9bo6GiGy+V2u7Fv3768j3/yySdx/fXXF7Rtmun+ERqFFNF0ZWJu\n//a7PvROhpFmWfzlRc6KbLORqGc/rGJfh8wNJPfQH/7wB2i1Wng8HvT29oJhmIw42GLlTGKPgTVU\n0P3pp59GT08P3nzzzYIeL7YGfrW4WCphMV61woJUmsP5LYYK7NF8Iok0/uV/emFQyfDP1y6HVDL/\nPWnUm6reFjnDMHA4HHxCayqVgs/nw+zsLAYHB5FKpTLiYGq1OuO9pIK1CC6XC8PDw/zvIyMj/Jst\n5JVXXsHDDz+MN998s6DVEqD8xNGlRqUuxFsvdePWS90V2VYuhn1RHBoJgGGAQCwFs4aOIyuUbMEh\nC1VWqxXAXBzM7/djdnYWJ0+eRCQSgU6nyytgC9GIwlZ1wdqwYQP6+vrQ398Pl8uFZ599dl6Q/ODB\ng7jnnnvw4osv8hX1hSCmflgkXtaIF0GtWWHT4utXLYNOKa2LWNVzkGolWOj1JRIJL07CONjMzAwG\nBgYQCAQQj8dx5swZvsGhmOJgVRcsmUyGHTt24Nprr0U6ncadd96JNWvW4MEHH8T69euxefNm/MM/\n/ANCoRA+//nPAwDa2trwwgsvLLptsbmE1UYsiwgMw1S17KfQPCkxUuz5FcbB2trakEgk8MEHH0Cj\n0WBiYgInT57MEDmTydRw4+mF1CSGtWnTJmzatCnjbw899BD/8yuvvFLSdsUkWGIRk0agkd+nRrew\nCkEmk8HpdMLpnFtQSSaTfD5Yf38/0uk0jEYjbDYbn/TaKDRU0L1YJBJJ2YNUKecm9RadepFLcOVy\neUYcLJ1OIxAINKQxIGrBEtNFRy2sxqHc8yCm6y6bQo6d9MBvxNK0xtsjSslQUaw+Yn9/G8GlLQcq\nWDWCikltqMV7LOYbngoWpaFYKqIYSbL422ePYNuvTyDFVvaYKtEPq9bPrRRiFyxRx7DERC0sLDFf\niNlMR9I4PRWBRMIgHE/BqKbJpZWAChalIBrlIpkIxGFUy0oqavZHk/j1hx5owyw2VGHfhLQZ5fjm\n9d3QKqQNJVblWmeNcB00wj6UChWsGlJvl+DQiB//tPcUOixqPHbL2qKf/9KJKfzonUG4NBxuvLoK\nO/gnyI195fKmvP8nY+uFX/F4HPF4HFqtFjabDQaDIefN2SjCUQ/qfQ2WCxWsGlGrG2ShC5L8q9SQ\n0KXtJlzcZkS3MljaBhaBiFAoFEIkEsHw8DBOTYZhVrCQc3NTpbO7xgq/TCYTGIZBMpnk2xJrNBq+\n93quhnjFInYLq5h9qPe+5oIKVg2pdwxrXasRT916QU4XK5pM4+hYEGtb9FD9yV0MxVN49K1BdNu0\n+MsLHeiwaPD9z6/Bj3+7D6c8Iay0Lz6dh0yQJhZQtlWUSCTm9c4nY8/2jUTx2P4ZdDdr8H8/dx7k\ncvmidW/JZBIKhQKtra18X3Wv14ve3l5EIhEYDAbEYjEkEom6jKivtwg0gmiWAxWscwynMfdN+sM3\nB/DCYQ+uXW3F/dd1AwAOjQTwu2OTeO2UBJ+5wA4Jw+CP/bN4+kQC/z10HD//0nk5BYj00GcYhp8g\nRL6USiV0Ol3G37JFyOfzYXp6Gh1KK+RSH6wGdVHiQm5IYV/1trY2vqPn4cOHcfz4cXAcB7PZDIvF\nArPZXPBQiHqsMFYKKlh1Rixvfq3ysEp9jWA8idloCr8/MYUvrjVgxBeBSc7hylYFdDIWPe8fBMOl\nEYiyaFJw6DKwGB0d5UVIo9HwhbOVGmG2vs2EPVsvgVZRmW4CpKOnRqPB2rVrIZPJ+OGop0+fhkQi\n4d1Ho9GYM9Nb7FnyVLDqTCN8ajUKuQYWkKlB+b7IOPsPzsYAAEoJiz3vj+HFMxFs7DLgsq5m/Mdb\n45hKa/HXf96Knj4vQqkhGExNWLVqRdWPyaCq3iUqlUozeqonEgl4vV6Mjo7i+PHjUKvVvIBVYjBq\nI1yrVLCWALU4iZWysDiOQyqVyik+8Xgc4XAYBw4c4I+JTJRWKpW89SN0ycg0oa2qCbx8chrfuLoT\n7w/5IRscgrPJCN2fblSGYfBvvzuNk54Qkmng9V4vvnVDBd6YOpDvfCsUCgQlenzrvWFs7HbiznUO\nTE9P84NRDQYDIpEIEolERV+3ljSCaJbDOS9YjdJYL58IZQenGYaBVCrNECCFQgGNRgOZTIZwOIxL\nLrmk6MLVz1zgwGcucACYG2qxcYUFTZo5MXuyWQObXok9748hFE9hYjaK81v0SKRYKGRLq1iidzKE\nsdko/nB2Bl/9RCfa2trQ1tYGjuMQCARw5MgR9Pb2gmVZPv7V1NRU8lDUelDva70cxPMu56GR33xh\nrlA8HsfExAQYhskQIdLCI5cInZhl8NtTcfzNx1pxcYtpURFKp9NgGKbsKnuGYWDRftTErdWsBgB8\n+TI34ikWP3xzAD2Ds3itdxrXra5eI7568KlVNkglDM7LWgFlGAZGoxFarRYrV66EUqnE7Owspqen\ncebMGUgkEjQ1NfHTbPLFv+p9vTbCPpSD6AULmFs6L/UmLdZVEy7T53PLhCJExIf0nl9shUzII+8e\nwwejIbzbH8T57txJlNnHUm0+GPYjxQGpJIvJYGmuUTHEUyyUNbTiFDIJrl9jX/RxUqmUj28B4If6\njo+P48SJE1CpVPz/dTpdxsplPaGCVWdI19FyBItl2QXjQmSZHvgoVyjbHROukOVyD44fPw6bzQad\nbvHcJcLfXtmO1b06/NVFjpKOrVLMhBP46b4R/EWnGQ9c341PP7ofCRZ48r0hvHxyGgopg9suc8Go\nluMit7Hs1yM31VN/GMbu98fw91d34lOrrEVtg2U5pFgup8tajTFfCoUCDocDDsfcuYpEIpiZmcGZ\nM2cQCoWg1+thMBjq3hSPCladIYMo5PLMZMiFVsiEIhSJRHDw4MF57phKpYLBYMgQoVqf6JV2XUHJ\nmUKqEVR9vdeLPR+M44PhAH56+0W4vkOC/x5gMRtJYSYSAAAcGQ3ArFVg120XwmGoTELm8EwUyTSH\nMX+s6Od+/ZdHcWIiiB23XIBuW3Hv4UIU+v5qNBpoNBq43W5wHIdgMAiPx4NQKIT33nsPJpOJj39l\nX7vVhApWgbz44ou49957kU6nsXXrVmzbti3j//F4HLfddhvef/99WCwW7N69Gx0dHTm3xXEc3njj\nDXg8HgSDQXzjG9/ALbfcAp1Oxy/TMwwDuVw+zxrKXiE7dOgQzj///KpfNGLs1jDgjaDfG8Xly0y4\nemUzPr5izv35VIcC73jS8IaT/GOjKQ7KeBoGlQyBWGrRdASO4zAVSuC9sz5c0dWEZt38wQd/f00n\nblhrw4Wu4ucjjsxGkUqzmA4l0J0jzFbLQaoMw8BgMEAulyMYDOKiiy7C7OwsvF4v+vv7AYCPfy3U\n6bNSq8y0NGcR0uk0vvKVr+Dll1+G2+3Ghg0bsHnz5oxx9U8++STMZjNOnz6NZ599Fvfddx92796d\nc3sMw+Cll15CU1MTpFIprrjiCnR3d8NsNvPL9IVCG+vl55/2noQnGMc3r+vGQ59egX/89Ql855Wz\n2NyWQq53eDaWwj3PHMbh0RAMahl+ddclsOpzz5j85m968UafF2mWww3nB/HA9d3zHqNTyrC+zVTQ\nvg54I3imZwyfXm3BRe1K/PCWC+EJxLA2h9jVq6cVEQsSoG9qakJ3dzeSySRmZmbg8Xhw6tQpKBQK\nPv6l1+v567lS1lEjClGh1ESw9u/fj+XLl6OzsxMAcMstt2Dv3r0ZgrV3715861vfAgDceOON+OpX\nv7rgCdq+fTsA4KWXXsKnPvUpNDUtHpTORa0ES4zCeEmbEfsH/VjWrEEglsJrp7xIpDk8PgvkO5JD\noyEAwGw0hTuf/hCfPt+OW9a3ZNQvTgbjeK13GolkGmqFFBe4yk+ReP7QBF447MFsOIGL2i2wG5Sw\nGwobyFtv5HI57HY77Pa5YH80GuWtr2AwCJ1Ox7uP5UJdwgIYHR1Fa2sr/7vb7ca+ffvyPkYmk8Fo\nNMLr9fJZyPmQSqWimJxTi4uk0qL4jWu6Mn6//XI3dv5xBMl0Ya8xMBPDjrcG8fbpGdx//XKM+mIA\nA3zv9X6E4nPB51Qsje+9dhbfeeUsbr/MjfXtJmi4ZNGtcDdfYMdsLIXPrl34eimXWnRrUKvVcLvd\nfPwrFArB6/XixIkTCAaDOHbsGC9gxc4QFNuHZjZLJuheKrW0fBrhYomnWOz84zDazWpcV+Qw07/7\neAdkEgZ7PxjCWLjw5304FsRNTx7M+T8WwFR4bgX2R+8MYsebA9DIJfjhpuL2rcuqxb/esALJZHLR\nx9bTyigl/qXX66HX6+F2u9HT04OWlhZMT09jcHAQLMtmxL8W62ZBLawCcLlcGB4e5n8fGRmBy+XK\n+Ri3241UKgW/38/nuCxEucNUa+kS1oLFXufQiB/Pvj8GhVSCq89rhlxauC3DMAz+98fa8ZsPhsrd\nzZzEUnPnIZhg8fZQFD2zQ/jMBXbY8sTB6kG9+sEThFOagbkKiZmZGUxNTaG3txdyuZyPf+VqYEgF\nqwA2bNiAvr4+9Pf3w+Vy4dlnn8UzzzyT8ZjNmzdj165d+LM/+zP88pe/xFVXXVXQGysWwQIaw8Ja\n49Tj6hXNWNasKUqsgDmx++XBCYxHqrRzAn7fF4A3EUIsxeLvPt5R8PMa4T1eiHLEItexyWQy2Gw2\n2GxzFmksFsPMzAwGBwcRDAb5BobNzc1Qq9VUsAp6EZkMO3bswLXXXot0Oo0777wTa9aswYMPPoj1\n69dj8+bN2LJlC7785S9j+fLlaGpqwrPPPlvQtqVSKZ/USVkcnVKGb+ZYkVsIluPw2ikvvv96P6bC\niZwrhJVmJMjhY90mXL3Cgm+/fAbjgTi+dcOKinRvqEbiaKGvWw6F7LdKpUJLSwtaWlrAcRzC4TC8\nXi9OnjyJaDQKqVQKnU6HRCJRdPyrEahZDGvTpk3YtGlTxt8eeugh/meVSoX/+q//Knq7YolhiXGV\nkHBwOIDtvz+NZJrDpe1G9I/PoL86XZJ5ZFLgniva0GpW4bfHJiFhgLPT4ZyZ9BzH4Y2+GWgVUlzU\nollwu9FEuqyxYZVIa6jV8xmGgU6ng06nQ3t7O1iW5TuvHjp0COl0OqOAe7H4VyMg+qC7RCIpa5VQ\nzEKSi2ocS3uTGh0WDaZDcVi0CrwVqvhLZKBg5tImvvH8CWzsbkKK5dBt0+KCPAmkfVMR/PtLp8Ew\nwM++fD4cebqTjviiuOfnhyDFosWHAAAgAElEQVRPx3D5ZRxKrRGvh3VWiedLJBKo1WrodDo+Vixs\nYCisjzQYik/WrQWiF6xyLaxaIWZhbNYpcP91y3H3M0fwRt8MjErAW3y1TEEwAP712lY884EHoTQD\np0EJtVyKyzvMkOS5WVuMSpzn0EGvlEGnzH9Jh+IpJFIsYkmu5EEc9RykWulMd5lMBqvVCqt1rk4z\nHo/D6/VieHgYgUAAa9as4XPDGgXRC1a5MSwxC0kuqhVQ7WhS4//Z2IH9Az7895HJqrzGBS49/s//\nWgkDoug2pNHavgwahRQ3rnNCs0CbZJ1Shu9/fg0ALNhc7zyHHj/8woXoPXaobn28aukSFrsNpVKZ\nEf9qxB5fou++RmNY1YXlOLx6chrHJ0L4X2vt+N8f64DrT7XEcgnKDsA79XJoZIBOKcH1q61YZpmL\nQckkDC9SWmXlCs9X2HVoUpV+2dczraHagiWkEn3VqkHj7VGRVCKtgZKf94f8+Pffn8Y/Pn8CKZaD\ny6TCv/yZGvdd04kmjQwKKSAv8i1k/vSlU0rwvZvWwqBRQiGV4tKOwuoGK0E9znsjpBQ0wj6UQ+PZ\nfEVSrmABtcndEauF1dGkRluTGp0WDaSC63x4NgZPaM4V18glSCVZcJgTr+Qih6lXSvHa1y4Hy81Z\nUpe0GRFLpOE2qat3IBWinjd8LS2sRuWcFyyxCkk+Kn0sVr0ST9164by/n56aq80xKKWw6BTwBGJI\npbk50+lPp8OgAIKJjwqlLRoZvvLxDqx26qGUfRST2v6Z84reL47j5o2oj8fjiMViMBgMcDgc0GgW\nTnGoNbVOa6jWNuqJ6AWLxrDqw+fXORFNpvF3H+/ASrsOkUQaM+EEdu0bwe9PeiGXABwjgVoBxBIs\nzFo5fvLlC/kYVTYvHZ/EE+8Nz23PkOJHbmULknAYh3A4q1KphMFggMFgQCKRwIkTJxCPx2GxWGC1\nWvP2WS+Fet7wYhabSiB6waqES0jJT8/gLI5NhHDTxU6o5R9ZRdetsWUUT0+FQvj/XuuHy6iCVimF\nBBwYRoJEMgVGAnxxvQttJiVmAyFw6SQvPoFwDE994MPpmQQ8YQ6/2XcSqhVycBwHuVzOD2k1m828\nOC0kPGQEfUdHB9LpNLxeL99nXavVwmazlZW3V+/E0XKhFladEYtLKNYi6//3pdOYCsXhCcRx4zon\nOpvnW0gsy6Kn34u+yTBi8QQsKgarmxVwaSXYfSKFNh2DC+Tj+M/fjWFPbwo3rNDiAqcWRq0ScU6D\nQ1NeMBI57ru2HVef14xUJIDZ2Vl0dXXl2KPCkUqlfJ0daVM8NTWFaDSKffv28TlIwiER1Ya6hOVB\nBWuJuWqVhOM4fHatFb89Nom9hydwoH8aj1xjRSwWw4cffoi+6Rh0cqBZI0W3Qo4vr9UiyUmw+2gA\nJ2dYfPw8JzRnR6DUqnD5ZRfh2HvDYKSjGIsr8eofZyGTMPj5HRfhy5dxsOkVuOH8uSRFbxWKq0mb\nYoPBAI/Hg3Xr1mFqagpnzpxBOByG2WyG1WotqESlnt0aqGCJHJlMJori50YSRjI9Ojs2lCtOdKFG\ngZYLVHj8gxTWuzQwGAyYnJxEyujGU++cgVohxe4tF0MmYXApAG84AU/iLC5pM+L6CxxoMmixzPKn\nuYaXunCBSw+7XolvPH8CzVo5NEoZ/vrPWuftX7XLXxQKBVwuF1wuF1iWhc/nw9TUFPr6+qBUKmG1\nWmGz2aDKU+ZTCtQlLB/RC1a5HUfF6qrlg+M4RCKRvGJErFESHyJxocXiRNdc+tHP/f39sBk1kMsk\nsOsVkAgOzaJV4OHNH636/Xmnmf9ZIZNgQ/tcrtWzd66by8dqgJtHIpFkzBgMh8OYmprCkSNHkEql\nYLFYYLPZYDSWP8KsHKhLuEQESywuYTmvw7LsPPERChLptBmJRHD27Fl+5YysnhFxqlS5RatZjf/a\ncjFkUiZvjd9ClPKcWqHVaqHVatHR0YFUKoXp6WkMDw/j6NGj/ARvi8VS9KSlRkhrEDvnvGDVG5JP\nlMsaisfj/NgyMsBVKEQajYb/mcxN3L9/P84///ya7LtCJkGK5bD9pT5Ekyzuv245VPLGb1FSDDKZ\njB+QynEc3nnnHQSDQfT390MqlfKBe61WKwoxEbvoLQnBasTi5+w4USAQAMdx8Pv9BeUTCQe7FtsD\nqZbMhBN4o28GHIDR2Ri6rNp5j0mzHB55+QzC8RS+eX13RnqEmGAYBlKpFN3d3eju7kYsFsP09DR6\ne3sRjUbR1NTEB+5zpV40goVF5xLWmXokjhIhypXUmCtOpFQqwXEcP86ciFEjFpcWi02vxD98shOx\nJJsz5QEAZqNJvNHrBQdg2BfDCtt8URMjKpWKn27DsixmZmYwOTmJU6dOQaPR8NaXUvlRT3oxCVYj\nInrBqmQMa6E4UTwe5y05qVSa4ZoJrSKlUplzWXxkZAQSiaTugdtq8MnzrAv+36JVYNunuhBOpLHc\nWptymZMTQZyYCGLTGjuURVp00UQa33vtDNxmNW69rHXxJ2AucN/c3Izm5ma+NfHU1BQ+/PBDsCyL\n5uZmSCSSsq15Klgip5BVwoXiRH6/HyzLYmhoCBKJJGPlTKlUQqvV8r+TOBElP8k0i9d7vVhp16G9\n6aNi5qtWVndeYDbffOEERnxRKGUSbDrfUdRzj44F8D9HJyCVSPC5dS1QL9CLKxfC1sTLli1DMpnE\n9PQ0hoaGEA6HEYvFYLVaYbFYiloEoWkNIhcsjuMQi8UwOzuLDz/8EFardZ4gkRNERId8JxaRSqXi\nG5dVk1qtRtY71+u5QxP4wev96LJq8NPb11X1tX70ziD+5+gkHvr0Slzkzmzpe+1qG9457cUaZ/Gt\nfi9wGXDTJS64TeqixSoXcrkcTqcTwNwqblNTE5+0KpfLYbPZYLVaFy3Wpi5hnQRrZmYGN998MwYG\nBtDR0YE9e/bwc9YIhw4dwt/8zd8gEAhAKpXi/vvvx80338z//9/+7d/w61//GrFYDE1NTbDb7bjm\nmmug1WrR1NRUcJzI7/dX5RjPFTiOwzd/04sz02H8x+dW45cHxxFNsnCZ5hIuf3tsEiO+KO64vLXi\nXT4PjwYRiKX+NJwiU5juuqIDd13RUdJ2lXIp/u4T5ZUF5YKs9pK5gitWrEA0GsXU1FRBxdqVEhsq\nWEWyfft2XH311di2bRu2b9+O7du345FHHsl4jEajwU9/+lN0d3djbGwMl1xyCa699lqYTHOJhw88\n8AAeeOAB/OQnP4HX68Udd9xR8v7UKnG0nATXYl6nlrAccGQsgFiSxag/huVWLWJJFndc3opEisX/\nfa0fKZbDulYjnzRaKf5lUzdOTIQyklMbnezzo1ar0dbWhra2toxi7ePHj0On08Fms6G5ubliI7nq\nbYGXS10Ea+/evXjjjTcAALfffjs2btw4T7BWrFjB/9zS0gKbzYapqSlesAi042h9kUoY/MfnVsMT\njOPiViMubjUileZ4a+q2y1wYmolijVNf8DYLtSRsemVDTYVejMXEIl+x9sGDBwHMrUrKZLKy2zSL\n+Zqvi2B5PB7ep3c4HPB4PAs+fv/+/UgkEjmr9ytRmlMry0fsn2756LJqM/KvFLKPbohbL3XXY5ca\nlmJyoEixdldXFxKJBE6fPo2ZmRm89957MJlMsNlsRc8TpIKVh2uuuQYTExPz/v7www9n/M4wzIJv\n4Pj4OL785S9j165dOeNR5RY/L2UhoTQW5VxnCoUCTU1NfK+vUou1qWDl4ZVXXsn7P7vdjvHxcTid\nToyPj8Nms+V8XCAQwA033ICHH34Yl19+ec7HiKU051xZJaTkp1LdGgop1iaB++zXE7tg1SXVevPm\nzdi1axcAYNeuXfjMZz4z7zGJRAJ/+Zd/idtuuw033nhj3m2JpVsDhVIu+cSGFGpv2LABGzZsgMFg\nwMjICN59910cPnwY4+PjfHE8Lc0pgW3btuGmm27Ck08+ifb2duzZswcA0NPTg8cffxxPPPEE9uzZ\ng7feegterxc7d+4EAOzcuRMXXXRRxrbE0q2BCiOlFrWE2cXagUAAk5OTGBgYgFQqRTQaRSQSgcFg\naEhBWoy6CJbFYsGrr7467+/r16/HE088AQC49dZbceutty66rUo08FtK/bDEeBGeS9Ty/DAMA6PR\nCKPRyBdr79+/H6dPn+bzFxcq1m5ERJ3pDlTGJawV1RbGZJrFvvEUDJ4QVtp1VX0tSvHUe/KzSqWC\nQqHARRddBIZhCirWbjSWhGCJxSWsNm+fnsEvTiXxxuQp/HLrJVV/PUpxNFJ7mYWKtdPpNKxWK1av\nXt1wFjsVLJF0HC2EbpsWDi2Djy+3VPV1KPWhWsXPuYq1Z2dnG06sgCUgWGIZQlELWs1q3LdeiUsv\n7aj3rpTFUl2cqISFVItaQlKQ3YiIXrDEYmHV2pI7MOiHw6BEm6DFi5hY6KYi7YJisRg/nl743Ww2\nw+l0wmg0NqSVUCqkePpchgrWEks3YBgGPUN+PPDfp6BWSPH8XZeI6qYVtpaempqaJ0bpdDqjXRBp\nD6TX6/lgcSwWw9DQEILBIN/Jw2w21/19aIQYltg55wWrVtRSGB0GJbQKKVbYazfRuBCIGOWzjFKp\nFP8+CUWJtAs660tCq1LkbcUMzCUcW61WOBwOvm3x2NgYjh8/DrPZDLvdXtcPqEZwCcXMkhAsmume\nSatZjefqYFktJEYk05r0uSeWkclk4n8n3TcnJycRiUTQ0dHBb3vAG8F9L/RCKmHw8zvWwaBa/NLN\nXgmbmZmBx+NBOBzGkSNHYLfb+dbFtaASaQ3nOqIXLLEUP9daGCstVsIJQMlkEv39/bwYJRIJAHPn\nQihGBoMBVquVF6Ny9smgksGgkkGjkEJVQiNAhmH4+jufzwe32w2Px4O+vj7odDo4HA40NzcX1fmg\nFKhLWB6iFyzaD2s+xV7Y6XQ6wxoS/pxIJMBxHKRSKS9EHMdBp9PBYrFApVJBLpdX/X1s0irw8zvW\ngWHKH8LKMAzf9ZOUr3g8Hpw5cwZqtRoOhwNWq7ViQ2cJlfjAWorXazGc84IF1K40px4mPZkElO2i\nCXveSyQSXoxUKhU/tp5kRmffJD6fD1brwpNyqoFUUvmbNbt8JRQKYWJiAgMDA1AqlbDb7bDZbEVP\nec5Fpbo1nMssCcE6V2NYLMvyAzeEgnTkyBF+UCuZBETESK1Ww2w2L6nZiJWCYRjo9Xro9Xp0d3cj\nHA5jYmICPT09kMvlsNvtNWn2mA/qEi4BwVqqMSzhaLJcgWwyNVo4G5G4Z11dXVCr1VSMykSr1aKr\nqwtdXV2IRCLweDyIRqM4cOAAbDYb7Hb7og3zhNC0hvIRvWCJKa2BwHEckslkzphRLBbLm2tEYkb5\nhrUCcwNbqVhVHo1Gg46ODoyPj2Pt2rXweDw4fPgwOI6DzWaDw+GAWi3OJF0xcc4LVqUtrHy5RsFg\nEKFQiK/Ryl7eJ7lGZGDruUyju+gqlQrt7e1ob29HPB7H5OQkjh49inQ6zVteWq123vOohVU+or8z\nai1Yi+UaMQwzb3nfZDJBp9NhZmYG5513XtUvuka/4QuhFjfmgQEflDIJLnAbC35O9n4plUq0trai\ntbUViUQCk5OTOHnyJJ/AarfbodN9lMArFsFqVGEUvWDJZLKKuYTpdDqvGJWbaxQIBBYduFEJGvVC\nazTOToex7fljYBgGe+7agCbt4nP/FvsgUCgUcLvdcLvdSKVSmJycxOnTpxGNRtHc3IxYLFb2h8m5\nfn5FL1iFWlgk1yh7aT8cDiMUCmH//v2QSqUZYlTJXKNz/UJrNJq1CrSY1NAppdApC78NCj2PMpkM\nLS0taGlpQSqVwvT0NCYmJuD3+/mYV7HF2UvBci6XJSNYg4ODMBqNOQPZAPgVNSJGJNcIAAYHB3HB\nBRdUfV+XSt+tpYBBLcfP/rq4Joelvq+kz7rP54PNZkM6ncbw8DCOHj2KpqYmOByOgoqzaQyrToI1\nMzODm2++GQMDA+jo6MCePXt48cgmEAhg9erV+OxnP4sdO3bwfz98+DDuvvtuxGIxeDwe3H///Xjw\nwQd5UTIajXzi40IrZtFotOLHV09qcUE3uiBW8z0oNwZFRnTZbLZ5xdkmkwkOhyNvj3WaeFonwdq+\nfTuuvvpqbNu2Ddu3b8f27dvnjaonPPDAA7jyyivn/X3VqlV48803wTAMPvaxj+Hpp58uaV9qGcQU\n8wUTT7H41aFxLLfOX/2qBySBk+M4cBzH56UBcwsjMpms4qkdlS6tyS7O9vl8mJiYwMmTJ2E0Gmte\nnC0G6iJYe/fuxRtvvAEAuP3227Fx48acgvX+++/D4/HguuuuQ09PT8b/SKlEOp1uqLSGpcp7Z314\n/O0hGNQyPHhJdUWeZVlwHMefVyJKuSACQFJBGIbhy5FI7/JqiFcpLHSdMQyDpqYmNDU1geM4zM7O\nZhRn2+12Pj+vnNcXu0tZF8HyeDxwOp0AAIfDAY/HM+8xLMvi7//+7/H0008vOEVaIpGUJTiNmule\nDtV4nbUtelzcasAFLgOAibK2lcs6Aj4SH7lcjsHBQT4pU6VSQSKR8Kus5HELiRARPaF4SaXSsoWr\nFoKRrzh7cnIS0WgUbrcbNput6Hw9KlgLcM0112BiYv6F/fDDD2f8nm+p/9FHH8WmTZvgdrsXfJ1K\nnIClZGFV64Js1inwHzeuAQAcOJBfsIRiRL4L3TUh5NxLpVLeGmIYBlarFWazGZOTk+jt7QUw98Fm\ns9mgUCyefgB89EHGsiyi0Sj/RVILkslk0QJWj+tEWJwdj8fR3NyMcDiMAwcOQKFQ8O9LIcXZVLAW\nYCGryG63Y3x8HE6nE+Pj4zkb3v/hD3/A22+/jUcffRShUAiJRAI6nQ7bt2+v6H7SGFZxCN01oSjl\ngggSsQSIIJGfF0KY0xSLxTAxMYFDhw5BLpfzN6lEIuFXg2OxGC9IwtVhuVwOlUrFf+n1erS1tSGd\nTiOVSkEikfDCVYh41dsl02q1cDqdWL58OV+c/f7770Mmk/GdJfLNFaSCVSKbN2/Grl27sG3bNuza\ntQuf+cxn5j3m5z//Of/zzp070dPTU3GxApaOkAgp53iIIJHt5LKOdDodDh8+zAeFScIs+apUvIhl\nWV6I5HI5LBYLgsEgTp8+jePHj0MikUCj0UCv10Oj0fCdKEjqymI3J8uyYFkWqVSKX8FbSLzq3c8q\nW3ByFWcfOnQIDMPAbrfPK86mglUi27Ztw0033YQnn3wS7e3t2LNnDwCgp6cHjz/+OD+uvlYspRjW\nQhdkduyI/JxvO/mso7Vr1yISiWB8fBwffvghdDodnE4nmpqairohSAF4LguJtMbJto5ITEsmkyEY\nDGJ8fBwzMzMAAIPBUFQyplCYssVLKpVmuKrC96VUqil4Go0Gy5Ytw7Jly/hUH2Fxtt1uXxI1qnU5\nAovFgldffXXe39evX59TrO644w7ccccdVdkXsX/iCBG6a6lUal4wW4jQGiommE3QaDTo6upCZ2cn\n/H4/xsfH0dvbC4vFAqfTCZ1Ox4/iEgpRNBrl3TWZTMaLkVqt5qsKVCpVQa2KDQYDDAZDRkrAqVOn\nYLFY4HA4oNfryxKvZDKZIV6VKAGrheDlKs4+fvw4EokEUqkUwuFwzuLsSu1nNRG/5JaJWFYJcwWz\nc21Pr9fj2LFjsNvtsFqtfOJsMWJU6P4IrSOFQgG9Xg+v14vR0VGwLAuVSgWDwQCdTgeVSgWTyVSw\nu1YMwpQAlmXh9XoxMDCASCTCT9BZ7AYVQsSL4zjE43GEQiHEYjGEQiH+A6EU17cSQyiKfd+ExdnB\nYBCHDx/GyZMnEY/H+fdGWJzd6FDBagBXDZjvri2WsyO8YYSCtHLlSsTjcYyPj+Po0aNQq9VoaWmB\nxWIp6qIkXSmE1hH5Ijes0DrS6/V8EbhcLkcqlYLH48HExASSySR/Y1TbLZFIJLBarbBarUilUpia\nmkJvby+SyeS8uA5pkkiOUfidFLvL5XKo1WqoVCpotVo4HA5+AhCJeRUqXvVuLyOTyaBWq3HxxRfn\nLM622+0wGAwNLV7nvGDVCuJmkNW1fEv9AOa5acUGs5VKJTo6OtDe3o5AIICxsTH09fWhubkZLS0t\n0Gg0SCaT81w1srpG3CByo6pUKpjNZv73Qtw1uVzOr/JFIhG+1XCp8a5SkEqlfBA+FArxlhfLspBK\npXxPMnJcJGivVqtz9rIXInQbgeLFqx4IBU9YnJ1OpzE1NYWBgQGEQiFYLBa0traiqampzns8n3Ne\nsCpx0xQSzGYYBolEAidPnkRLSwv/SVYtd40s90ejUX4y8vT0NIaHhwGAd9fI6prBYODdtUrfcBqN\nBp2dnVi2bBkCgcC8eJdery9pu+Q4c1lIpG22UJCcTieWLVsGhmHg9XoxNTUFhmFgMplgs9mKGvGV\nL+ZF/idcrCDU28LK93ypVAqHw8EPn52enkYsFiv5darJkhCsart0C9WtZbNQMPvyyy+Hz+fD6Ogo\nzpw5A6fTCYfDUXAyJCGdTs9z1cjvxF0TDp3QarWwWCxQq9WQy+W8y0gqDEwmE0wmU9UtHmESJLkx\nzp49i1gsBrvdDofDkbEML0xryP6ezy0VriIudDwGgwHLli3jc5kOHDjAu3wWi6Uo0SbiRa6TapUG\n1aL4WSKRFC3etWRJCFY5lFK3lp2ZDRRuHZHgcDKZxPj4OA4ePAiNRgOXy8V3rBAu92cLk3BGIPki\nVf7C6ckLoVKpsGzZMnR0dMDv92NsbAynTp2CzWaD0+mERpN/FHylIF0LtFotgsEgpqamMDw8zLtr\n5EsoSEajsajjLASSy9TZ2YlAIICJiQmcPn2af61C2r4Ij4l8J+JFmkKSgSL1buDXyPGpQlgSglVM\n7lGuC4ZhGPT29qKlpQU6na7kpf5CICtPJBnSarUiEAjg6NGjSCaTfJBXo9FkWA5qtbri7hpxh0wm\nE9LpNCYnJ3HixAlwHIeWlpaS6tWEZAfuc+VZEXfNbDbD6XSCYRj4fD5MT09DrVbz8a5qx4WE1l+x\naRLCc0qOk/xM3ETSEJLjOMRisZIsr3qsMjYaS0KwUqnUosHsfAFshmFw2WWXYXp6GoODg0in03C5\nXLDb7SXdJOQTNZd1lCuuotFo+PwjiUQCj8eDsbExJJNJ2Gw2NDc31+Qik0qlcDqdcDqdiEajGB8f\nx4EDB2AwGNDS0pLTZSSWYC5Byg7ckzwr8vtC763FYkFXVxcf7+rr60NTUxP/gVIL11WYJjE1NYUz\nZ84gEonwKRrELRcKklqt5l3w5uZmqNXqea5pvqLsQuoa6x0DawSYIlW7IWtYzjvvPDidTmzZsgXX\nX389/8lVinUUjUYxOjqKqakpNDc3w+128+ObhBNxciVDZmdnC1fZyHJ/oQSDQYyOjvJdKltaWmo6\nRoos+U9NTWFiYgKRSIRfISQ3KVkmFx5rNSxBkls1Pj6OaDSaM95VDsL0hmzxJekNZKJROp1GOBwG\ny7JwOBxwOp0lnxeheBVSGrR//36sW7eu5CnUfr8fw8PDOP/88xd9rFQqrXVmfEFKuiQECwD6+vrw\n6KOP4tVXX8XnPvc53H777TmLqheCmPaxWAyRSARerxc+nw8sy0Imk/Ff2UJUiNVQKsRVGx0dhUQi\ngcvlgtVqLfu1is1BUigUiEaj8Pl8/H7UIzibTCb5/C6GYeB0Ohd1XcmxCt21bEFSKBS8hUTOKVmk\nyGWVkAk5ZD9IjlepYkJWGknYIpd47d+/HxdffHHJQjI7O4vR0VGsWbNm0cdSwaoR4XAYP//5z/Hj\nH/8Y3d3duOeee7B+/XowDDNv1UloIQndtWwhYlkWHo8Hs7OzsNvtcLlceSviq31sY2NjmJ6ehsVi\ngcvlypvBnSuuQr4TC4ncpNkW0mI5SJFIBGNjY5iamoLRaERLS0vRAxUqAXFdPR4PNBoNTCYTFAoF\nf9xC8VUoFBlCJBTicvebdJOYnJyEXC6H0+mE1WotWcxziZdMJsOBAwewfv36koXE5/NhfHwcq1ev\nXvSxMpms1h9G56ZgEaampvC1r30Nb7/9NhKJBPR6PR5//HE+7yhblBZbBgfmrJ2JiQmMjo5CpVLB\n7XYXtYpUKViWxeTkJIaHh5FKpWA0GjNu1FyxMuH3cicAETiO43uSh8Nh2O12OJ3OirlqwtcRJroK\nv5OVN4lEgnQ6jWQyCb1eD7vdDovFUvEyoMUgaRJTU1Mlp0mQOGg0GkUkEkEkEkEsFsPs7Cz+4i/+\ngj9/xVrZMzMz8Hg8WLVq1aKPpYJVYyKRCN555x20t7dDJpPhZz/7GZ577jlce+212LJlC9rb28va\nvt/vx8jICILBIFpaWuB0Okt2B3JRaA6STCbjx5UZjUa0trbWJKcqG+KqjY+P81nUhbquJDaYHUMS\nFkoL3VOhhZQtSNnxLpKqUWkRLeSYSJrEzMwMjEYjnE4nTCYTAOR0T8m5JXFQoTVILF+y7YU6SuTD\n6/VicnKSCpZYSCaT+NWvfoXHHnsMer0ed911F6666qqy4kEkn2psbAx6vR5utxtG4+KThIWfotnf\n8wXvhdZgNhzH8YXH8XgcLS0tcDgcdWkpInRdzWYzWlpaoFKpMlw1oVsOfNRoLzuOVI6FlEwmMTk5\nifHx8YLjXZUgO4gfjUbh9/sRDAaRTCYhk8mg1Wqh0+l4i3+hc5sLodsIFFYaRLL7zzvvvEW3TwWr\nwTh8+DB27NiBAwcO4Etf+hK+9KUvFSQ0+SC5OyMjI4jFYvygzFwtVrJzkLK/l3uhxONxjI2NwePx\nQK/Xw+VyVT3GlCvFIRqNIhQK8W1atFotzGYzdDpdhoVUi9q7aDSKiYkJeDwevmtnOfldxEXNtpCI\nRSgM4gu/JBIJvF4vv/JaSjeJbAoVr+npaXi9XqxcuXLRbVLBalBmZ2fxk5/8BLt27cL69etx9913\nL7rsmyu9IVuQUqkUUhexCnIAABSsSURBVKkUNBoNmpubYTKZqrqamG8/SSlQJBIpuRQIAO+yZSdG\nZudc5bKQJBIJkskkJiYmMD4+DoVCgZaWlrqMsCKu2vj4OHw+H5qamvh6RqGgkzyrXMfMcRy/Wpx9\nzMVYhKSbBOlokatLaLEsJF5k1XvFihWLbocKVoPDsixefvll/OAHP0AwGMTNN9+MtrY2fnpJruZz\ni+UgcRyH6elpjIyMIJ1O89NO6lHNT1zX8fHxjFIgcnMJBThbhHMlgZKvUiykUCiEsbExeL1ePiG0\n1ALoUiHDKcg0mng8nrFiSCzgXHGzapy/RCIBj8cDj8dTsTQJ8p38PDU1hVgsRgVrKXHPPffg3Xff\nRTgchtlsxlVXXYUvfOELcDgcZcVUotEoRkZGMD09DavVCpfLVdNEUKHF4PP54PV6EYvF+HybfDGk\nalqEJEA+NjaGeDzOdwwoxQLMRphTJ3TdyCoqWbQQ5ltFIhHMzMyAYRg4HI66tRUWCimZjFNImoTw\nHJMVxuyFC5fLxa9aLlQaRAVLhESjUfziF7/Af/7nf6KtrQ333HMPLr/88rJiQSSna3R0FFKpFG63\nuyLlN8IgfraFJIyZZac3kIGd5GKuVSlQNolEgncZVSoV33Qw3w2VvbIo/BJmp+eyChezWoTxLo1G\nA6fTWXRqQqUIhUKYmJjA9PR0RhlXthCTovFccTPhh6ywKJuU6uQSLypYIobjOOzbtw8/+MEP0NfX\nhzvuuAM33XRT2V0NQqEQRkZG4PP54HA40NLSkjchNdeqYrYg5UqMLDSIX+9SICGk6eDMzAzftx1A\nxjED81cWhcv/lcozEw66yBfvqiTZwXyhCBORSaVSfL1ic3MzNBpNSeKSXRokrGtUKBRUsJYCHo8H\nP/7xj7Fnzx584hOfwF133YXOzs6ytplKpTA2NobR0VHIZDIYDAZIJJKcaQ65YiqVvLCqVQqUC2Gu\nWfaKGzlmhmF4l6a5uZnP7q+1tSPM74pEInw9Y7GinuuYSWJovmC+RqPJSPYliynj4+MIBAIlDd0g\nkO4RAwMD6O/vx9mzZzEwMAC73Y5/+Zd/KWpbZUIFq5qkUins3bsXjz32GORyOe666y586lOfynsj\nZV+ouZrukXhRJBLhV41aW1trnvRIKKYUKBfC8qDsL+Ex51ptyxZh0nRwYmICWq0WLS0tNWmznAth\nv3oAGfEukpVP4kf5YmfC/KtyYoWkEeJCaRKkMqK/vx/9/f0YHBzkfw6FQlAqlWhvb0dnZyffG2zl\nypVobW2t2HtWAFSwasXx48fxve99D2+99RauvPJKtLS04OKLL4bdbucv1IUspFyB3WQyibGxMYyP\nj0Ov16O1tZV3jWoNabFCpuG0tLTw7XfyuTC52q4Iv0oNZpO0hLGxMczOzsJqtcLpdJaVx1QswuB2\nIBCA1+tFOBwGgIx+ZtnHXMlKCCHkg+Hs2bMYHh7G8PAwvv/97/OxK6lUCpvNxrepJsLU1dVVlxrQ\nPFDBqiWf/exn+dYjiUQCa9euxRe+8AVcdNFFZa00kXq9kZERxONxuFwuOByOmsUXsgPbwWAQfr+f\nX2HUaDR8Imj2zVntG4EMTxgbG0M6neaFtNyVvcUsQ+EChvCLLBzMzMzwDQkrNYWGWFLEbSNW0sDA\nAAKBAG8lEUGyWCw4efIkQqEQtm/f3iiitBBUsOoFx3F47bXXsGPHDkxPT2PLli347Gc/W/ZyfSwW\n4zPYm5qa4Ha7y7YsslcXsxNg8608KRQKzMzMNEQpEDD33pDODTqdDi0tLQsWphMhznbdslcYsy2l\nQj4oWJbFzMwMxsfH+aLwxeJdRCQHBwf5eBL5Pjo6inQ6jebmZnR1dc2zkupRO1oFloZgvfjii7j3\n3nuRTqexdetWbNu2LefjnnvuOdx44418C45GYWhoCI8//jh+85vf4NOf/jS2bNmClpaWsrZJPm1H\nRkbAcRzcbnfewDgJqhZTaEt+LzSmIowv1aoUKB8cx8Hv92N0dBSzs7P8IFehpZgvuF3JFUYCmf83\nPj4OlmXx9ttv48ILL8T09HSGMPn9figUigwriQhSW1tbQd1ERI74BSudTmPFihV4+eWX4Xa7sWHD\nBvziF7+Y188nGAzihhtuQCKRwI4dOxpKsAixWAx79uzBj370I9jtdtxzzz244ooryr4Iw+EwhoaG\n4PV6+WJacnMmk0l+gk6uwHalraFKlgIV8loLxc/IcZOMdolEAofDAbfbXdVYUiKRmGclDQwMYGRk\nhF9MOHnyJNra2vCNb3yDD3LXawGhgRC/YP3hD3/At771Lbz00ksAgH//938HAPzTP/1TxuO+9rWv\n4ZOf/CS+/e1v4zvf+U5DCpaQnp4e7NixA0ePHsVtt92GW265BTqdLu/jCym0ValUSKVSCAaDkMvl\naGtrq1sZENnnhUqBCoG4q2TZn7hvwjws4dI/WX3LZY1Eo1GMjY1hcnISBoOhZCuQuHtklU0oTD6f\nL6eV1NnZiY6ODn6/OI7D6Ogo3G53Ua+9xCnoRDT0EIrR0dGMpVW32419+/ZlPOaDDz7A8PAwbrjh\nBnz729+u9S6WxPr167Fz505MT0/jiSeewNVXX43zzz8fGzZsgMFgwPr163MW2grHXS1UJhQMBjEy\nMoL+/n44HA64XK6qWDkLQUSztbUVgUAAo6Oj6O3thd1u5xNkCwluC1MArFZryfWLarWaFw/SVePk\nyZPzmg4Sy21oaAgDAwN8XhKxkpLJJJqamvgVt5UrV2LTpk3o6uoq2EpiGIaKVYk0tGAtBsuy+PrX\nv46dO3fWe1dKorm5GXq9HjqdDmNjY9i9ezesVitMJhOuvPJKaDSakiwkvV6PVatWIZVKYWJigp99\n6Ha7ax6gJcJDpsiQcfHAnKhptVreSiKDUMtJe1gMhmFgNpvBcRzOnj2Lw4cP48c//jF+97vf8T3U\nlUol2tra0NnZic7OTlx33XW8lVSL1U9KfkTtEvr9fnR1dfHu1MTEBJqamvDCCy80vFuYj97eXvzw\nhz/E66+/jptuugm33XYbmpuby9omCUSPjIwgHA7zK3qViOUIx8ULXbd8q4zEdUskEhgbG6tKKRCx\nkoaHh3mXjbhtw8PDSCaTMJvNGXlJCoUCf/zjH3HppZfi5ptvrsh+UIpC/DGsVCqFFStW4NVXX4XL\n5cKGDRvwzDPP5J36sXHjRnznO9/B9PT0giuL3/3ud/HEE09AJpPBarXiqaeeKrtlcqUJhUL42c9+\nhieffBKrVq3CPffcg4svvrjs7RKhmJiYgMFggNvtXjAhldSu5crcJsFtUrOY3T2zkBQAUgo0NjYG\nhmEKLgXiOA6zs7O8yyYMcHu9XshkMrS2tmLZsmX8altnZyfa29tr3uedUhDiFywA+O1vf4uvfe1r\nSKfTuPPOO3H//ffjwQcfxPr167F58+aMx27cuBGPPPIIvvjFLy64svj666/jsssug0ajwWOPPYY3\n3ngDu3fvrvWhFQTHcXjrrbewY8cOjI6O4s4778Rf/dVflV2uk52QarVaodVqM2JKJIYmDG5XMzlU\nWAoEzC0mmM3meXlJQ0NDSCaTMBqNvNsmTAOoR2NAStksDcEqlkJXFgkHDx7EV7/6Vbz77rs128dS\nGR0dxY9+9CM8//zzuO6667B169aC6r1y9RgnLlw6nQYw59qR6UIOhwNms7nq3VGJq0rEiFhLQ0ND\nkMlkOHHiBEKhEDZu3Igrr7wSy5cv52NJ1Epacoh/lbAUCllZFPLkk0/i+uuvr8WulY3L5cJDDz2E\nb37zm3juuedw1113wWQy4e6778Yll1zC52IJRSlXj3GdTsevuAmD28KE1ImJiQUTUguB4zik0+mM\nWJLQSkokEjAYDLzb1tnZiauuugrLly/nX3doaKjgLpmUpc+SE6xiePrpp9HT04M333yz3rtSFAqF\nAl/4whfg8Xjwk5/8BLfffjsMBgOuuOIK3HvvvTCZTDAYDEV3SJVIJLDZbLDZbAiHwxgdHcXZs2dh\ntVrhdrtzuqGkGJkEtoWxpKmpKb5JIQlub9y4EVu2bEFHRwdUKtWi+9bW1lbSe0RZmiw5wXK5XBge\nHuZ/HxkZgcvlmve4V155BQ8//DDefPPNukxxrgRf/OIXcffdd0Oj0cDn8+Gpp57C1q1bcdlll+Hu\nu+8uaP5cPrRaLVasWMGnRvT09GDnzp1obm6GQqHAwMAABgcHeTdSmCi5ceNGLF++vK6Jq5SlyZKL\nYRWysnjw4EHceOONePHFF9Hd3c3/fbG6xXg8jttuuw3vv/8+LBYLdu/ejY6OjlodWkGwLIsXX3wR\nP/zhDxGNRrF161Z8+tOfXjCviXTWzJW9PTk5CalUCpfLBZvNhlOnTsHr9WL37t1Yvnw51Go1jSVR\nKsG5GXQHFl9ZvOaaa3DkyBE4nU4Ac27H888/v2jd4qOPPorDhw/j8ccfx7PPPovnn3++YVcXAeDM\nmTN49NFH8fvf/x6f+9zn8MlPfhLBYBBn///27i2kyT+MA/i3ki4ssBh0oCCb81DCMqi0i9KcWhgY\nga4iyQuDMCO6aDYIu4jUFeSVSRdFHi4M68IFoaCBJWjpKikVSlKj0gZt5TKweXj+F7Wh/3lY5uZe\n+35g4Ls97v29gs/7O789Pe6E5KolrVixwmO/JI1G4973aqKRkRGfrcej+aWgzQO8u+u5Hnvt5WvR\nam5ulpSUFPdxYWGhFBYWTopJSUmR5uZmEREZGRkRlUol4+Pjfi3nXPz48UNKS0tFrVZLbm6uFBcX\ni9lsltevX8vQ0JAiroH+3OjoqKjVann37p38/PlTtFqtdHZ2esQ5HA7Zs2ePxMbGSltb2wKUVES8\nzEGLrg9rrrwZXZwYExQUhJCQENhstr+eie5rwcHByMnJQU5OzkIXhfyotbXVPRUEAI4ePQqz2eyx\n20l+fj4uXLigiLW47BElWgB1dXWIjIyERqOByWTy+Ly4uBhbt26FVquFTqfD+/fv//gcU92EP336\nNClm4uYBSsCE9Zs3o4sTY0ZHRzE4OAiVSuXXcpLyjY2NITc3F7W1tejq6kJVVRW6uromxWzfvh0W\niwWvXr1Ceno68vLy5r0crs0Drl+/Pu/f7StMWL/t3LkT3d3d6O3thdPpxN27dz2W/qSlpaG8vBwA\ncP/+fSQmJnqMkNntdiQnJyM8PBzJycn4+vWrx7na29uxe/duREdHQ6vVBnTHPc2/iU215cuXu5tq\nE+3bt8/93Mu4uDh8/Pjxj88z2034+/fv6OjoQEJCAkJDQ/H06VOkpaXBYrHM8cp8jwnrt6CgIJSU\nlGD//v3YsmUL9Ho9oqOjcenSJTx48AAAkJ2dDZvNBo1Gg+Li4imr8iaTCTqdDt3d3dDpdFPGBAcH\no6KiAp2dnairq8O5c+fw7ds3n18jBQZvmmoTzXU1xmw34ZCQEPdWzX19fYiLiwv8nU687Z2XRT5K\nOF8iIiKkv79fRET6+/slIiJi1t/RarXy9u1bXxeNZlFbWysRERESFhYmRUVFHp8PDw+LXq+XsLAw\n2bVrl/T29s7pPPfu3ZPs7Gz3cUVFheTm5k4ZW1lZKbGxsTI8PDyncz18+FDCw8NFrVbLlStXREQk\nPz9fzGazR2x8fHzAjxIyYc2zkJAQ98/j4+OTjqfy7NkziYqKkrGxMV8XjWbgzRSAGzduyKlTp0RE\npKqqSvR6/ZzO5c0UGhGR+vp6iYqKEqvVOqfzKAwTlq/odDqJjo72eNXU1HgkqFWrVk37Pa4aWEtL\ni6+LTLPw5zy8kZER2bx5s/T09LiTY0dHx6SYFy9eiFqt/pdq3pyH5SsNDQ3TfrZ27VoMDAxg/fr1\nGBgYwJo1a6aMczgcOHjwIAoKChAXF+eropKX/DkPb2J/qWs1hqu/1LUaw2AwYGhoCBkZGQB+rcZw\n9aX+y5iw5plrJNFoNKK8vByHDh3yiHE6nTh8+DBOnDiB9PT0BSil8tjtdhw5cgR9fX0IDQ1FdXU1\nVq9ePSmmvb0dOTk5cDgcWLZsGS5evBiw2x2npqYiNTV10nuXL192/zzTTfFfxlHCeWY0GlFfX4/w\n8HA0NDS4125ZLBacPHkSAFBdXY0nT56grKwMMTExiImJwePHj2edDuHicDiwceNGnDlzxi/XFAh8\nPfrKeXgK4W3bUdiH5VMGg8E9MlVUVCR5eXnTxp49e1aOHTs27cjSYuTr0Vdv+pVKSkomdbpnZGT8\n4VXQDLzKQaxhBQiz2YysrCwAQFZWFmpqaqaMe/78OaxWK1JSUvxZvAVntVrdu2usW7cOVqt1xvjW\n1lY4nU6EhYV59f3zNQ+PfMzbzCasYfmUN9MhxsbGJD4+Xj58+CB37txZ8BqWzWaTpKQk0Wg0kpSU\nJHa7fdrYwcFB2bBhw4xl5ujrP42jhIEmKSkJnz9/9ni/oKBg0vGSJUum3BSvtLQUqampAfPUYFe/\nktFohMlkgslkwtWrV6eMzc/Px969e2f8Po6+0myYsPzob/8hW1pa0NTUhNLSUgwNDcHpdGLlypUL\n1jQxm81obGwE8KsZ63rM2v+5mrEHDhyY8zo1jr4SADYJA8X58+cndbobDIYZ4wOhSejPZuyXL18k\nMTFRNBqN6HQ6sdlsIiLS1tbmXuZSWVkpQUFBsm3bNvfr5cuXczof+R2bhEpiNBqh1+tx+/ZtbNq0\nCdXV1QB+TYe4efMmbt26tSDlCpRmrEqlwqNHjzze37Fjh/tvk5mZiczMzL86DwW2RbmnO/lHZGQk\nGhsb3c3YhIQEvHnzZlLM8ePH0dTUhKVLl7qbsadPn+YIG/3fv/sQCvIPg8EAlUrl7nS32+24du3a\ntPFlZWWwWCwoKSnxYylJIbxKWJyHRXPmzax+ovnEGhYRBQLWsIhocWHCIiLFYMIiIsVgwiIixWDC\nIiLFYMIiIsVgwiIixWDCIiLFYMIiIsVgwiIixWDCIiLFYMIiIsVgwiIixWDCIiLFYMIiIsVgwiIi\nxWDCIiLF+NOn5ni1KyARkS+whkVEisGERUSKwYRFRIrBhEVEisGERUSKwYRFRIrBhEVEisGERUSK\nwYRFRIrBhEVEivEfD4It0zup6B8AAAAASUVORK5CYII=\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "2 (-0.1694611712186425, 0.5825921777221892)\n", + "3 (-0.2628166483508216, 0.5641037192609575)\n", + "4 (-0.27953310001779486, 0.6531302498446571)\n" + ] + } + ], + "source": [ + "# Train the GAN.\n", + "for _ in range(n_epochs):\n", + " loss, duration = gan._single_epoch_train(all_pc_data, batch_size, noise_params)\n", + " epoch = int(gan.sess.run(gan.increment_epoch))\n", + " print epoch, loss\n", + "\n", + " if save_gan_model and epoch in saver_step:\n", + " checkpoint_path = osp.join(train_dir, MODEL_SAVER_ID)\n", + " gan.saver.save(gan.sess, checkpoint_path, global_step=gan.epoch)\n", + "\n", + " if save_synthetic_samples and epoch in saver_step:\n", + " syn_data = gan.generate(n_syn_samples, noise_params)\n", + " np.savez(osp.join(synthetic_data_out_dir, 'epoch_' + str(epoch)), syn_data)\n", + " for k in range(3): # plot three (synthetic) random examples.\n", + " plot_3d_point_cloud(syn_data[k][:, 0], syn_data[k][:, 1], syn_data[k][:, 2],\n", + " in_u_sphere=True)\n", + "\n", + " train_stats.append((epoch, ) + loss)" + ] + }, + { + "cell_type": "code", + "execution_count": 27, + "metadata": { + "collapsed": false, + "deletable": true, + "editable": true + }, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAEWCAYAAACaBstRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/h+qAO4r4UH7sAAAAABJRU5ErkJggg==\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "if plot_train_curve:\n", + " x = range(len(train_stats))\n", + " d_loss = [t[1] for t in train_stats]\n", + " g_loss = [t[2] for t in train_stats]\n", + " plt.plot(x, d_loss, '--')\n", + " plt.plot(x, g_loss)\n", + " plt.title('GAN training. (%s)' %(class_name))\n", + " plt.legend(['Discriminator', 'Generator'], loc=0)\n", + " \n", + " plt.tick_params(axis='x', which='both', bottom='off', top='off')\n", + " plt.tick_params(axis='y', which='both', left='off', right='off')\n", + " \n", + " plt.xlabel('Epochs.') \n", + " plt.ylabel('Loss.')" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "TensorFlow1", + "language": "python", + "name": "tf1" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 2 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython2", + "version": "2.7.12" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/notebooks/train_single_class_ae.ipynb b/notebooks/train_single_class_ae.ipynb index 842667b..af6ac28 100644 --- a/notebooks/train_single_class_ae.ipynb +++ b/notebooks/train_single_class_ae.ipynb @@ -16,28 +16,6 @@ "metadata": { "collapsed": false }, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Picking GPU 1\n" - ] - } - ], - "source": [ - "!! DELETE\n", - "from general_tools.notebook.gpu_utils import setup_one_gpu\n", - "GPU = 1\n", - "setup_one_gpu(GPU)" - ] - }, - { - "cell_type": "code", - "execution_count": 2, - "metadata": { - "collapsed": false - }, "outputs": [], "source": [ "import os.path as osp\n", @@ -49,12 +27,13 @@ "from latent_3d_points.src.in_out import snc_category_to_synth_id, create_dir, PointCloudDataSet, \\\n", " load_all_point_clouds_under_folder\n", "\n", - "from latent_3d_points.src.tf_utils import reset_tf_graph" + "from latent_3d_points.src.tf_utils import reset_tf_graph\n", + "from latent_3d_points.src.general_utils import plot_3d_point_cloud" ] }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 2, "metadata": { "collapsed": false }, @@ -74,7 +53,7 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": 3, "metadata": { "collapsed": false }, @@ -107,7 +86,7 @@ }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 4, "metadata": { "collapsed": false }, @@ -146,7 +125,7 @@ }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 5, "metadata": { "collapsed": true }, @@ -157,7 +136,7 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 6, "metadata": { "collapsed": true }, @@ -169,28 +148,7 @@ }, { "cell_type": "code", - "execution_count": 9, - "metadata": { - "collapsed": false - }, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "{'filter_sizes': [1], 'n_filters': [64, 128, 128, 256, 128], 'verbose': True, 'b_norm': True, 'strides': [1]}\n", - "{'b_norm_finish': False, 'verbose': True, 'b_norm': False, 'layer_sizes': [256, 256, 6144]}\n" - ] - } - ], - "source": [ - "print enc_args\n", - "print dec_args" - ] - }, - { - "cell_type": "code", - "execution_count": 10, + "execution_count": 12, "metadata": { "collapsed": false }, @@ -217,6 +175,30 @@ "conf.save(osp.join(train_dir, 'configuration'))" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "If you ran the above lines, you can reload a saved model like this:" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "load_pre_trained_ae = False\n", + "restore_epoch = 500\n", + "if load_pre_trained_ae:\n", + " conf = Conf.load(train_dir + '/configuration')\n", + " reset_tf_graph()\n", + " ae = PointNetAutoEncoder(conf.experiment_name, conf)\n", + " ae.restore_model(conf.train_dir, epoch=restore_epoch)" + ] + }, { "cell_type": "markdown", "metadata": {}, @@ -226,50 +208,11 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 16, "metadata": { "collapsed": false }, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Building Encoder\n", - "encoder_conv_layer_0 conv params = 256 bnorm params = 128\n", - "Tensor(\"single_class_ae_2/Relu:0\", shape=(?, 2048, 64), dtype=float32)\n", - "output size: 131072 \n", - "\n", - "encoder_conv_layer_1 conv params = 8320 bnorm params = 256\n", - "Tensor(\"single_class_ae_2/Relu_1:0\", shape=(?, 2048, 128), dtype=float32)\n", - "output size: 262144 \n", - "\n", - "encoder_conv_layer_2 conv params = 16512 bnorm params = 256\n", - "Tensor(\"single_class_ae_2/Relu_2:0\", shape=(?, 2048, 128), dtype=float32)\n", - "output size: 262144 \n", - "\n", - "encoder_conv_layer_3 conv params = 33024 bnorm params = 512\n", - "Tensor(\"single_class_ae_2/Relu_3:0\", shape=(?, 2048, 256), dtype=float32)\n", - "output size: 524288 \n", - "\n", - "encoder_conv_layer_4 conv params = 32896 bnorm params = 256\n", - "Tensor(\"single_class_ae_2/Relu_4:0\", shape=(?, 2048, 128), dtype=float32)\n", - "output size: 262144 \n", - "\n", - "Tensor(\"single_class_ae_2/Max:0\", shape=(?, 128), dtype=float32)\n", - "Building Decoder\n", - "decoder_fc_0 FC params = 33024 Tensor(\"single_class_ae_2/Relu_5:0\", shape=(?, 256), dtype=float32)\n", - "output size: 256 \n", - "\n", - "decoder_fc_1 FC params = 65792 Tensor(\"single_class_ae_2/Relu_6:0\", shape=(?, 256), dtype=float32)\n", - "output size: 256 \n", - "\n", - "decoder_fc_2 FC params = 1579008 Tensor(\"single_class_ae_2/decoder_fc_2/BiasAdd:0\", shape=(?, 6144), dtype=float32)\n", - "output size: 6144 \n", - "\n" - ] - } - ], + "outputs": [], "source": [ "reset_tf_graph()\n", "ae = PointNetAutoEncoder(conf.experiment_name, conf)" @@ -284,149 +227,11 @@ }, { "cell_type": "code", - "execution_count": 12, + "execution_count": 17, "metadata": { "collapsed": false }, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "('Epoch:', '0001', 'training time (minutes)=', '0.1416', 'loss=', '0.003298247')\n", - "INFO:tensorflow:../data/single_class_ae/models.ckpt-1 is not in all_model_checkpoint_paths. Manually adding it.\n", - "('Epoch:', '0002', 'training time (minutes)=', '0.1346', 'loss=', '0.001645958')\n", - "('Epoch:', '0003', 'training time (minutes)=', '0.1385', 'loss=', '0.001443318')\n", - "('Epoch:', '0004', 'training time (minutes)=', '0.1375', 'loss=', '0.001287838')\n", - "('Epoch:', '0005', 'training time (minutes)=', '0.1431', 'loss=', '0.001243037')\n", - "('Epoch:', '0006', 'training time (minutes)=', '0.1421', 'loss=', '0.001230350')\n", - "('Epoch:', '0007', 'training time (minutes)=', '0.1393', 'loss=', '0.001117009')\n", - "('Epoch:', '0008', 'training time (minutes)=', '0.1438', 'loss=', '0.001081870')\n", - "('Epoch:', '0009', 'training time (minutes)=', '0.1392', 'loss=', '0.001050571')\n", - "('Epoch:', '0010', 'training time (minutes)=', '0.1409', 'loss=', '0.001035428')\n", - "INFO:tensorflow:../data/single_class_ae/models.ckpt-10 is not in all_model_checkpoint_paths. Manually adding it.\n", - "('Epoch:', '0011', 'training time (minutes)=', '0.1506', 'loss=', '0.000998460')\n", - "('Epoch:', '0012', 'training time (minutes)=', '0.1413', 'loss=', '0.000984788')\n", - "('Epoch:', '0013', 'training time (minutes)=', '0.1506', 'loss=', '0.000969033')\n", - "('Epoch:', '0014', 'training time (minutes)=', '0.1441', 'loss=', '0.000966409')\n", - "('Epoch:', '0015', 'training time (minutes)=', '0.1450', 'loss=', '0.000922309')\n", - "('Epoch:', '0016', 'training time (minutes)=', '0.1397', 'loss=', '0.000911922')\n", - "('Epoch:', '0017', 'training time (minutes)=', '0.1445', 'loss=', '0.000921226')\n", - "('Epoch:', '0018', 'training time (minutes)=', '0.1424', 'loss=', '0.000889343')\n", - "('Epoch:', '0019', 'training time (minutes)=', '0.1427', 'loss=', '0.000888205')\n", - "('Epoch:', '0020', 'training time (minutes)=', '0.1441', 'loss=', '0.000875273')\n", - "INFO:tensorflow:../data/single_class_ae/models.ckpt-20 is not in all_model_checkpoint_paths. Manually adding it.\n", - "('Epoch:', '0021', 'training time (minutes)=', '0.1451', 'loss=', '0.000863822')\n", - "('Epoch:', '0022', 'training time (minutes)=', '0.1432', 'loss=', '0.000864306')\n", - "('Epoch:', '0023', 'training time (minutes)=', '0.1468', 'loss=', '0.000865318')\n", - "('Epoch:', '0024', 'training time (minutes)=', '0.1444', 'loss=', '0.000819852')\n", - "('Epoch:', '0025', 'training time (minutes)=', '0.1494', 'loss=', '0.000829801')\n", - "('Epoch:', '0026', 'training time (minutes)=', '0.1426', 'loss=', '0.000838800')\n", - "('Epoch:', '0027', 'training time (minutes)=', '0.1465', 'loss=', '0.000839270')\n", - "('Epoch:', '0028', 'training time (minutes)=', '0.1435', 'loss=', '0.000797047')\n", - "('Epoch:', '0029', 'training time (minutes)=', '0.1408', 'loss=', '0.000836910')\n", - "('Epoch:', '0030', 'training time (minutes)=', '0.1464', 'loss=', '0.000788118')\n", - "INFO:tensorflow:../data/single_class_ae/models.ckpt-30 is not in all_model_checkpoint_paths. Manually adding it.\n", - "('Epoch:', '0031', 'training time (minutes)=', '0.1431', 'loss=', '0.000789107')\n", - "('Epoch:', '0032', 'training time (minutes)=', '0.1440', 'loss=', '0.000801768')\n", - "('Epoch:', '0033', 'training time (minutes)=', '0.1431', 'loss=', '0.000778435')\n", - "('Epoch:', '0034', 'training time (minutes)=', '0.1452', 'loss=', '0.000743628')\n", - "('Epoch:', '0035', 'training time (minutes)=', '0.1412', 'loss=', '0.000771024')\n", - "('Epoch:', '0036', 'training time (minutes)=', '0.1453', 'loss=', '0.000777234')\n", - "('Epoch:', '0037', 'training time (minutes)=', '0.1452', 'loss=', '0.000751956')\n", - "('Epoch:', '0038', 'training time (minutes)=', '0.1401', 'loss=', '0.000772422')\n", - "('Epoch:', '0039', 'training time (minutes)=', '0.1457', 'loss=', '0.000757981')\n", - "('Epoch:', '0040', 'training time (minutes)=', '0.1433', 'loss=', '0.000742244')\n", - "INFO:tensorflow:../data/single_class_ae/models.ckpt-40 is not in all_model_checkpoint_paths. Manually adding it.\n", - "('Epoch:', '0041', 'training time (minutes)=', '0.1452', 'loss=', '0.000728385')\n", - "('Epoch:', '0042', 'training time (minutes)=', '0.1426', 'loss=', '0.000726597')\n", - "('Epoch:', '0043', 'training time (minutes)=', '0.1435', 'loss=', '0.000729684')\n", - "('Epoch:', '0044', 'training time (minutes)=', '0.1477', 'loss=', '0.000727496')\n", - "('Epoch:', '0045', 'training time (minutes)=', '0.1386', 'loss=', '0.000737438')\n", - "('Epoch:', '0046', 'training time (minutes)=', '0.1450', 'loss=', '0.000736978')\n", - "('Epoch:', '0047', 'training time (minutes)=', '0.1448', 'loss=', '0.000711031')\n", - "('Epoch:', '0048', 'training time (minutes)=', '0.1424', 'loss=', '0.000707406')\n", - "('Epoch:', '0049', 'training time (minutes)=', '0.1446', 'loss=', '0.000726043')\n", - "('Epoch:', '0050', 'training time (minutes)=', '0.1419', 'loss=', '0.000704496')\n", - "INFO:tensorflow:../data/single_class_ae/models.ckpt-50 is not in all_model_checkpoint_paths. Manually adding it.\n", - "('Epoch:', '0051', 'training time (minutes)=', '0.1438', 'loss=', '0.000712091')\n", - "('Epoch:', '0052', 'training time (minutes)=', '0.1432', 'loss=', '0.000730910')\n", - "('Epoch:', '0053', 'training time (minutes)=', '0.1410', 'loss=', '0.000685905')\n", - "('Epoch:', '0054', 'training time (minutes)=', '0.1468', 'loss=', '0.000702954')\n", - "('Epoch:', '0055', 'training time (minutes)=', '0.1427', 'loss=', '0.000678213')\n", - "('Epoch:', '0056', 'training time (minutes)=', '0.1405', 'loss=', '0.000686704')\n", - "('Epoch:', '0057', 'training time (minutes)=', '0.1389', 'loss=', '0.000681744')\n", - "('Epoch:', '0058', 'training time (minutes)=', '0.1512', 'loss=', '0.000691520')\n", - "('Epoch:', '0059', 'training time (minutes)=', '0.1416', 'loss=', '0.000685650')\n", - "('Epoch:', '0060', 'training time (minutes)=', '0.1415', 'loss=', '0.000690602')\n", - "INFO:tensorflow:../data/single_class_ae/models.ckpt-60 is not in all_model_checkpoint_paths. Manually adding it.\n", - "('Epoch:', '0061', 'training time (minutes)=', '0.1406', 'loss=', '0.000691447')\n", - "('Epoch:', '0062', 'training time (minutes)=', '0.1405', 'loss=', '0.000669356')\n", - "('Epoch:', '0063', 'training time (minutes)=', '0.1419', 'loss=', '0.000678524')\n", - "('Epoch:', '0064', 'training time (minutes)=', '0.1413', 'loss=', '0.000682174')\n", - "('Epoch:', '0065', 'training time (minutes)=', '0.1433', 'loss=', '0.000670896')\n", - "('Epoch:', '0066', 'training time (minutes)=', '0.1413', 'loss=', '0.000681451')\n", - "('Epoch:', '0067', 'training time (minutes)=', '0.1487', 'loss=', '0.000667460')\n", - "('Epoch:', '0068', 'training time (minutes)=', '0.1436', 'loss=', '0.000655880')\n", - "('Epoch:', '0069', 'training time (minutes)=', '0.1431', 'loss=', '0.000650574')\n", - "('Epoch:', '0070', 'training time (minutes)=', '0.1422', 'loss=', '0.000668619')\n", - "INFO:tensorflow:../data/single_class_ae/models.ckpt-70 is not in all_model_checkpoint_paths. Manually adding it.\n", - "('Epoch:', '0071', 'training time (minutes)=', '0.1402', 'loss=', '0.000642797')\n", - "('Epoch:', '0072', 'training time (minutes)=', '0.1428', 'loss=', '0.000662679')\n", - "('Epoch:', '0073', 'training time (minutes)=', '0.1423', 'loss=', '0.000665245')\n", - "('Epoch:', '0074', 'training time (minutes)=', '0.1452', 'loss=', '0.000647563')\n", - "('Epoch:', '0075', 'training time (minutes)=', '0.1467', 'loss=', '0.000634363')\n", - "('Epoch:', '0076', 'training time (minutes)=', '0.1422', 'loss=', '0.000644973')\n", - "('Epoch:', '0077', 'training time (minutes)=', '0.1452', 'loss=', '0.000635988')\n", - "('Epoch:', '0078', 'training time (minutes)=', '0.1468', 'loss=', '0.000630166')\n", - "('Epoch:', '0079', 'training time (minutes)=', '0.1433', 'loss=', '0.000645955')\n", - "('Epoch:', '0080', 'training time (minutes)=', '0.1414', 'loss=', '0.000634762')\n", - "INFO:tensorflow:../data/single_class_ae/models.ckpt-80 is not in all_model_checkpoint_paths. Manually adding it.\n", - "('Epoch:', '0081', 'training time (minutes)=', '0.1445', 'loss=', '0.000634045')\n", - "('Epoch:', '0082', 'training time (minutes)=', '0.1475', 'loss=', '0.000624762')\n", - "('Epoch:', '0083', 'training time (minutes)=', '0.1482', 'loss=', '0.000618184')\n", - "('Epoch:', '0084', 'training time (minutes)=', '0.1415', 'loss=', '0.000617901')\n", - "('Epoch:', '0085', 'training time (minutes)=', '0.1467', 'loss=', '0.000614790')\n", - "('Epoch:', '0086', 'training time (minutes)=', '0.1441', 'loss=', '0.000635329')\n", - "('Epoch:', '0087', 'training time (minutes)=', '0.1429', 'loss=', '0.000635824')\n", - "('Epoch:', '0088', 'training time (minutes)=', '0.1483', 'loss=', '0.000622333')\n", - "('Epoch:', '0089', 'training time (minutes)=', '0.1430', 'loss=', '0.000626940')\n", - "('Epoch:', '0090', 'training time (minutes)=', '0.1435', 'loss=', '0.000618810')\n", - "INFO:tensorflow:../data/single_class_ae/models.ckpt-90 is not in all_model_checkpoint_paths. Manually adding it.\n", - "('Epoch:', '0091', 'training time (minutes)=', '0.1412', 'loss=', '0.000624059')\n", - "('Epoch:', '0092', 'training time (minutes)=', '0.1445', 'loss=', '0.000628070')\n", - "('Epoch:', '0093', 'training time (minutes)=', '0.1414', 'loss=', '0.000619484')\n", - "('Epoch:', '0094', 'training time (minutes)=', '0.1418', 'loss=', '0.000616793')\n", - "('Epoch:', '0095', 'training time (minutes)=', '0.1437', 'loss=', '0.000612060')\n", - "('Epoch:', '0096', 'training time (minutes)=', '0.1449', 'loss=', '0.000602396')\n", - "('Epoch:', '0097', 'training time (minutes)=', '0.1396', 'loss=', '0.000616112')\n", - "('Epoch:', '0098', 'training time (minutes)=', '0.1393', 'loss=', '0.000612572')\n", - "('Epoch:', '0099', 'training time (minutes)=', '0.1412', 'loss=', '0.000605171')\n", - "('Epoch:', '0100', 'training time (minutes)=', '0.1439', 'loss=', '0.000611150')\n", - "INFO:tensorflow:../data/single_class_ae/models.ckpt-100 is not in all_model_checkpoint_paths. Manually adding it.\n", - "('Epoch:', '0101', 'training time (minutes)=', '0.1397', 'loss=', '0.000608574')\n" - ] - }, - { - "ename": "KeyboardInterrupt", - "evalue": "", - "output_type": "error", - "traceback": [ - "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", - "\u001b[0;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)", - "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m()\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[0mbuf_size\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m1\u001b[0m \u001b[0;31m# Make 'training_stats' file to flush each output line regarding training.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2\u001b[0m \u001b[0mfout\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mopen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mosp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mjoin\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mconf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_dir\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'train_stats.txt'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'a'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbuf_size\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 3\u001b[0;31m \u001b[0mtrain_stats\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mae\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mall_pc_data\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mconf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlog_file\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mfout\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 4\u001b[0m \u001b[0mfout\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mclose\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m/orions4-zfs/projects/optas/Testing/optas/latent_3d_points/src/autoencoder.pyc\u001b[0m in \u001b[0;36mtrain\u001b[0;34m(self, train_data, configuration, log_file, held_out_data)\u001b[0m\n\u001b[1;32m 181\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 182\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0m_\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mxrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mc\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtraining_epochs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 183\u001b[0;31m \u001b[0mloss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mduration\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_single_epoch_train\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrain_data\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mc\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 184\u001b[0m \u001b[0mepoch\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msess\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mepoch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0massign_add\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstant\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m1.0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 185\u001b[0m \u001b[0mstats\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mepoch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mloss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mduration\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m/orions4-zfs/projects/optas/Testing/optas/latent_3d_points/src/point_net_ae.pyc\u001b[0m in \u001b[0;36m_single_epoch_train\u001b[0;34m(self, train_data, configuration, only_fw)\u001b[0m\n\u001b[1;32m 126\u001b[0m \u001b[0m_\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mloss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbatch_i\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moriginal_data\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 127\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 128\u001b[0;31m \u001b[0m_\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mloss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbatch_i\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 129\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 130\u001b[0m \u001b[0;31m# Compute average loss\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m/orions4-zfs/projects/optas/Testing/optas/latent_3d_points/src/autoencoder.pyc\u001b[0m in \u001b[0;36mpartial_fit\u001b[0;34m(self, X, GT)\u001b[0m\n\u001b[1;32m 129\u001b[0m \u001b[0m_\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mloss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mrecon\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msess\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_step\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mx_reconstr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m{\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgt\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mGT\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 130\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 131\u001b[0;31m \u001b[0m_\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mloss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mrecon\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msess\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_step\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mx_reconstr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m{\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mX\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 132\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 133\u001b[0m \u001b[0mis_training\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mFalse\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msession\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msess\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m/orions4-zfs/projects/optas/Virt_Env/tf_1.3/local/lib/python2.7/site-packages/tensorflow/python/client/session.pyc\u001b[0m in \u001b[0;36mrun\u001b[0;34m(self, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m 893\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 894\u001b[0m result = self._run(None, fetches, feed_dict, options_ptr,\n\u001b[0;32m--> 895\u001b[0;31m run_metadata_ptr)\n\u001b[0m\u001b[1;32m 896\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mrun_metadata\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 897\u001b[0m \u001b[0mproto_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf_session\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTF_GetBuffer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrun_metadata_ptr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m/orions4-zfs/projects/optas/Virt_Env/tf_1.3/local/lib/python2.7/site-packages/tensorflow/python/client/session.pyc\u001b[0m in \u001b[0;36m_run\u001b[0;34m(self, handle, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m 1122\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mfinal_fetches\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0mfinal_targets\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mhandle\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mfeed_dict_tensor\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1123\u001b[0m results = self._do_run(handle, final_targets, final_fetches,\n\u001b[0;32m-> 1124\u001b[0;31m feed_dict_tensor, options, run_metadata)\n\u001b[0m\u001b[1;32m 1125\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1126\u001b[0m \u001b[0mresults\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m/orions4-zfs/projects/optas/Virt_Env/tf_1.3/local/lib/python2.7/site-packages/tensorflow/python/client/session.pyc\u001b[0m in \u001b[0;36m_do_run\u001b[0;34m(self, handle, target_list, fetch_list, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m 1319\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mhandle\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1320\u001b[0m return self._do_call(_run_fn, self._session, feeds, fetches, targets,\n\u001b[0;32m-> 1321\u001b[0;31m options, run_metadata)\n\u001b[0m\u001b[1;32m 1322\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1323\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_prun_fn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_session\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhandle\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeeds\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetches\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m/orions4-zfs/projects/optas/Virt_Env/tf_1.3/local/lib/python2.7/site-packages/tensorflow/python/client/session.pyc\u001b[0m in \u001b[0;36m_do_call\u001b[0;34m(self, fn, *args)\u001b[0m\n\u001b[1;32m 1325\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m_do_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1326\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1327\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1328\u001b[0m \u001b[0;32mexcept\u001b[0m \u001b[0merrors\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mOpError\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1329\u001b[0m \u001b[0mmessage\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcompat\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mas_text\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmessage\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m/orions4-zfs/projects/optas/Virt_Env/tf_1.3/local/lib/python2.7/site-packages/tensorflow/python/client/session.pyc\u001b[0m in \u001b[0;36m_run_fn\u001b[0;34m(session, feed_dict, fetch_list, target_list, options, run_metadata)\u001b[0m\n\u001b[1;32m 1304\u001b[0m return tf_session.TF_Run(session, options,\n\u001b[1;32m 1305\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetch_list\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtarget_list\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1306\u001b[0;31m status, run_metadata)\n\u001b[0m\u001b[1;32m 1307\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1308\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m_prun_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msession\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhandle\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetch_list\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;31mKeyboardInterrupt\u001b[0m: " - ] - } - ], + "outputs": [], "source": [ "buf_size = 1 # Make 'training_stats' file to flush each output line regarding training.\n", "fout = open(osp.join(conf.train_dir, 'train_stats.txt'), 'a', buf_size)\n", @@ -443,14 +248,14 @@ }, { "cell_type": "code", - "execution_count": 13, + "execution_count": 9, "metadata": { "collapsed": false }, "outputs": [], "source": [ "feed_pc, feed_model_names, _ = all_pc_data.next_batch(10)\n", - "reconstructions = ae.reconstruct(feed_pc)\n", + "reconstructions = ae.reconstruct(feed_pc)[0]\n", "latent_codes = ae.transform(feed_pc)" ] }, @@ -462,6 +267,46 @@ "source": [ "Use any plotting mechanism such as matplotlib to visualize the results." ] + }, + { + "cell_type": "code", + "execution_count": 27, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAASwAAAElCAYAAABect+9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAASwAAAElCAYAAABect+9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "i = 2\n", + "plot_3d_point_cloud(reconstructions[i][:, 0], \n", + " reconstructions[i][:, 1], \n", + " reconstructions[i][:, 2], in_u_sphere=True);\n", + "\n", + "i = 4\n", + "plot_3d_point_cloud(reconstructions[i][:, 0], \n", + " reconstructions[i][:, 1], \n", + " reconstructions[i][:, 2], in_u_sphere=True);" + ] } ], "metadata": { diff --git a/src/autoencoder.py b/src/autoencoder.py index 52bc894..838efcf 100755 --- a/src/autoencoder.py +++ b/src/autoencoder.py @@ -13,9 +13,7 @@ from . in_out import create_dir, pickle_data, unpickle_data from . general_utils import apply_augmentations, iterate_in_chunks -from . neural_net import Neural_Net - -model_saver_id = 'models.ckpt' +from . neural_net import Neural_Net, MODEL_SAVER_ID class Configuration(): @@ -103,18 +101,7 @@ def __init__(self, name, graph, configuration): self.gt = tf.placeholder(tf.float32, out_shape) else: self.gt = self.x - - def restore_model(self, model_path, epoch, verbose=False): - '''Restore all the variables of a saved auto-encoder model. - ''' - self.saver.restore(self.sess, osp.join(model_path, model_saver_id + '-' + str(int(epoch)))) - - if self.epoch.eval(session=self.sess) != epoch: - warnings.warn('Loaded model\'s epoch doesn\'t match the requested one.') - else: - if verbose: - print('Model restored in epoch {0}.'.format(epoch)) - + def partial_fit(self, X, GT=None): '''Trains the model with mini-batches of input data. If GT is not None, then the reconstruction loss compares the output of the net that is fed X, with the GT. @@ -143,7 +130,7 @@ def reconstruct(self, X, GT=None, compute_loss=True): if compute_loss: loss = self.loss else: - loss = tf.no_op() + loss = self.no_op if GT is None: return self.sess.run((self.x_reconstr, loss), feed_dict={self.x: X}) @@ -181,7 +168,7 @@ def train(self, train_data, configuration, log_file=None, held_out_data=None): for _ in xrange(c.training_epochs): loss, duration = self._single_epoch_train(train_data, c) - epoch = int(self.sess.run(self.epoch.assign_add(tf.constant(1.0)))) + epoch = int(self.sess.run(self.increment_epoch)) stats.append((epoch, loss, duration)) if epoch % c.loss_display_step == 0: @@ -191,7 +178,7 @@ def train(self, train_data, configuration, log_file=None, held_out_data=None): # Save the models checkpoint periodically. if c.saver_step is not None and (epoch % c.saver_step == 0 or epoch - 1 == 0): - checkpoint_path = osp.join(c.train_dir, model_saver_id) + checkpoint_path = osp.join(c.train_dir, MODEL_SAVER_ID) self.saver.save(self.sess, checkpoint_path, global_step=self.epoch) if c.exists_and_is_not_none('summary_step') and (epoch % c.summary_step == 0 or epoch - 1 == 0): @@ -236,37 +223,7 @@ def evaluate(self, in_data, configuration, ret_pre_augmentation=False): return reconstructions, data_loss, np.squeeze(feed_data), ids, np.squeeze(original_data), pre_aug else: return reconstructions, data_loss, np.squeeze(feed_data), ids, np.squeeze(original_data) - - def evaluate_one_by_one(self, in_data, configuration): - '''Evaluates every data point separately to recover the loss on it. Thus, the batch_size = 1 making it - a slower than the 'evaluate' method. - ''' - - if self.is_denoising: - original_data, ids, feed_data = in_data.full_epoch_data(shuffle=False) - if feed_data is None: - feed_data = original_data - feed_data = apply_augmentations(feed_data, configuration) # This is a new copy of the batch. - else: - original_data, ids, _ = in_data.full_epoch_data(shuffle=False) - feed_data = apply_augmentations(original_data, configuration) - - n_examples = in_data.num_examples - assert(len(original_data) == n_examples) - - feed_data = np.expand_dims(feed_data, 1) - original_data = np.expand_dims(original_data, 1) - reconstructions = np.zeros([n_examples] + self.n_output) - losses = np.zeros([n_examples]) - - for i in xrange(n_examples): - if self.is_denoising: - reconstructions[i], losses[i] = self.reconstruct(feed_data[i], original_data[i]) - else: - reconstructions[i], losses[i] = self.reconstruct(feed_data[i]) - - return reconstructions, losses, np.squeeze(feed_data), ids, np.squeeze(original_data) - + def embedding_at_tensor(self, dataset, conf, feed_original=True, apply_augmentation=False, tensor_name='bottleneck'): ''' Observation: the NN-neighborhoods seem more reasonable when we do not apply the augmentation. @@ -299,8 +256,7 @@ def embedding_at_tensor(self, dataset, conf, feed_original=True, apply_augmentat embedding = np.vstack(embedding) return feed, embedding, ids - - + def get_latent_codes(self, pclouds, batch_size=100): ''' Convenience wrapper of self.transform to get the latent (bottle-neck) codes for a set of input point clouds. diff --git a/src/general_utils.py b/src/general_utils.py index 48b2ab0..ccc0459 100644 --- a/src/general_utils.py +++ b/src/general_utils.py @@ -6,6 +6,8 @@ import numpy as np from numpy.linalg import norm +import matplotlib.pylab as plt +from mpl_toolkits.mplot3d import Axes3D def rand_rotation_matrix(deflection=1.0, seed=None): @@ -104,3 +106,41 @@ def unit_cube_grid_point_cloud(resolution, clip_sphere=False): grid = grid[norm(grid, axis=1) <= 0.5] return grid, spacing + +def plot_3d_point_cloud(x, y, z, show=True, show_axis=True, in_u_sphere=False, marker='.', s=8, alpha=.8, figsize=(5, 5), elev=10, azim=240, axis=None, title=None, *args, **kwargs): + + if axis is None: + fig = plt.figure(figsize=figsize) + ax = fig.add_subplot(111, projection='3d') + else: + ax = axis + fig = axis + + if title is not None: + plt.title(title) + + sc = ax.scatter(x, y, z, marker=marker, s=s, alpha=alpha, *args, **kwargs) + ax.view_init(elev=elev, azim=azim) + + if in_u_sphere: + ax.set_xlim3d(-0.5, 0.5) + ax.set_ylim3d(-0.5, 0.5) + ax.set_zlim3d(-0.5, 0.5) + else: + miv = 0.7 * np.min([np.min(x), np.min(y), np.min(z)]) # Multiply with 0.7 to squeeze free-space. + mav = 0.7 * np.max([np.max(x), np.max(y), np.max(z)]) + ax.set_xlim(miv, mav) + ax.set_ylim(miv, mav) + ax.set_zlim(miv, mav) + plt.tight_layout() + + if not show_axis: + plt.axis('off') + + if 'c' in kwargs: + plt.colorbar(sc) + + if show: + plt.show() + + return fig \ No newline at end of file diff --git a/src/generators_discriminators.py b/src/generators_discriminators.py index 87fa030..791b522 100755 --- a/src/generators_discriminators.py +++ b/src/generators_discriminators.py @@ -14,7 +14,7 @@ from . tf_utils import expand_scope_by_name -def mlp_discriminator(in_signal, non_linearity=tf.nn.relu, reuse=False, scope=None, b_norm=[True], dropout_prob=None): +def mlp_discriminator(in_signal, non_linearity=tf.nn.relu, reuse=False, scope=None, b_norm=True, dropout_prob=None): ''' used in nips submission. ''' encoder_args = {'n_filters': [64, 128, 256, 256, 512], 'filter_sizes': [1, 1, 1, 1, 1], 'strides': [1, 1, 1, 1, 1]} @@ -32,7 +32,7 @@ def mlp_discriminator(in_signal, non_linearity=tf.nn.relu, reuse=False, scope=No return d_prob, d_logit -def point_cloud_generator(z, pc_dims, layer_sizes=[64, 128, 512, 1024], non_linearity=tf.nn.relu, b_norm=[False], b_norm_last=False, dropout_prob=None): +def point_cloud_generator(z, pc_dims, layer_sizes=[64, 128, 512, 1024], non_linearity=tf.nn.relu, b_norm=False, b_norm_last=False, dropout_prob=None): ''' used in nips submission. ''' @@ -42,6 +42,7 @@ def point_cloud_generator(z, pc_dims, layer_sizes=[64, 128, 512, 1024], non_line out_signal = decoder_with_fc_only(z, layer_sizes=layer_sizes, non_linearity=non_linearity, b_norm=b_norm) out_signal = non_linearity(out_signal) + if dropout_prob is not None: out_signal = dropout(out_signal, dropout_prob) @@ -70,33 +71,33 @@ def convolutional_discriminator(in_signal, non_linearity=tf.nn.relu, return d_prob, d_logit -def latent_code_generator(z, out_dim, layer_sizes=[64, 128], b_norm=[False]): +def latent_code_generator(z, out_dim, layer_sizes=[64, 128], b_norm=False): layer_sizes = layer_sizes + out_dim out_signal = decoder_with_fc_only(z, layer_sizes=layer_sizes, b_norm=b_norm) out_signal = tf.nn.relu(out_signal) return out_signal -def latent_code_discriminator(in_singnal, layer_sizes=[64, 128, 256, 256, 512], b_norm=[False], non_linearity=tf.nn.relu, reuse=False, scope=None): +def latent_code_discriminator(in_singnal, layer_sizes=[64, 128, 256, 256, 512], b_norm=False, non_linearity=tf.nn.relu, reuse=False, scope=None): layer_sizes = layer_sizes + [1] d_logit = decoder_with_fc_only(in_singnal, layer_sizes=layer_sizes, non_linearity=non_linearity, b_norm=b_norm, reuse=reuse, scope=scope) d_prob = tf.nn.sigmoid(d_logit) return d_prob, d_logit -def latent_code_discriminator_two_layers(in_singnal, layer_sizes=[256, 512], b_norm=[False], non_linearity=tf.nn.relu, reuse=False, scope=None): - ''' used in nips submission. +def latent_code_discriminator_two_layers(in_signal, layer_sizes=[256, 512], b_norm=False, non_linearity=tf.nn.relu, reuse=False, scope=None): + ''' Used in ICML submission. ''' layer_sizes = layer_sizes + [1] - d_logit = decoder_with_fc_only(in_singnal, layer_sizes=layer_sizes, non_linearity=non_linearity, b_norm=b_norm, reuse=reuse, scope=scope) + d_logit = decoder_with_fc_only(in_signal, layer_sizes=layer_sizes, non_linearity=non_linearity, b_norm=b_norm, reuse=reuse, scope=scope) d_prob = tf.nn.sigmoid(d_logit) return d_prob, d_logit -def latent_code_generator_two_layers(z, out_dim, layer_sizes=[128], b_norm=[False]): - ''' used in nips submission. +def latent_code_generator_two_layers(z, out_dim, layer_sizes=[128], b_norm=False): + ''' Used in ICML submission. ''' layer_sizes = layer_sizes + out_dim out_signal = decoder_with_fc_only(z, layer_sizes=layer_sizes, b_norm=b_norm) - out_signal = tf.nn.relu(out_signal) # I could have added batch-norm before relu here. + out_signal = tf.nn.relu(out_signal) return out_signal diff --git a/src/neural_net.py b/src/neural_net.py index f328cd4..194ca2b 100755 --- a/src/neural_net.py +++ b/src/neural_net.py @@ -22,7 +22,21 @@ def __init__(self, name, graph): with tf.variable_scope(name): with tf.device('/cpu:0'): self.epoch = tf.get_variable('epoch', [], initializer=tf.constant_initializer(0), trainable=False) + self.increment_epoch = self.epoch.assign_add(tf.constant(1.0)) + + self.no_op = tf.no_op() def is_training(self): is_training_op = self.graph.get_collection('is_training') return self.sess.run(is_training_op)[0] + + def restore_model(self, model_path, epoch, verbose=False): + '''Restore all the variables of a saved model. + ''' + self.saver.restore(self.sess, osp.join(model_path, MODEL_SAVER_ID + '-' + str(int(epoch)))) + + if self.epoch.eval(session=self.sess) != epoch: + warnings.warn('Loaded model\'s epoch doesn\'t match the requested one.') + else: + if verbose: + print('Model restored in epoch {0}.'.format(epoch)) diff --git a/src/vanilla_gan.py b/src/vanilla_gan.py index 402e2a8..d023f17 100755 --- a/src/vanilla_gan.py +++ b/src/vanilla_gan.py @@ -1,5 +1,5 @@ ''' -Created on Apr 27, 2017 +Created on 2018 Author: Achlioptas Panos (Github ID: optas) ''' @@ -57,10 +57,6 @@ def generator_noise_distribution(self, n_samples, ndims, mu, sigma): return np.random.normal(mu, sigma, (n_samples, ndims)) def _single_epoch_train(self, train_data, batch_size, noise_params): - ''' - see: http://blog.aylien.com/introduction-generative-adversarial-networks-code-tensorflow/ - http://wiseodd.github.io/techblog/2016/09/17/gan-tensorflow/ - ''' n_examples = train_data.num_examples epoch_loss_d = 0. epoch_loss_g = 0. diff --git a/src/w_gan_gp.py b/src/w_gan_gp.py index f41c052..54ed7dc 100755 --- a/src/w_gan_gp.py +++ b/src/w_gan_gp.py @@ -1,5 +1,5 @@ ''' -Created on May 22, 2017 +Created on May 22, 2018 Author: Achlioptas Panos (Github ID: optas) '''