config.py

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Fri May  8 21:56:19 2020

@author: akshitac8
"""
import argparse

parser = argparse.ArgumentParser()
parser.add_argument('--gzsl', action='store_true', default=False,help='enable generalized zero-shot learning')
parser.add_argument('--preprocessing', action='store_true',default=True, help='enbale MinMaxScaler on visual features')
parser.add_argument('--standardization', action='store_true', default=False)
parser.add_argument('--workers', type=int,help='number of data loading workers', default=16)
parser.add_argument('--batch_size', type=int,default=64, help='input batch size')
parser.add_argument('--resSize', type=int, default=4096, help='size of visual features')
parser.add_argument('--attSize', type=int, default=300, help='size of semantic features')
parser.add_argument('--ndh', type=int, default=4096,help='size of the hidden units in discriminator')
parser.add_argument('--nepoch', type=int, default=2000, help='number of epochs to train for')
parser.add_argument('--critic_iter', type=int, default=5, help='critic iteration, following WGAN-GP')
parser.add_argument('--lambda1', type=float, default=10,help='gradient penalty regularizer, following WGAN-GP')
parser.add_argument('--lr', type=float, default=0.001,help='learning rate to train GANs ')
parser.add_argument('--classifier_lr', type=float, default=0.001,help='learning rate to train softmax classifier')
parser.add_argument('--beta1', type=float, default=0.5, help='beta1 for adam. default=0.5')
parser.add_argument('--cuda', action='store_true',default=True, help='enables cuda')
parser.add_argument('--manualSeed', type=int, help='manual seed')
parser.add_argument('--nclass_all', type=int, default=1006,help='number of all classes')
parser.add_argument('--nseen_class', type=int, default=925,help='number of seen classes')
parser.add_argument("--encoder_layer_sizes1", type=int, default=4096)
parser.add_argument("--encoder_layer_sizes2", type=int, default=4096)
parser.add_argument("--decoder_layer_sizes1", type=int, default=4096)
parser.add_argument("--decoder_layer_sizes2", type=int, default=4096)
parser.add_argument('--gammaD', type=int, default=10,help='weight on the W-GAN loss')
parser.add_argument('--gammaG', type=int, default=10,help='weight on the W-GAN loss')
parser.add_argument('--validation', action='store_true',default=False, help='enable cross validation mode')
parser.add_argument('--summary', type=str,default='test dataset type 2 split', help='details regarding the code')
parser.add_argument('--N', type=int, default=10,help='number of seen and unseen labels per image')
parser.add_argument('--syn_num', type=int, default=1, help='number features to generate per class')
parser.add_argument('--classifier_epoch', type=int,default=25, help='classifier training epochs')
parser.add_argument('--classifier_batch_size', type=int,default=100, help='classifier training batch size')
parser.add_argument('--fake_batch_size', type=int,default=100, help=' synthezise batch size')

parser.add_argument("--late_fusion", action='store_true', default=False)
parser.add_argument("--early_fusion", action='store_true', default=False)
parser.add_argument("--hybrid_fusion", action='store_true', default=False)
parser.add_argument("--trimmed_train", action='store_true', default=False)

parser.add_argument('--per_seen', type=float, default=0.10 ,help='percent of seen classes')
parser.add_argument('--per_unseen', type=float, default=0.40 ,help='percent of unseen classes')
parser.add_argument('--per_seen_unseen', type=float, default=0.50 ,help='percent of seen unseen classes')

parser.add_argument("--hiddensize", type=int, default=4096)

parser.add_argument('--seen_classifier_epoch', type=int,default=25, help='classifier training epochs')
parser.add_argument('--seen_classifier_batch_size', type=int,default=100, help='classifier training batch size')
parser.add_argument("--train", action='store_true', default=False)

parser.add_argument('--val_per_seen', type=float, default=0.10 ,help='percent of val seen classes')
parser.add_argument('--val_syn_num', type=int, default=1, help='number val features to generate per class')
parser.add_argument('--test_epoch', type=int, default=1, help='epoch at which to test the model')

opt = parser.parse_args()
opt.encoder_layer_sizes = [opt.encoder_layer_sizes1, opt.encoder_layer_sizes2]
opt.decoder_layer_sizes = [opt.decoder_layer_sizes1, opt.decoder_layer_sizes2]
opt.encoder_layer_sizes[0] = opt.resSize
opt.decoder_layer_sizes[-1] = opt.resSize