# 'qv_arch': [[x_size,200],[200,200],[200,z_size*2]],
# 'qz_arch': [[x_size+z_size,200],[200,200],[200,z_size*2]],
# 'rv_arch': [[x_size+z_size,200],[200,200],[200,z_size*2]],
# 'flow_hidden_size': 100
# }
# # model = VAE(hyper_config)
# else:
# print ('What')
# fadas
trained_model = VAE(hyper_config)
# model.load_params(home+'/Documents/tmp/first_try/'+args.model+'/params_'+args.model+'_2800.pt')
trained_model.load_params(home+'/Documents/tmp/new_training/standard/params_standard_1000.pt')
# now to reset encoder, or even load a new one with different architecture.
# could maybe init a new one. and get model to ignore the old one.
# antoher option is init the right model then just load the params you want
# dont init q
# how does it know which params are the decoder ones, besides shape?
#ah they have keys names! mine are mostly numbers
# # print (model.state_dict())
hyper_config = {
'x_size':
x_size,
'z_size':
z_size,
'act_func':
F.tanh, # F.relu,
'encoder_arch': [[x_size, l_size], [l_size, l_size],
[l_size, z_size * 2]],
'decoder_arch': [[z_size, l_size], [l_size, l_size],
[l_size, x_size]],
'q_dist':
standard, #hnf,#aux_nf,#flow1,#,
}
model = VAE(hyper_config)
path_to_save_variables = this_dir + '/params_standard' + '_'
# elif args.model == 'flow1':
# this_dir = directory+'/flow1'
# # if not os.path.exists(this_dir):
# # os.makedirs(this_dir)
# # print ('Made directory:'+this_dir)
# # experiment_log = this_dir+'/log.txt'
# # with open(experiment_log, "a") as myfile:
# # myfile.write("Flow1" +'\n')
which_model = 'standard'
if which_model == 'standard':
print(which_model)
this_dir = directory+'/standard'
hyper_config = {
'x_size': x_size,
'z_size': z_size,
'act_func': F.tanh,# F.relu,
'encoder_arch': [[x_size,l_size],[l_size,l_size],[l_size,z_size*2]],
'decoder_arch': [[z_size,l_size],[l_size,l_size],[l_size,x_size]],
'q_dist': standard,#hnf,#aux_nf,#flow1,#,
}
model = VAE(hyper_config)
path_to_save_variables=this_dir+'/params_standard' +'_'
# elif args.model == 'flow1':
# this_dir = directory+'/flow1'
# # if not os.path.exists(this_dir):
# # os.makedirs(this_dir)
# # print ('Made directory:'+this_dir)
# 'decoder_arch': [[z_size,200],[200,200],[200,x_size]],
# 'q_dist': hnf,#aux_nf,#flow1,#standard,#, #, #, #,#, #,# ,
# 'n_flows': 2,
# 'qv_arch': [[x_size,200],[200,200],[200,z_size*2]],
# 'qz_arch': [[x_size+z_size,200],[200,200],[200,z_size*2]],
# 'rv_arch': [[x_size+z_size,200],[200,200],[200,z_size*2]],
# 'flow_hidden_size': 100
# }
# # model = VAE(hyper_config)
else:
print('What')
fadas
model = VAE(hyper_config)
print('done init')
model.load_params(home + '/Documents/tmp/large_N_time/' + args.model +
'/params_' + args.model + '_1.pt')
#Train params
learning_rate = .001
batch_size = 100
k = 1
epochs = 3000
#save params and compute IW and AIS
start_at = 100
save_freq = 300
display_epoch = 10
# with open(experiment_log, "a") as myfile:
# myfile.write("Standard" +'\n')
print('Init standard model')
hyper_config = {
'x_size': x_size,
'z_size': z_size,
'act_func': F.tanh, # F.relu,
'encoder_arch': [[x_size, 200], [200, 200], [200, z_size * 2]],
'decoder_arch': [[z_size, 200], [200, 200], [200, x_size]],
'q_dist': standard_layernorm #standard,#hnf,#aux_nf,#flow1,#,
}
model = VAE(hyper_config)
# elif model_ == 'flow1':
# this_dir = directory+'/flow1'
# if not os.path.exists(this_dir):
# os.makedirs(this_dir)
# print ('Made directory:'+this_dir)
# experiment_log = this_dir+eval_file
# with open(experiment_log, "a") as myfile:
# myfile.write("Flow1" +'\n')
# print('Init flow model')
# hyper_config = {
'qv_arch': [[x_size,200],[200,200],[200,z_size*2]],
'qz_arch': [[x_size+z_size,200],[200,200],[200,z_size*2]],
'rv_arch': [[x_size+z_size,200],[200,200],[200,z_size*2]],
'flow_hidden_size': 100
}
# model = VAE(hyper_config)
else:
print ('What')
fadas
model = VAE(hyper_config)
# model.load_params(home+'/Documents/tmp/first_try/'+args.model+'/params_'+args.model+'_2800.pt')
#Train params
# learning_rate = .001
batch_size = 50
k = 1
# epochs = 3000
#save params and compute IW and AIS
start_at = 100
save_freq = 300
display_epoch = 10
# 'decoder_arch': [[z_size,200],[200,200],[200,x_size]],
# 'q_dist': aux_nf,#aux_nf,#flow1,#standard,#, #, #, #,#, #,# ,
# 'n_flows': 2,
# 'qv_arch': [[x_size,200],[200,200],[200,z_size*2]],
# 'qz_arch': [[x_size+z_size,200],[200,200],[200,z_size*2]],
# 'rv_arch': [[x_size+z_size,200],[200,200],[200,z_size*2]],
# 'flow_hidden_size': 100
# }
# # model = VAE(hyper_config)
# else:
# print ('What')
# fadas
trained_model = VAE(hyper_config)
# model.load_params(home+'/Documents/tmp/first_try/'+args.model+'/params_'+args.model+'_2800.pt')
trained_model.load_params(
home + '/Documents/tmp/new_training/standard/params_standard_1000.pt')
# now to reset encoder, or even load a new one with different architecture.
# could maybe init a new one. and get model to ignore the old one.
# antoher option is init the right model then just load the params you want
# dont init q
# how does it know which params are the decoder ones, besides shape?
#ah they have keys names! mine are mostly numbers
# # print (model.state_dict())
# print (model.state_dict().keys())
# fsdaas
# myfile.write("Standard" +'\n')
print('Init standard model')
hyper_config = {
'x_size': x_size,
'z_size': z_size,
'act_func': F.tanh,# F.relu,
'encoder_arch': [[x_size,200],[200,200],[200,z_size*2]],
'decoder_arch': [[z_size,200],[200,200],[200,x_size]],
'q_dist': standard_layernorm #standard,#hnf,#aux_nf,#flow1,#,
}
model = VAE(hyper_config)
# elif model_ == 'flow1':
# this_dir = directory+'/flow1'
# if not os.path.exists(this_dir):
# os.makedirs(this_dir)
# print ('Made directory:'+this_dir)
# experiment_log = this_dir+eval_file
# with open(experiment_log, "a") as myfile:
# myfile.write("Flow1" +'\n')
hyper_config = {
'x_size':
x_size,
'z_size':
z_size,
'act_func':
F.tanh, # F.relu,
'encoder_arch': [[x_size, l_size], [l_size, l_size],
[l_size, z_size * 2]],
'decoder_arch': [[z_size, l_size], [l_size, l_size],
[l_size, x_size]],
'q_dist':
standard, #hnf,#aux_nf,#flow1,#,
}
model = VAE(hyper_config)
elif model_ == 'flow1':
this_dir = directory + '/flow1'
if not os.path.exists(this_dir):
os.makedirs(this_dir)
print('Made directory:' + this_dir)
experiment_log = this_dir + eval_file
with open(experiment_log, "a") as myfile:
myfile.write("Flow1" + '\n')
print('Init flow model')
hyper_config = {
myfile.write("Standard" +'\n')
print('Init standard model')
hyper_config = {
'x_size': x_size,
'z_size': z_size,
'act_func': F.tanh,# F.relu,
'encoder_arch': [[x_size,200],[200,200],[200,z_size*2]],
'decoder_arch': [[z_size,200],[200,200],[200,x_size]],
'q_dist': standard,#hnf,#aux_nf,#flow1,#,
}
model = VAE(hyper_config)
elif args.model == 'flow1':
this_dir = directory+'/flow1'
if not os.path.exists(this_dir):
os.makedirs(this_dir)
print ('Made directory:'+this_dir)
experiment_log = this_dir+eval_file
with open(experiment_log, "a") as myfile:
myfile.write("Flow1" +'\n')
# 'q_dist': hnf,#aux_nf,#flow1,#standard,#, #, #, #,#, #,# ,
# 'n_flows': 2,
# 'qv_arch': [[x_size,200],[200,200],[200,z_size*2]],
# 'qz_arch': [[x_size+z_size,200],[200,200],[200,z_size*2]],
# 'rv_arch': [[x_size+z_size,200],[200,200],[200,z_size*2]],
# 'flow_hidden_size': 100
# }
# # model = VAE(hyper_config)
else:
print ('What')
fadas
model = VAE(hyper_config)
print ('done init')
model.load_params(home+'/Documents/tmp/large_N_time/'+args.model+'/params_'+args.model+'_1.pt')
#Train params
learning_rate = .001
batch_size = 100
k = 1
epochs = 3000
#save params and compute IW and AIS
start_at = 100
save_freq = 300
display_epoch = 10
