update last 2

4fd56d5e · Julien Dejasmin · 31edd1e6 · 31edd1e6 · 31edd1e6 · 31edd1e6
Commit 4fd56d5e authored 4 years ago by Julien Dejasmin
--- a/OAR.2068271.stderr
+++ b/OAR.2068271.stderr
-/data1/home/julien.dejasmin/anaconda3/envs/pytorch/lib/python3.7/site-packages/torch/nn/_reduction.py:43: UserWarning: size_average and reduce args will be deprecated, please use reduction='sum' instead.
-  warnings.warn(warning.format(ret))
-## OAR [2020-06-24 16:29:57] Job 2068271 KILLED ##
--- a/OAR.2068271.stdout
+++ b/OAR.2068271.stdout
-Namespace(batch_size=256, beta=4, ckpt_dir='checkpoints', ckpt_name='last', cont_capacity=None, dataset='rendered_chairs', disc_capacity=None, epochs=400, experiment_name='beta_VAE_bs_256', gpu_devices=[0, 1], is_beta_VAE=True, latent_name='', latent_spec_cont=10, latent_spec_disc=None, load_expe_name='', load_model_checkpoint=True, lr=0.0001, num_worker=4, print_loss_every=50, record_loss_every=50, save_model=True, save_reconstruction_image=False, save_step=1, verbose=True)
-load dataset: rendered_chairs, with: 69120 train images of shape: (3, 64, 64)
-use 2 gpu who named:
-GeForce RTX 2080 Ti
-GeForce RTX 2080 Ti
-DataParallel(
-  (module): VAE(
-    (img_to_last_conv): Sequential(
-      (0): Conv2d(3, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (1): ReLU()
-      (2): Conv2d(32, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (3): ReLU()
-      (4): Conv2d(32, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (5): ReLU()
-      (6): Conv2d(64, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (7): ReLU()
-    )
-    (last_conv_to_continuous_features): Sequential(
-      (0): Conv2d(64, 256, kernel_size=(4, 4), stride=(1, 1))
-      (1): ReLU()
-    )
-    (features_to_hidden_continue): Sequential(
-      (0): Linear(in_features=256, out_features=20, bias=True)
-      (1): ReLU()
-    )
-    (latent_to_features): Sequential(
-      (0): Linear(in_features=10, out_features=256, bias=True)
-      (1): ReLU()
-    )
-    (features_to_img): Sequential(
-      (0): ConvTranspose2d(256, 64, kernel_size=(4, 4), stride=(1, 1))
-      (1): ReLU()
-      (2): ConvTranspose2d(64, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (3): ReLU()
-      (4): ConvTranspose2d(64, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (5): ReLU()
-      (6): ConvTranspose2d(32, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (7): ReLU()
-      (8): ConvTranspose2d(32, 3, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (9): Sigmoid()
-    )
-  )
-)
-The number of parameters of model is 765335
-don't use continuous capacity
-=> loaded checkpoint 'trained_models/rendered_chairs/beta_VAE_bs_256/checkpoints/last (iter 171)'
-0/69092	Loss: 147.665
-12800/69092	Loss: 158.181
-25600/69092	Loss: 159.590
-38400/69092	Loss: 158.551
-51200/69092	Loss: 158.112
-64000/69092	Loss: 157.238
-Training time 0:03:48.393375
-Epoch: 1 Average loss: 158.38
-=> saved checkpoint 'trained_models/rendered_chairs/beta_VAE_bs_256/checkpoints/last' (iter 172)
-0/69092	Loss: 159.925
-12800/69092	Loss: 158.833
-25600/69092	Loss: 159.460
-38400/69092	Loss: 157.054
-51200/69092	Loss: 157.525
-64000/69092	Loss: 158.482
-Training time 0:03:41.025229
-Epoch: 2 Average loss: 158.12
-=> saved checkpoint 'trained_models/rendered_chairs/beta_VAE_bs_256/checkpoints/last' (iter 173)
-0/69092	Loss: 152.283
-12800/69092	Loss: 157.230
-25600/69092	Loss: 157.103
-38400/69092	Loss: 158.817
--- a/OAR.2068272.stderr
+++ b/OAR.2068272.stderr
-/data1/home/julien.dejasmin/anaconda3/envs/pytorch/lib/python3.7/site-packages/torch/nn/_reduction.py:43: UserWarning: size_average and reduce args will be deprecated, please use reduction='sum' instead.
-  warnings.warn(warning.format(ret))
-## OAR [2020-06-24 16:29:57] Job 2068272 KILLED ##
--- a/OAR.2068272.stdout
+++ b/OAR.2068272.stdout
-Namespace(batch_size=64, beta=4, ckpt_dir='checkpoints', ckpt_name='last', cont_capacity=None, dataset='rendered_chairs', disc_capacity=None, epochs=400, experiment_name='beta_VAE_bs_64', gpu_devices=[0, 1], is_beta_VAE=True, latent_name='', latent_spec_cont=10, latent_spec_disc=None, load_expe_name='', load_model_checkpoint=True, lr=0.0001, num_worker=4, print_loss_every=50, record_loss_every=50, save_model=True, save_reconstruction_image=False, save_step=1, verbose=True)
-load dataset: rendered_chairs, with: 69120 train images of shape: (3, 64, 64)
-use 2 gpu who named:
-Tesla K80
-Tesla K80
-DataParallel(
-  (module): VAE(
-    (img_to_last_conv): Sequential(
-      (0): Conv2d(3, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (1): ReLU()
-      (2): Conv2d(32, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (3): ReLU()
-      (4): Conv2d(32, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (5): ReLU()
-      (6): Conv2d(64, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (7): ReLU()
-    )
-    (last_conv_to_continuous_features): Sequential(
-      (0): Conv2d(64, 256, kernel_size=(4, 4), stride=(1, 1))
-      (1): ReLU()
-    )
-    (features_to_hidden_continue): Sequential(
-      (0): Linear(in_features=256, out_features=20, bias=True)
-      (1): ReLU()
-    )
-    (latent_to_features): Sequential(
-      (0): Linear(in_features=10, out_features=256, bias=True)
-      (1): ReLU()
-    )
-    (features_to_img): Sequential(
-      (0): ConvTranspose2d(256, 64, kernel_size=(4, 4), stride=(1, 1))
-      (1): ReLU()
-      (2): ConvTranspose2d(64, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (3): ReLU()
-      (4): ConvTranspose2d(64, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (5): ReLU()
-      (6): ConvTranspose2d(32, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (7): ReLU()
-      (8): ConvTranspose2d(32, 3, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (9): Sigmoid()
-    )
-  )
-)
-The number of parameters of model is 765335
-don't use continuous capacity
-=> loaded checkpoint 'trained_models/rendered_chairs/beta_VAE_bs_64/checkpoints/last (iter 304)'
-0/69092	Loss: 153.033
-3200/69092	Loss: 151.558
-6400/69092	Loss: 150.062
-9600/69092	Loss: 154.221
-12800/69092	Loss: 151.615
-16000/69092	Loss: 153.931
-19200/69092	Loss: 153.015
-22400/69092	Loss: 153.433
-25600/69092	Loss: 151.584
-28800/69092	Loss: 154.002
-32000/69092	Loss: 153.302
-35200/69092	Loss: 151.154
-38400/69092	Loss: 152.741
-41600/69092	Loss: 153.137
-44800/69092	Loss: 152.327
-48000/69092	Loss: 152.095
-51200/69092	Loss: 150.511
-54400/69092	Loss: 151.692
-57600/69092	Loss: 151.058
-60800/69092	Loss: 155.055
-64000/69092	Loss: 150.898
-67200/69092	Loss: 155.036
-Training time 0:01:58.162781
-Epoch: 1 Average loss: 152.50
-=> saved checkpoint 'trained_models/rendered_chairs/beta_VAE_bs_64/checkpoints/last' (iter 305)
-0/69092	Loss: 143.092
-3200/69092	Loss: 153.010
-6400/69092	Loss: 155.530
-9600/69092	Loss: 155.364
-12800/69092	Loss: 151.123
-16000/69092	Loss: 154.058
-19200/69092	Loss: 151.657
-22400/69092	Loss: 153.832
-25600/69092	Loss: 151.606
-28800/69092	Loss: 153.287
-32000/69092	Loss: 150.634
-35200/69092	Loss: 152.244
-38400/69092	Loss: 152.817
-41600/69092	Loss: 152.458
-44800/69092	Loss: 153.017
-48000/69092	Loss: 151.141
-51200/69092	Loss: 154.387
-54400/69092	Loss: 149.705
-57600/69092	Loss: 151.770
-60800/69092	Loss: 151.618
-64000/69092	Loss: 152.999
-67200/69092	Loss: 154.086
-Training time 0:01:57.931589
-Epoch: 2 Average loss: 152.66
-=> saved checkpoint 'trained_models/rendered_chairs/beta_VAE_bs_64/checkpoints/last' (iter 306)
-0/69092	Loss: 149.850
-3200/69092	Loss: 153.249
-6400/69092	Loss: 153.873
-9600/69092	Loss: 153.335
-12800/69092	Loss: 153.749
-16000/69092	Loss: 148.933
-19200/69092	Loss: 154.976
-22400/69092	Loss: 153.382
-25600/69092	Loss: 151.603
-28800/69092	Loss: 152.808
-32000/69092	Loss: 151.032
-35200/69092	Loss: 151.926
-38400/69092	Loss: 155.662
-41600/69092	Loss: 150.252
-44800/69092	Loss: 152.976
-48000/69092	Loss: 153.162
-51200/69092	Loss: 153.542
-54400/69092	Loss: 152.422
-57600/69092	Loss: 150.808
-60800/69092	Loss: 152.001
-64000/69092	Loss: 153.256
-67200/69092	Loss: 152.930
-Training time 0:01:57.656130
-Epoch: 3 Average loss: 152.64
-=> saved checkpoint 'trained_models/rendered_chairs/beta_VAE_bs_64/checkpoints/last' (iter 307)
-0/69092	Loss: 161.186
-3200/69092	Loss: 151.119
-6400/69092	Loss: 154.165
-9600/69092	Loss: 150.897
-12800/69092	Loss: 152.695
-16000/69092	Loss: 151.460
-19200/69092	Loss: 154.138
-22400/69092	Loss: 153.239
-25600/69092	Loss: 152.014
-28800/69092	Loss: 151.173
-32000/69092	Loss: 155.464
-35200/69092	Loss: 154.407
-38400/69092	Loss: 150.135
-41600/69092	Loss: 154.000
-44800/69092	Loss: 152.920
-48000/69092	Loss: 151.982
-51200/69092	Loss: 155.144
-54400/69092	Loss: 151.637
-57600/69092	Loss: 150.082
-60800/69092	Loss: 153.631
-64000/69092	Loss: 152.659
-67200/69092	Loss: 153.400
-Training time 0:01:58.079109
-Epoch: 4 Average loss: 152.67
-=> saved checkpoint 'trained_models/rendered_chairs/beta_VAE_bs_64/checkpoints/last' (iter 308)
-0/69092	Loss: 173.856
-3200/69092	Loss: 152.023
-6400/69092	Loss: 153.829
-9600/69092	Loss: 150.343
-12800/69092	Loss: 151.734
-16000/69092	Loss: 152.576
-19200/69092	Loss: 153.110
-22400/69092	Loss: 155.344
-25600/69092	Loss: 151.083
-28800/69092	Loss: 151.468
-32000/69092	Loss: 150.122
-35200/69092	Loss: 150.961
-38400/69092	Loss: 152.861
-41600/69092	Loss: 150.400
-44800/69092	Loss: 154.243
-48000/69092	Loss: 156.053
-51200/69092	Loss: 151.848
-54400/69092	Loss: 154.020
-57600/69092	Loss: 152.957
-60800/69092	Loss: 154.637
-64000/69092	Loss: 151.895
-67200/69092	Loss: 152.106
-Training time 0:01:58.719240
-Epoch: 5 Average loss: 152.53
-=> saved checkpoint 'trained_models/rendered_chairs/beta_VAE_bs_64/checkpoints/last' (iter 309)
-0/69092	Loss: 161.423
-3200/69092	Loss: 151.636
-6400/69092	Loss: 150.785
--- a/OAR.2068273.stderr
+++ b/OAR.2068273.stderr
-/data1/home/julien.dejasmin/anaconda3/envs/pytorch/lib/python3.7/site-packages/torch/nn/_reduction.py:43: UserWarning: size_average and reduce args will be deprecated, please use reduction='sum' instead.
-  warnings.warn(warning.format(ret))
-## OAR [2020-06-24 16:29:57] Job 2068273 KILLED ##
--- a/OAR.2068273.stdout
+++ b/OAR.2068273.stdout
-Namespace(batch_size=256, beta=None, ckpt_dir='checkpoints', ckpt_name='last', cont_capacity=None, dataset='rendered_chairs', disc_capacity=None, epochs=400, experiment_name='VAE_bs_256', gpu_devices=[0, 1], is_beta_VAE=False, latent_name='', latent_spec_cont=10, latent_spec_disc=None, load_expe_name='', load_model_checkpoint=True, lr=0.0001, num_worker=4, print_loss_every=50, record_loss_every=50, save_model=True, save_reconstruction_image=False, save_step=1, verbose=True)
-load dataset: rendered_chairs, with: 69120 train images of shape: (3, 64, 64)
-use 2 gpu who named:
-GeForce GTX 1080 Ti
-GeForce GTX 1080 Ti
-DataParallel(
-  (module): VAE(
-    (img_to_last_conv): Sequential(
-      (0): Conv2d(3, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (1): ReLU()
-      (2): Conv2d(32, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (3): ReLU()
-      (4): Conv2d(32, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (5): ReLU()
-      (6): Conv2d(64, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (7): ReLU()
-    )
-    (last_conv_to_continuous_features): Sequential(
-      (0): Conv2d(64, 256, kernel_size=(4, 4), stride=(1, 1))
-      (1): ReLU()
-    )
-    (features_to_hidden_continue): Sequential(
-      (0): Linear(in_features=256, out_features=20, bias=True)
-      (1): ReLU()
-    )
-    (latent_to_features): Sequential(
-      (0): Linear(in_features=10, out_features=256, bias=True)
-      (1): ReLU()
-    )
-    (features_to_img): Sequential(
-      (0): ConvTranspose2d(256, 64, kernel_size=(4, 4), stride=(1, 1))
-      (1): ReLU()
-      (2): ConvTranspose2d(64, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (3): ReLU()
-      (4): ConvTranspose2d(64, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (5): ReLU()
-      (6): ConvTranspose2d(32, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (7): ReLU()
-      (8): ConvTranspose2d(32, 3, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (9): Sigmoid()
-    )
-  )
-)
-The number of parameters of model is 765335
-don't use continuous capacity
-=> loaded checkpoint 'trained_models/rendered_chairs/VAE_bs_256/checkpoints/last (iter 143)'
-0/69092	Loss: 116.257
-12800/69092	Loss: 116.865
-25600/69092	Loss: 115.704
-38400/69092	Loss: 116.840
-51200/69092	Loss: 116.758
-64000/69092	Loss: 116.974
-Training time 0:05:34.767176
-Epoch: 1 Average loss: 116.75
-=> saved checkpoint 'trained_models/rendered_chairs/VAE_bs_256/checkpoints/last' (iter 144)
-0/69092	Loss: 115.425
-12800/69092	Loss: 116.366
-25600/69092	Loss: 115.519
-38400/69092	Loss: 116.832
-51200/69092	Loss: 116.514
--- a/OAR.2068274.stderr
+++ b/OAR.2068274.stderr
-/data1/home/julien.dejasmin/anaconda3/envs/pytorch/lib/python3.7/site-packages/torch/nn/_reduction.py:43: UserWarning: size_average and reduce args will be deprecated, please use reduction='sum' instead.
-  warnings.warn(warning.format(ret))
-## OAR [2020-06-24 16:29:57] Job 2068274 KILLED ##
--- a/OAR.2068274.stdout
+++ b/OAR.2068274.stdout
-Namespace(batch_size=64, beta=None, ckpt_dir='checkpoints', ckpt_name='last', cont_capacity=None, dataset='rendered_chairs', disc_capacity=None, epochs=400, experiment_name='VAE_bs_64', gpu_devices=[0, 1], is_beta_VAE=False, latent_name='', latent_spec_cont=10, latent_spec_disc=None, load_expe_name='', load_model_checkpoint=True, lr=0.0001, num_worker=4, print_loss_every=50, record_loss_every=50, save_model=True, save_reconstruction_image=False, save_step=1, verbose=True)
-load dataset: rendered_chairs, with: 69120 train images of shape: (3, 64, 64)
-use 2 gpu who named:
-GeForce RTX 2080 Ti
-GeForce RTX 2080 Ti
-DataParallel(
-  (module): VAE(
-    (img_to_last_conv): Sequential(
-      (0): Conv2d(3, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (1): ReLU()
-      (2): Conv2d(32, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (3): ReLU()
-      (4): Conv2d(32, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (5): ReLU()
-      (6): Conv2d(64, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (7): ReLU()
-    )
-    (last_conv_to_continuous_features): Sequential(
-      (0): Conv2d(64, 256, kernel_size=(4, 4), stride=(1, 1))
-      (1): ReLU()
-    )
-    (features_to_hidden_continue): Sequential(
-      (0): Linear(in_features=256, out_features=20, bias=True)
-      (1): ReLU()
-    )
-    (latent_to_features): Sequential(
-      (0): Linear(in_features=10, out_features=256, bias=True)
-      (1): ReLU()
-    )
-    (features_to_img): Sequential(
-      (0): ConvTranspose2d(256, 64, kernel_size=(4, 4), stride=(1, 1))
-      (1): ReLU()
-      (2): ConvTranspose2d(64, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (3): ReLU()
-      (4): ConvTranspose2d(64, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (5): ReLU()
-      (6): ConvTranspose2d(32, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (7): ReLU()
-      (8): ConvTranspose2d(32, 3, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (9): Sigmoid()
-    )
-  )
-)
-The number of parameters of model is 765335
-don't use continuous capacity
-=> loaded checkpoint 'trained_models/rendered_chairs/VAE_bs_64/checkpoints/last (iter 139)'
-0/69092	Loss: 110.847
-3200/69092	Loss: 115.390
-6400/69092	Loss: 112.621
-9600/69092	Loss: 112.659
-12800/69092	Loss: 114.223
-16000/69092	Loss: 112.621
-19200/69092	Loss: 115.178
-22400/69092	Loss: 113.874
-25600/69092	Loss: 113.745
-28800/69092	Loss: 114.605
-32000/69092	Loss: 113.612
-35200/69092	Loss: 114.793
-38400/69092	Loss: 116.839
-41600/69092	Loss: 114.658
-44800/69092	Loss: 114.111
-48000/69092	Loss: 112.756
-51200/69092	Loss: 114.016
-54400/69092	Loss: 114.316
-57600/69092	Loss: 112.581
-60800/69092	Loss: 112.926
-64000/69092	Loss: 112.703
-67200/69092	Loss: 112.203
-Training time 0:04:30.608806
-Epoch: 1 Average loss: 113.81
-=> saved checkpoint 'trained_models/rendered_chairs/VAE_bs_64/checkpoints/last' (iter 140)
-0/69092	Loss: 102.792
-3200/69092	Loss: 112.805
-6400/69092	Loss: 114.190
-9600/69092	Loss: 114.078
-12800/69092	Loss: 113.312
-16000/69092	Loss: 112.534
-19200/69092	Loss: 113.381
-22400/69092	Loss: 114.327
-25600/69092	Loss: 114.343
-28800/69092	Loss: 114.635
-32000/69092	Loss: 114.228
-35200/69092	Loss: 113.063
-38400/69092	Loss: 113.853
-41600/69092	Loss: 114.526
-44800/69092	Loss: 113.829
-48000/69092	Loss: 114.176
-51200/69092	Loss: 113.607
-54400/69092	Loss: 113.869
-57600/69092	Loss: 113.135
-60800/69092	Loss: 113.620
-64000/69092	Loss: 114.794
-67200/69092	Loss: 113.848
-Training time 0:04:18.213843
-Epoch: 2 Average loss: 113.78
-=> saved checkpoint 'trained_models/rendered_chairs/VAE_bs_64/checkpoints/last' (iter 141)
-0/69092	Loss: 123.686
-3200/69092	Loss: 115.639
-6400/69092	Loss: 113.484
-9600/69092	Loss: 112.216
-12800/69092	Loss: 112.320
-16000/69092	Loss: 115.109
--- a/OAR.2068275.stderr
+++ b/OAR.2068275.stderr
-/data1/home/julien.dejasmin/anaconda3/envs/pytorch/lib/python3.7/site-packages/torch/nn/_reduction.py:43: UserWarning: size_average and reduce args will be deprecated, please use reduction='sum' instead.
-  warnings.warn(warning.format(ret))
-## OAR [2020-06-24 16:29:57] Job 2068275 KILLED ##
--- a/OAR.2068275.stdout
+++ b/OAR.2068275.stdout
-Namespace(batch_size=64, beta=4, ckpt_dir='checkpoints', ckpt_name='last', cont_capacity=None, dataset='rendered_chairs', disc_capacity=None, epochs=400, experiment_name='beta_VAE_bs_64_ls_15', gpu_devices=[0, 1], is_beta_VAE=True, latent_name='', latent_spec_cont=15, latent_spec_disc=None, load_expe_name='', load_model_checkpoint=False, lr=0.0001, num_worker=4, print_loss_every=50, record_loss_every=50, save_model=True, save_reconstruction_image=False, save_step=1, verbose=True)
-creare new diretory experiment: rendered_chairs/beta_VAE_bs_64_ls_15
-load dataset: rendered_chairs, with: 69120 train images of shape: (3, 64, 64)
-use 2 gpu who named:
-GeForce RTX 2080 Ti
-GeForce RTX 2080 Ti
-DataParallel(
-  (module): VAE(
-    (img_to_last_conv): Sequential(
-      (0): Conv2d(3, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (1): ReLU()
-      (2): Conv2d(32, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (3): ReLU()
-      (4): Conv2d(32, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (5): ReLU()
-      (6): Conv2d(64, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (7): ReLU()
-    )
-    (last_conv_to_continuous_features): Sequential(
-      (0): Conv2d(64, 256, kernel_size=(4, 4), stride=(1, 1))
-      (1): ReLU()
-    )
-    (features_to_hidden_continue): Sequential(
-      (0): Linear(in_features=256, out_features=30, bias=True)
-      (1): ReLU()
-    )
-    (latent_to_features): Sequential(
-      (0): Linear(in_features=15, out_features=256, bias=True)
-      (1): ReLU()
-    )
-    (features_to_img): Sequential(
-      (0): ConvTranspose2d(256, 64, kernel_size=(4, 4), stride=(1, 1))
-      (1): ReLU()
-      (2): ConvTranspose2d(64, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (3): ReLU()
-      (4): ConvTranspose2d(64, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (5): ReLU()
-      (6): ConvTranspose2d(32, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (7): ReLU()
-      (8): ConvTranspose2d(32, 3, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (9): Sigmoid()
-    )
-  )
-)
-The number of parameters of model is 769185
-don't use continuous capacity
-=> no checkpoint found at 'trained_models/rendered_chairs/beta_VAE_bs_64_ls_15/checkpoints/last'
-0/69092	Loss: 2891.769
-3200/69092	Loss: 2825.642
-6400/69092	Loss: 1205.228
-9600/69092	Loss: 556.734
-12800/69092	Loss: 484.027
-16000/69092	Loss: 465.437
-19200/69092	Loss: 448.356
-22400/69092	Loss: 437.048
-25600/69092	Loss: 401.074
-28800/69092	Loss: 289.548
-32000/69092	Loss: 238.668
-35200/69092	Loss: 234.046
-38400/69092	Loss: 226.796
-41600/69092	Loss: 222.254
-44800/69092	Loss: 225.392
-48000/69092	Loss: 221.208
-51200/69092	Loss: 220.618
-54400/69092	Loss: 213.913
-57600/69092	Loss: 210.644
-60800/69092	Loss: 213.087
-64000/69092	Loss: 216.019
-67200/69092	Loss: 208.903
-Training time 0:04:53.937253
-Epoch: 1 Average loss: 460.47
-=> saved checkpoint 'trained_models/rendered_chairs/beta_VAE_bs_64_ls_15/checkpoints/last' (iter 1)
-0/69092	Loss: 188.720
-3200/69092	Loss: 204.200
-6400/69092	Loss: 204.811
-9600/69092	Loss: 200.863
-12800/69092	Loss: 202.928
-16000/69092	Loss: 205.920
-19200/69092	Loss: 195.467
-22400/69092	Loss: 199.383
-25600/69092	Loss: 200.195
-28800/69092	Loss: 196.327
-32000/69092	Loss: 196.155
-35200/69092	Loss: 193.812
-38400/69092	Loss: 193.730
-41600/69092	Loss: 194.544
-44800/69092	Loss: 196.956
-48000/69092	Loss: 190.843
-51200/69092	Loss: 192.657
-54400/69092	Loss: 189.354
-57600/69092	Loss: 192.162
-60800/69092	Loss: 189.223
--- a/OAR.2068276.stderr
+++ b/OAR.2068276.stderr
-/data1/home/julien.dejasmin/anaconda3/envs/pytorch/lib/python3.7/site-packages/torch/nn/_reduction.py:43: UserWarning: size_average and reduce args will be deprecated, please use reduction='sum' instead.
-  warnings.warn(warning.format(ret))
-## OAR [2020-06-24 16:39:19] Job 2068276 KILLED ##
--- a/OAR.2068276.stdout
+++ b/OAR.2068276.stdout
-Namespace(batch_size=64, beta=4, ckpt_dir='checkpoints', ckpt_name='last', cont_capacity=None, dataset='rendered_chairs', disc_capacity=None, epochs=400, experiment_name='beta_VAE_bs_64_ls_20', gpu_devices=[0, 1], is_beta_VAE=True, latent_name='', latent_spec_cont=20, latent_spec_disc=None, load_expe_name='', load_model_checkpoint=False, lr=0.0001, num_worker=4, print_loss_every=50, record_loss_every=50, save_model=True, save_reconstruction_image=False, save_step=1, verbose=True)
-creare new diretory experiment: rendered_chairs/beta_VAE_bs_64_ls_20
-load dataset: rendered_chairs, with: 69120 train images of shape: (3, 64, 64)
-use 2 gpu who named:
-GeForce RTX 2080 Ti
-GeForce RTX 2080 Ti
-DataParallel(
-  (module): VAE(
-    (img_to_last_conv): Sequential(
-      (0): Conv2d(3, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (1): ReLU()
-      (2): Conv2d(32, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (3): ReLU()
-      (4): Conv2d(32, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (5): ReLU()
-      (6): Conv2d(64, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (7): ReLU()
-    )
-    (last_conv_to_continuous_features): Sequential(
-      (0): Conv2d(64, 256, kernel_size=(4, 4), stride=(1, 1))
-      (1): ReLU()
-    )
-    (features_to_hidden_continue): Sequential(
-      (0): Linear(in_features=256, out_features=40, bias=True)
-      (1): ReLU()
-    )
-    (latent_to_features): Sequential(
-      (0): Linear(in_features=20, out_features=256, bias=True)
-      (1): ReLU()
-    )
-    (features_to_img): Sequential(
-      (0): ConvTranspose2d(256, 64, kernel_size=(4, 4), stride=(1, 1))
-      (1): ReLU()
-      (2): ConvTranspose2d(64, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (3): ReLU()
-      (4): ConvTranspose2d(64, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (5): ReLU()
-      (6): ConvTranspose2d(32, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (7): ReLU()
-      (8): ConvTranspose2d(32, 3, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (9): Sigmoid()
-    )
-  )
-)
-The number of parameters of model is 773035
-don't use continuous capacity
-=> no checkpoint found at 'trained_models/rendered_chairs/beta_VAE_bs_64_ls_20/checkpoints/last'
-0/69092	Loss: 3095.371
-3200/69092	Loss: 3013.162
-6400/69092	Loss: 1142.397
-9600/69092	Loss: 583.239
-12800/69092	Loss: 500.184
-16000/69092	Loss: 479.822
-19200/69092	Loss: 462.136
-22400/69092	Loss: 446.272
-25600/69092	Loss: 445.532
-28800/69092	Loss: 451.249
-32000/69092	Loss: 444.728
-35200/69092	Loss: 436.249
-38400/69092	Loss: 434.357
-41600/69092	Loss: 450.662
-44800/69092	Loss: 436.829
-48000/69092	Loss: 444.385
-51200/69092	Loss: 442.255
-54400/69092	Loss: 431.708
-57600/69092	Loss: 444.837
-60800/69092	Loss: 445.616
-64000/69092	Loss: 438.358
-67200/69092	Loss: 448.892
-Training time 0:04:22.264375
-Epoch: 1 Average loss: 608.54
-=> saved checkpoint 'trained_models/rendered_chairs/beta_VAE_bs_64_ls_20/checkpoints/last' (iter 1)
-0/69092	Loss: 419.617
-3200/69092	Loss: 441.728
-6400/69092	Loss: 436.680
-9600/69092	Loss: 440.386
-12800/69092	Loss: 437.215
-16000/69092	Loss: 439.624
-19200/69092	Loss: 436.731
-22400/69092	Loss: 436.506
-25600/69092	Loss: 439.104
-28800/69092	Loss: 433.798
-32000/69092	Loss: 443.554
-35200/69092	Loss: 442.369
-38400/69092	Loss: 441.461
-41600/69092	Loss: 428.932
-44800/69092	Loss: 458.124
-48000/69092	Loss: 440.610
-51200/69092	Loss: 433.183
-54400/69092	Loss: 445.103
-57600/69092	Loss: 448.074
-60800/69092	Loss: 437.783
-64000/69092	Loss: 442.622
-67200/69092	Loss: 435.505
-Training time 0:04:38.790417
-Epoch: 2 Average loss: 439.99
-=> saved checkpoint 'trained_models/rendered_chairs/beta_VAE_bs_64_ls_20/checkpoints/last' (iter 2)
-0/69092	Loss: 419.540
-3200/69092	Loss: 431.518
-6400/69092	Loss: 444.740
-9600/69092	Loss: 437.046
-12800/69092	Loss: 446.255
--- a/OAR.2068277.stderr
+++ b/OAR.2068277.stderr
-/data1/home/julien.dejasmin/anaconda3/envs/pytorch/lib/python3.7/site-packages/torch/nn/_reduction.py:43: UserWarning: size_average and reduce args will be deprecated, please use reduction='sum' instead.
-  warnings.warn(warning.format(ret))
-## OAR [2020-06-24 16:40:03] Job 2068277 KILLED ##
--- a/OAR.2068277.stdout
+++ b/OAR.2068277.stdout
-Namespace(batch_size=64, beta=4, ckpt_dir='checkpoints', ckpt_name='last', cont_capacity=None, dataset='rendered_chairs', disc_capacity=None, epochs=400, experiment_name='beta_VAE_bs_64_ls_5', gpu_devices=[0, 1], is_beta_VAE=True, latent_name='', latent_spec_cont=5, latent_spec_disc=None, load_expe_name='', load_model_checkpoint=False, lr=0.0001, num_worker=4, print_loss_every=50, record_loss_every=50, save_model=True, save_reconstruction_image=False, save_step=1, verbose=True)
-creare new diretory experiment: rendered_chairs/beta_VAE_bs_64_ls_5
-load dataset: rendered_chairs, with: 69120 train images of shape: (3, 64, 64)
-use 2 gpu who named:
-GeForce RTX 2080 Ti
-GeForce RTX 2080 Ti
-DataParallel(
-  (module): VAE(
-    (img_to_last_conv): Sequential(
-      (0): Conv2d(3, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (1): ReLU()
-      (2): Conv2d(32, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (3): ReLU()
-      (4): Conv2d(32, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (5): ReLU()
-      (6): Conv2d(64, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (7): ReLU()
-    )
-    (last_conv_to_continuous_features): Sequential(
-      (0): Conv2d(64, 256, kernel_size=(4, 4), stride=(1, 1))
-      (1): ReLU()
-    )
-    (features_to_hidden_continue): Sequential(
-      (0): Linear(in_features=256, out_features=10, bias=True)
-      (1): ReLU()
-    )
-    (latent_to_features): Sequential(
-      (0): Linear(in_features=5, out_features=256, bias=True)
-      (1): ReLU()
-    )
-    (features_to_img): Sequential(
-      (0): ConvTranspose2d(256, 64, kernel_size=(4, 4), stride=(1, 1))
-      (1): ReLU()
-      (2): ConvTranspose2d(64, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (3): ReLU()
-      (4): ConvTranspose2d(64, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (5): ReLU()
-      (6): ConvTranspose2d(32, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (7): ReLU()
-      (8): ConvTranspose2d(32, 3, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (9): Sigmoid()
-    )
-  )
-)
-The number of parameters of model is 761485
-don't use continuous capacity
-=> no checkpoint found at 'trained_models/rendered_chairs/beta_VAE_bs_64_ls_5/checkpoints/last'
-0/69092	Loss: 2840.949
-3200/69092	Loss: 2696.334
-6400/69092	Loss: 878.800
-9600/69092	Loss: 529.528
-12800/69092	Loss: 473.136
-16000/69092	Loss: 459.979
-19200/69092	Loss: 427.518
-22400/69092	Loss: 312.154
-25600/69092	Loss: 251.443
-28800/69092	Loss: 243.150
-32000/69092	Loss: 238.743
-35200/69092	Loss: 232.739
-38400/69092	Loss: 233.067
-41600/69092	Loss: 230.312
-44800/69092	Loss: 223.895
-48000/69092	Loss: 230.252
-51200/69092	Loss: 228.705
-54400/69092	Loss: 232.979
-57600/69092	Loss: 229.967
-60800/69092	Loss: 224.165
-64000/69092	Loss: 224.971
-67200/69092	Loss: 228.978
-Training time 0:04:33.155919
-Epoch: 1 Average loss: 426.85
-=> saved checkpoint 'trained_models/rendered_chairs/beta_VAE_bs_64_ls_5/checkpoints/last' (iter 1)
-0/69092	Loss: 220.633
-3200/69092	Loss: 223.511
-6400/69092	Loss: 224.541
-9600/69092	Loss: 225.067
-12800/69092	Loss: 222.152
-16000/69092	Loss: 222.201
-19200/69092	Loss: 225.091
-22400/69092	Loss: 220.790
-25600/69092	Loss: 219.536
-28800/69092	Loss: 214.187
-32000/69092	Loss: 221.030
-35200/69092	Loss: 217.724
-38400/69092	Loss: 214.937
-41600/69092	Loss: 221.160
-44800/69092	Loss: 214.006
-48000/69092	Loss: 216.696
-51200/69092	Loss: 208.186
-54400/69092	Loss: 198.221
-57600/69092	Loss: 202.476
-60800/69092	Loss: 191.258
-64000/69092	Loss: 196.500
-67200/69092	Loss: 196.036
-Training time 0:04:43.151928
-Epoch: 2 Average loss: 213.44
-=> saved checkpoint 'trained_models/rendered_chairs/beta_VAE_bs_64_ls_5/checkpoints/last' (iter 2)
-0/69092	Loss: 181.796
-3200/69092	Loss: 190.527
-6400/69092	Loss: 192.437
--- a/OAR.2068278.stderr
+++ b/OAR.2068278.stderr
-/data1/home/julien.dejasmin/anaconda3/envs/pytorch/lib/python3.7/site-packages/torch/nn/_reduction.py:43: UserWarning: size_average and reduce args will be deprecated, please use reduction='sum' instead.
-  warnings.warn(warning.format(ret))
-## OAR [2020-06-24 16:40:04] Job 2068278 KILLED ##
--- a/OAR.2068278.stdout
+++ b/OAR.2068278.stdout
-Namespace(batch_size=64, beta=None, ckpt_dir='checkpoints', ckpt_name='last', cont_capacity=None, dataset='rendered_chairs', disc_capacity=None, epochs=400, experiment_name='VAE_bs_64_ls_5', gpu_devices=[0, 1], is_beta_VAE=False, latent_name='', latent_spec_cont=5, latent_spec_disc=None, load_expe_name='', load_model_checkpoint=False, lr=0.0001, num_worker=4, print_loss_every=50, record_loss_every=50, save_model=True, save_reconstruction_image=False, save_step=1, verbose=True)
-creare new diretory experiment: rendered_chairs/VAE_bs_64_ls_5
-load dataset: rendered_chairs, with: 69120 train images of shape: (3, 64, 64)
-use 2 gpu who named:
-GeForce RTX 2080 Ti
-GeForce RTX 2080 Ti
-DataParallel(
-  (module): VAE(
-    (img_to_last_conv): Sequential(
-      (0): Conv2d(3, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (1): ReLU()
-      (2): Conv2d(32, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (3): ReLU()
-      (4): Conv2d(32, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (5): ReLU()
-      (6): Conv2d(64, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (7): ReLU()
-    )
-    (last_conv_to_continuous_features): Sequential(
-      (0): Conv2d(64, 256, kernel_size=(4, 4), stride=(1, 1))
-      (1): ReLU()
-    )
-    (features_to_hidden_continue): Sequential(
-      (0): Linear(in_features=256, out_features=10, bias=True)
-      (1): ReLU()
-    )
-    (latent_to_features): Sequential(
-      (0): Linear(in_features=5, out_features=256, bias=True)
-      (1): ReLU()
-    )
-    (features_to_img): Sequential(
-      (0): ConvTranspose2d(256, 64, kernel_size=(4, 4), stride=(1, 1))
-      (1): ReLU()
-      (2): ConvTranspose2d(64, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (3): ReLU()
-      (4): ConvTranspose2d(64, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (5): ReLU()
-      (6): ConvTranspose2d(32, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (7): ReLU()
-      (8): ConvTranspose2d(32, 3, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (9): Sigmoid()
-    )
-  )
-)
-The number of parameters of model is 761485
-don't use continuous capacity
-=> no checkpoint found at 'trained_models/rendered_chairs/VAE_bs_64_ls_5/checkpoints/last'
-0/69092	Loss: 2903.199
-3200/69092	Loss: 2704.227
-6400/69092	Loss: 982.409
-9600/69092	Loss: 523.329
-12800/69092	Loss: 363.221
-16000/69092	Loss: 276.723
-19200/69092	Loss: 250.323
-22400/69092	Loss: 229.455
-25600/69092	Loss: 220.906
-28800/69092	Loss: 220.425
-32000/69092	Loss: 216.671
-35200/69092	Loss: 213.795
-38400/69092	Loss: 211.775
-41600/69092	Loss: 205.819
-44800/69092	Loss: 206.077
-48000/69092	Loss: 208.456
-51200/69092	Loss: 210.820
-54400/69092	Loss: 204.282
-57600/69092	Loss: 204.736
-60800/69092	Loss: 204.576
-64000/69092	Loss: 200.975
-67200/69092	Loss: 198.954
-Training time 0:05:45.368750
-Epoch: 1 Average loss: 390.23
-=> saved checkpoint 'trained_models/rendered_chairs/VAE_bs_64_ls_5/checkpoints/last' (iter 1)
-0/69092	Loss: 193.590
-3200/69092	Loss: 194.946
-6400/69092	Loss: 199.709
-9600/69092	Loss: 200.276
-12800/69092	Loss: 193.180
-16000/69092	Loss: 192.735
-19200/69092	Loss: 190.256
-22400/69092	Loss: 190.013
-25600/69092	Loss: 188.325
-28800/69092	Loss: 189.229
-32000/69092	Loss: 189.882
-35200/69092	Loss: 190.248
-38400/69092	Loss: 188.566
-41600/69092	Loss: 185.072
-44800/69092	Loss: 189.564
-48000/69092	Loss: 181.238
-51200/69092	Loss: 181.543
-54400/69092	Loss: 182.526
--- a/OAR.2068279.stderr
+++ b/OAR.2068279.stderr
-/data1/home/julien.dejasmin/anaconda3/envs/pytorch/lib/python3.7/site-packages/torch/nn/_reduction.py:43: UserWarning: size_average and reduce args will be deprecated, please use reduction='sum' instead.
-  warnings.warn(warning.format(ret))
-## OAR [2020-06-24 16:40:20] Job 2068279 KILLED ##
--- a/OAR.2068279.stdout
+++ b/OAR.2068279.stdout
-Namespace(batch_size=64, beta=None, ckpt_dir='checkpoints', ckpt_name='last', cont_capacity=None, dataset='rendered_chairs', disc_capacity=None, epochs=400, experiment_name='VAE_bs_64_ls_15', gpu_devices=[0, 1], is_beta_VAE=False, latent_name='', latent_spec_cont=15, latent_spec_disc=None, load_expe_name='', load_model_checkpoint=False, lr=0.0001, num_worker=4, print_loss_every=50, record_loss_every=50, save_model=True, save_reconstruction_image=False, save_step=1, verbose=True)
-creare new diretory experiment: rendered_chairs/VAE_bs_64_ls_15
-load dataset: rendered_chairs, with: 69120 train images of shape: (3, 64, 64)
-use 2 gpu who named:
-GeForce RTX 2080 Ti
-GeForce RTX 2080 Ti
-DataParallel(
-  (module): VAE(
-    (img_to_last_conv): Sequential(
-      (0): Conv2d(3, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (1): ReLU()
-      (2): Conv2d(32, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (3): ReLU()
-      (4): Conv2d(32, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (5): ReLU()
-      (6): Conv2d(64, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (7): ReLU()
-    )
-    (last_conv_to_continuous_features): Sequential(
-      (0): Conv2d(64, 256, kernel_size=(4, 4), stride=(1, 1))
-      (1): ReLU()
-    )
-    (features_to_hidden_continue): Sequential(
-      (0): Linear(in_features=256, out_features=30, bias=True)
-      (1): ReLU()
-    )
-    (latent_to_features): Sequential(
-      (0): Linear(in_features=15, out_features=256, bias=True)
-      (1): ReLU()
-    )
-    (features_to_img): Sequential(
-      (0): ConvTranspose2d(256, 64, kernel_size=(4, 4), stride=(1, 1))
-      (1): ReLU()
-      (2): ConvTranspose2d(64, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (3): ReLU()
-      (4): ConvTranspose2d(64, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (5): ReLU()
-      (6): ConvTranspose2d(32, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (7): ReLU()
-      (8): ConvTranspose2d(32, 3, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (9): Sigmoid()
-    )
-  )
-)
-The number of parameters of model is 769185
-don't use continuous capacity
-=> no checkpoint found at 'trained_models/rendered_chairs/VAE_bs_64_ls_15/checkpoints/last'
-0/69092	Loss: 3015.823
-3200/69092	Loss: 2867.494
-6400/69092	Loss: 899.528
-9600/69092	Loss: 536.005
-12800/69092	Loss: 478.343
-16000/69092	Loss: 455.459
-19200/69092	Loss: 457.437
-22400/69092	Loss: 370.021
-25600/69092	Loss: 263.633
-28800/69092	Loss: 232.440
-32000/69092	Loss: 210.459
-35200/69092	Loss: 217.661
-38400/69092	Loss: 216.086
-41600/69092	Loss: 215.461
-44800/69092	Loss: 208.221
-48000/69092	Loss: 205.981
-51200/69092	Loss: 208.176
-54400/69092	Loss: 204.615
-57600/69092	Loss: 205.887
-60800/69092	Loss: 203.826
-64000/69092	Loss: 202.141
-67200/69092	Loss: 198.343
-Training time 0:03:51.208936
-Epoch: 1 Average loss: 427.71
-=> saved checkpoint 'trained_models/rendered_chairs/VAE_bs_64_ls_15/checkpoints/last' (iter 1)
-0/69092	Loss: 187.281
-3200/69092	Loss: 200.427
-6400/69092	Loss: 196.083
-9600/69092	Loss: 202.029
-12800/69092	Loss: 196.254
-16000/69092	Loss: 196.466
-19200/69092	Loss: 195.587
-22400/69092	Loss: 192.062
-25600/69092	Loss: 197.137
-28800/69092	Loss: 196.870
-32000/69092	Loss: 193.763
-35200/69092	Loss: 196.194
-38400/69092	Loss: 193.444
-41600/69092	Loss: 186.353
-44800/69092	Loss: 184.125
-48000/69092	Loss: 179.607
-51200/69092	Loss: 181.214
-54400/69092	Loss: 179.105
-57600/69092	Loss: 173.470
-60800/69092	Loss: 163.793
-64000/69092	Loss: 163.068
-67200/69092	Loss: 164.580
-Training time 0:03:46.574199
-Epoch: 2 Average loss: 186.57
-=> saved checkpoint 'trained_models/rendered_chairs/VAE_bs_64_ls_15/checkpoints/last' (iter 2)
-0/69092	Loss: 164.363
-3200/69092	Loss: 157.112
-6400/69092	Loss: 152.674
-9600/69092	Loss: 154.297
-12800/69092	Loss: 155.036
-16000/69092	Loss: 151.871
-19200/69092	Loss: 151.537
-22400/69092	Loss: 152.374
-25600/69092	Loss: 150.578
-28800/69092	Loss: 152.244
-32000/69092	Loss: 150.801
-35200/69092	Loss: 147.880
-38400/69092	Loss: 148.003
-41600/69092	Loss: 147.567
--- a/OAR.2068280.stderr
+++ b/OAR.2068280.stderr
-/data1/home/julien.dejasmin/anaconda3/envs/pytorch/lib/python3.7/site-packages/torch/nn/_reduction.py:43: UserWarning: size_average and reduce args will be deprecated, please use reduction='sum' instead.
-  warnings.warn(warning.format(ret))
-## OAR [2020-06-24 16:40:20] Job 2068280 KILLED ##
--- a/OAR.2068280.stdout
+++ b/OAR.2068280.stdout
-Namespace(batch_size=64, beta=None, ckpt_dir='checkpoints', ckpt_name='last', cont_capacity=None, dataset='rendered_chairs', disc_capacity=None, epochs=400, experiment_name='VAE_bs_64_ls_20', gpu_devices=[0, 1], is_beta_VAE=False, latent_name='', latent_spec_cont=20, latent_spec_disc=None, load_expe_name='', load_model_checkpoint=False, lr=0.0001, num_worker=4, print_loss_every=50, record_loss_every=50, save_model=True, save_reconstruction_image=False, save_step=1, verbose=True)
-creare new diretory experiment: rendered_chairs/VAE_bs_64_ls_20
-load dataset: rendered_chairs, with: 69120 train images of shape: (3, 64, 64)
-use 2 gpu who named:
-Tesla K80
-Tesla K80
-DataParallel(
-  (module): VAE(
-    (img_to_last_conv): Sequential(
-      (0): Conv2d(3, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (1): ReLU()
-      (2): Conv2d(32, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (3): ReLU()
-      (4): Conv2d(32, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (5): ReLU()
-      (6): Conv2d(64, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (7): ReLU()
-    )
-    (last_conv_to_continuous_features): Sequential(
-      (0): Conv2d(64, 256, kernel_size=(4, 4), stride=(1, 1))
-      (1): ReLU()
-    )
-    (features_to_hidden_continue): Sequential(
-      (0): Linear(in_features=256, out_features=40, bias=True)
-      (1): ReLU()
-    )
-    (latent_to_features): Sequential(
-      (0): Linear(in_features=20, out_features=256, bias=True)
-      (1): ReLU()
-    )
-    (features_to_img): Sequential(
-      (0): ConvTranspose2d(256, 64, kernel_size=(4, 4), stride=(1, 1))
-      (1): ReLU()
-      (2): ConvTranspose2d(64, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (3): ReLU()
-      (4): ConvTranspose2d(64, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (5): ReLU()
-      (6): ConvTranspose2d(32, 32, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (7): ReLU()
-      (8): ConvTranspose2d(32, 3, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1))
-      (9): Sigmoid()
-    )
-  )
-)
-The number of parameters of model is 773035
-don't use continuous capacity
-=> no checkpoint found at 'trained_models/rendered_chairs/VAE_bs_64_ls_20/checkpoints/last'
-0/69092	Loss: 2881.933
-3200/69092	Loss: 2719.794
-6400/69092	Loss: 873.853
-9600/69092	Loss: 554.953
-12800/69092	Loss: 492.325
-16000/69092	Loss: 482.132
-19200/69092	Loss: 452.762
-22400/69092	Loss: 330.168
-25600/69092	Loss: 255.698
-28800/69092	Loss: 239.942
-32000/69092	Loss: 228.030
-35200/69092	Loss: 218.689
-38400/69092	Loss: 215.508
-41600/69092	Loss: 216.039
-44800/69092	Loss: 210.728
-48000/69092	Loss: 214.618
-51200/69092	Loss: 213.710
-54400/69092	Loss: 209.650
-57600/69092	Loss: 205.299
-60800/69092	Loss: 203.167
-64000/69092	Loss: 204.888
-67200/69092	Loss: 199.354
-Training time 0:01:59.952372
-Epoch: 1 Average loss: 422.07
-=> saved checkpoint 'trained_models/rendered_chairs/VAE_bs_64_ls_20/checkpoints/last' (iter 1)
-0/69092	Loss: 213.822
-3200/69092	Loss: 195.598
-6400/69092	Loss: 188.690
-9600/69092	Loss: 189.246
-12800/69092	Loss: 185.152
-16000/69092	Loss: 182.199
-19200/69092	Loss: 178.716
-22400/69092	Loss: 175.190
-25600/69092	Loss: 173.751
-28800/69092	Loss: 173.114
-32000/69092	Loss: 169.251
-35200/69092	Loss: 165.858
-38400/69092	Loss: 162.986
-41600/69092	Loss: 160.892
-44800/69092	Loss: 157.779
-48000/69092	Loss: 159.470
-51200/69092	Loss: 155.455
-54400/69092	Loss: 153.642
-57600/69092	Loss: 152.026
-60800/69092	Loss: 154.318
-64000/69092	Loss: 149.577
-67200/69092	Loss: 149.924
-Training time 0:01:58.317271
-Epoch: 2 Average loss: 167.76
-=> saved checkpoint 'trained_models/rendered_chairs/VAE_bs_64_ls_20/checkpoints/last' (iter 2)
-0/69092	Loss: 165.835
-3200/69092	Loss: 147.308
-6400/69092	Loss: 149.297
-9600/69092	Loss: 142.005
-12800/69092	Loss: 146.004
-16000/69092	Loss: 143.574
-19200/69092	Loss: 144.306
-22400/69092	Loss: 142.412
-25600/69092	Loss: 141.625
-28800/69092	Loss: 141.184
-32000/69092	Loss: 140.178
-35200/69092	Loss: 142.290
-38400/69092	Loss: 140.434
-41600/69092	Loss: 138.432
-44800/69092	Loss: 141.047
-48000/69092	Loss: 138.527
-51200/69092	Loss: 140.962
-54400/69092	Loss: 137.973
-57600/69092	Loss: 137.226
-60800/69092	Loss: 136.121
-64000/69092	Loss: 138.238
-67200/69092	Loss: 138.245
-Training time 0:01:59.670387
-Epoch: 3 Average loss: 141.22
-=> saved checkpoint 'trained_models/rendered_chairs/VAE_bs_64_ls_20/checkpoints/last' (iter 3)
-0/69092	Loss: 142.078
-3200/69092	Loss: 136.431
-6400/69092	Loss: 135.250
-9600/69092	Loss: 136.257
-12800/69092	Loss: 135.379
-16000/69092	Loss: 135.078
-19200/69092	Loss: 134.206
-22400/69092	Loss: 135.185
-25600/69092	Loss: 132.867
-28800/69092	Loss: 137.278
-32000/69092	Loss: 134.216
-35200/69092	Loss: 134.020
-38400/69092	Loss: 130.385
-41600/69092	Loss: 135.043
-44800/69092	Loss: 133.981
-48000/69092	Loss: 132.545
-51200/69092	Loss: 135.545
-54400/69092	Loss: 134.631
-57600/69092	Loss: 131.294
-60800/69092	Loss: 132.418
-64000/69092	Loss: 131.634
-67200/69092	Loss: 133.258
-Training time 0:01:58.240577
-Epoch: 4 Average loss: 134.15
-=> saved checkpoint 'trained_models/rendered_chairs/VAE_bs_64_ls_20/checkpoints/last' (iter 4)
-0/69092	Loss: 142.708
-3200/69092	Loss: 132.561
-6400/69092	Loss: 130.874
-9600/69092	Loss: 131.468
-12800/69092	Loss: 131.411
-16000/69092	Loss: 131.824
-19200/69092	Loss: 131.005
-22400/69092	Loss: 132.192
-25600/69092	Loss: 132.671
-28800/69092	Loss: 131.997
-32000/69092	Loss: 128.867
-35200/69092	Loss: 130.574
-38400/69092	Loss: 132.202
-41600/69092	Loss: 129.819
-44800/69092	Loss: 131.265
-48000/69092	Loss: 129.098
-51200/69092	Loss: 130.616
-54400/69092	Loss: 130.498
-57600/69092	Loss: 126.580
-60800/69092	Loss: 130.212
-64000/69092	Loss: 132.173
-67200/69092	Loss: 129.980
-Training time 0:01:58.078942
-Epoch: 5 Average loss: 130.86
-=> saved checkpoint 'trained_models/rendered_chairs/VAE_bs_64_ls_20/checkpoints/last' (iter 5)
-0/69092	Loss: 124.281
-3200/69092	Loss: 128.822