main.py

import argparse
import numpy as np
import time
from data_loader import load_data, load_new_data
from train import train
import os
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '0'
os.environ["CUDA_VISIBLE_DEVICES"] = "1"



parser = argparse.ArgumentParser()

parser.add_argument('--dataset', type=str, default='ten_week', help='which dataset to use')
parser.add_argument('--title_len', type=int, default=10, help='the max length of title')
parser.add_argument('--session_len', type=int, default=10, help='the max length of session')
parser.add_argument('--aggregator', type=str, default='neighbor', help='which aggregator to use')
parser.add_argument('--n_epochs', type=int, default=1, help='the number of epochs')
parser.add_argument('--user_neighbor', type=int, default=30, help='the number of neighbors to be sampled')
parser.add_argument('--news_neighbor', type=int, default=10, help='the number of neighbors to be sampled')
parser.add_argument('--entity_neighbor', type=int, default=1, help='the number of neighbors to be sampled')
parser.add_argument('--user_dim', type=int, default=128, help='dimension of user and entity embeddings')
parser.add_argument('--cnn_out_size', type=int, default=128, help='dimension of cnn output')
parser.add_argument('--n_iter', type=int, default=2, help='number of iterations when computing entity representation')
parser.add_argument('--batch_size', type=int, default=128, help='batch size')
parser.add_argument('--l2_weight', type=float, default=5e-3, help='weight of l2 regularization')
parser.add_argument('--lr', type=float, default=0.0005, help='learning rate')  #3e-4
parser.add_argument('--save_path', type=str, default="./data/1week/hop2/version1/", help='model save path')
parser.add_argument('--test', type=int, default=0, help='test')
parser.add_argument('--use_group', type=int, default=1, help='whether use group')
parser.add_argument('--n_filters', type=int, default=64, help='number of filters for each size in KCNN')
parser.add_argument('--filter_sizes', type=int, default=[2, 3], nargs='+',
                    help='list of filter sizes, e.g., --filter_sizes 2 3')
parser.add_argument('--ncaps', type=int, default=7,
                    help='Maximum number of capsules per layer.')
parser.add_argument('--dcaps', type=int, default=0, help='Decrease this number of capsules per layer.')
parser.add_argument('--nhidden', type=int, default=16,
                        help='Number of hidden units per capsule.')
parser.add_argument('--routit', type=int, default=7,
                    help='Number of iterations when routing.')
parser.add_argument('--balance', type=float, default=0.004, help='learning rate')  #3e-4
parser.add_argument('--version', type=int, default=0,
                        help='Different version under the same set')

show_loss = True
show_time = False

#t = time()

args = parser.parse_args()
# data = load_data(args)
data = load_new_data(args)
train(args, data, show_loss)

#if show_time:
#    print('time used: %d s' % (time() - t))