from tvtk.api import tvtk, write_data import sys sys.path.append('../') import model.flow_net as flow_net from inputs.flow_data_queue import Sailfish_data from utils.experiment_manager import make_checkpoint_path from utils.boundary_utils import get_random_params import matplotlib.pyplot as plt FLAGS = tf.app.flags.FLAGS FLOW_DIR = make_checkpoint_path(FLAGS.base_dir_flow, FLAGS, network="flow") BOUNDARY_DIR = make_checkpoint_path(FLAGS.base_dir_boundary_flow, FLAGS, network="boundary") print(BOUNDARY_DIR) print(FLOW_DIR) shape = FLAGS.shape.split('x') shape = map(int, shape) def evaluate(): """Run Eval once. """ with tf.Session() as sess: # Make image placeholder param_inputs, _ = flow_net.inputs_boundary(FLAGS.nr_boundary_params, 1, shape) # Make boundary
FLAGS = tf.app.flags.FLAGS tf.app.flags.DEFINE_string('base_dir', '../checkpoints', """dir to store trained net """) tf.app.flags.DEFINE_integer('batch_size', 8, """ training batch size """) tf.app.flags.DEFINE_integer('max_steps', 500000, """ max number of steps to train """) tf.app.flags.DEFINE_float('keep_prob', 0.7, """ keep probability for dropout """) tf.app.flags.DEFINE_float('learning_rate', 1e-4, """ keep probability for dropout """) tf.app.flags.DEFINE_bool('display_test', True, """ display the test images """) tf.app.flags.DEFINE_string('test_set', "car", """ either car or random """) TEST_DIR = make_checkpoint_path(FLAGS.base_dir, FLAGS) def tryint(s): try: return int(s) except: return s def alphanum_key(s): return [tryint(c) for c in re.split('([0-9]+)', s)] def evaluate(): """Run Eval once.
import sys sys.path.append('../') import model.flow_net as flow_net from inputs.flow_data_queue import Sailfish_data from utils.experiment_manager import make_checkpoint_path from model.pressure import calc_force from model.velocity_norm import calc_velocity_norm import matplotlib.pyplot as plt from tqdm import * FLAGS = tf.app.flags.FLAGS FLOW_DIR = make_checkpoint_path(FLAGS.base_dir_flow, FLAGS, network="flow") shape = FLAGS.shape.split('x') shape = map(int, shape) batch_size = 10 def tryint(s): try: return int(s) except: return s def alphanum_key(s):
import matplotlib.pyplot as plt #SMALL_SIZE = 12 #MEDIUM_SIZE = 16 #BIGGER_SIZE = 18 #plt.rc('font', size=MEDIUM_SIZE) # controls default text sizes #plt.rc('axes', titlesize=MEDIUM_SIZE) # fontsize of the axes title #plt.rc('axes', labelsize=MEDIUM_SIZE) # fontsize of the x and y labels #plt.rc('legend', fontsize=SMALL_SIZE) # legend fontsize #plt.rc('figure', titlesize=BIGGER_SIZE) # fontsize of the figure title FLAGS = tf.app.flags.FLAGS FLOW_DIR = make_checkpoint_path(FLAGS.base_dir_heat, FLAGS, network="heat") BOUNDARY_DIR = make_checkpoint_path(FLAGS.base_dir_boundary_heat, FLAGS, network="boundary") print(FLOW_DIR) print(BOUNDARY_DIR) shape = FLAGS.shape.split('x') shape = map(int, shape) batch_size = 1 def evaluate(): """Run Eval once. Args:
import numpy as np import tensorflow as tf import sys sys.path.append('../') import model.flow_net as flow_net from inputs.vtk_data import VTK_data from inputs.boundary_data_queue import Boundary_data from utils.experiment_manager import make_checkpoint_path import matplotlib.pyplot as plt FLAGS = tf.app.flags.FLAGS TRAIN_DIR = make_checkpoint_path(FLAGS.base_dir_boundary_flow, FLAGS, network="boundary") shape = FLAGS.dims * [FLAGS.obj_size] def train(): """Train ring_net for a number of steps.""" with tf.Graph().as_default(): # global step counter global_step = tf.get_variable('global_step', [], initializer=tf.constant_initializer(0), trainable=False) # make inputs input_dims = FLAGS.nr_boundary_params
import numpy as np import tensorflow as tf import sys sys.path.append('../') import model.flow_net as flow_net from inputs.vtk_data import VTK_data from utils.experiment_manager import make_checkpoint_path import matplotlib.pyplot as plt FLAGS = tf.app.flags.FLAGS BOUNDARY_DIR = make_checkpoint_path(FLAGS.base_dir_boundary, FLAGS, network="boundary") # video init shape = FLAGS.shape.split('x') shape = map(int, shape) batch_size = 1 def tryint(s): try: return int(s) except: return s
import tensorflow as tf import sys sys.path.append('../') import model.flow_net as flow_net from inputs.heat_data_queue import Heat_Sink_data from utils.experiment_manager import make_checkpoint_path from model.optimizer import * from tqdm import * import matplotlib.pyplot as plt FLAGS = tf.app.flags.FLAGS TRAIN_DIR = make_checkpoint_path(FLAGS.base_dir_heat, FLAGS, network="heat") shape = FLAGS.shape.split('x') shape = map(int, shape) def train(): """Train ring_net for a number of steps.""" with tf.Graph().as_default(): # store grad and loss values grads = [] loss_gen = [] # global step counter global_step = tf.get_variable('global_step', [], initializer=tf.constant_initializer(0), trainable=False)