slurmm1.0-15067.out

Lmod has detected the following error: The following module(s) are unknown:
"buildenv-gcccuda/12.1.1-gcc12.3.0"

Please check the spelling or version number. Also try "module spider ..."
It is also possible your cache file is out-of-date; it may help to try:
  $ module --ignore_cache load "buildenv-gcccuda/12.1.1-gcc12.3.0"

Also make sure that all modulefiles written in TCL start with the string
#%Module



Already on 'bimamba'
Your branch is up to date with 'origin/bimamba'.
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/hydra/_internal/defaults_list.py:251: UserWarning: In 'training': Defaults list is missing `_self_`. See https://hydra.cc/docs/1.2/upgrades/1.0_to_1.1/default_composition_order for more information
  warnings.warn(msg, UserWarning)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/hydra/_internal/defaults_list.py:251: UserWarning: In 'training': Defaults list is missing `_self_`. See https://hydra.cc/docs/1.2/upgrades/1.0_to_1.1/default_composition_order for more information
  warnings.warn(msg, UserWarning)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/hydra/_internal/defaults_list.py:251: UserWarning: In 'training': Defaults list is missing `_self_`. See https://hydra.cc/docs/1.2/upgrades/1.0_to_1.1/default_composition_order for more information
  warnings.warn(msg, UserWarning)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/hydra/_internal/defaults_list.py:251: UserWarning: In 'training': Defaults list is missing `_self_`. See https://hydra.cc/docs/1.2/upgrades/1.0_to_1.1/default_composition_order for more information
  warnings.warn(msg, UserWarning)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/hydra/_internal/defaults_list.py:251: UserWarning: In 'training': Defaults list is missing `_self_`. See https://hydra.cc/docs/1.2/upgrades/1.0_to_1.1/default_composition_order for more information
  warnings.warn(msg, UserWarning)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/hydra/_internal/defaults_list.py:251: UserWarning: In 'training': Defaults list is missing `_self_`. See https://hydra.cc/docs/1.2/upgrades/1.0_to_1.1/default_composition_order for more information
  warnings.warn(msg, UserWarning)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/hydra/_internal/defaults_list.py:251: UserWarning: In 'training': Defaults list is missing `_self_`. See https://hydra.cc/docs/1.2/upgrades/1.0_to_1.1/default_composition_order for more information
  warnings.warn(msg, UserWarning)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/hydra/_internal/defaults_list.py:251: UserWarning: In 'training': Defaults list is missing `_self_`. See https://hydra.cc/docs/1.2/upgrades/1.0_to_1.1/default_composition_order for more information
  warnings.warn(msg, UserWarning)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lightning/fabric/__init__.py:40: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lightning/fabric/__init__.py:40: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lightning/fabric/__init__.py:40: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lightning/fabric/__init__.py:40: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lightning/fabric/__init__.py:40: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lightning/fabric/__init__.py:40: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lightning/fabric/__init__.py:40: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lightning/fabric/__init__.py:40: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=AlexNet_Weights.IMAGENET1K_V1`. You can also use `weights=AlexNet_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=AlexNet_Weights.IMAGENET1K_V1`. You can also use `weights=AlexNet_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=AlexNet_Weights.IMAGENET1K_V1`. You can also use `weights=AlexNet_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=AlexNet_Weights.IMAGENET1K_V1`. You can also use `weights=AlexNet_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=AlexNet_Weights.IMAGENET1K_V1`. You can also use `weights=AlexNet_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lpips/lpips.py:107: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  self.load_state_dict(torch.load(model_path, map_location='cpu'), strict=False)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lpips/lpips.py:107: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  self.load_state_dict(torch.load(model_path, map_location='cpu'), strict=False)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lpips/lpips.py:107: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  self.load_state_dict(torch.load(model_path, map_location='cpu'), strict=False)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=AlexNet_Weights.IMAGENET1K_V1`. You can also use `weights=AlexNet_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lpips/lpips.py:107: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  self.load_state_dict(torch.load(model_path, map_location='cpu'), strict=False)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lpips/lpips.py:107: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  self.load_state_dict(torch.load(model_path, map_location='cpu'), strict=False)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=AlexNet_Weights.IMAGENET1K_V1`. You can also use `weights=AlexNet_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lpips/lpips.py:107: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  self.load_state_dict(torch.load(model_path, map_location='cpu'), strict=False)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lpips/lpips.py:107: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  self.load_state_dict(torch.load(model_path, map_location='cpu'), strict=False)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.
  warnings.warn(
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=AlexNet_Weights.IMAGENET1K_V1`. You can also use `weights=AlexNet_Weights.DEFAULT` to get the most up-to-date weights.
  warnings.warn(msg)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lpips/lpips.py:107: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  self.load_state_dict(torch.load(model_path, map_location='cpu'), strict=False)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lightning/pytorch/plugins/precision/amp.py:54: `torch.cuda.amp.GradScaler(args...)` is deprecated. Please use `torch.amp.GradScaler('cuda', args...)` instead.
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lightning/pytorch/plugins/precision/amp.py:54: `torch.cuda.amp.GradScaler(args...)` is deprecated. Please use `torch.amp.GradScaler('cuda', args...)` instead.
/proj/cvl/users/x_fahkh2/WorldMem_Repro/experiments/exp_base.py:74: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  ckpt = torch.load(checkpoint_path, map_location=torch.device('cpu'))
/proj/cvl/users/x_fahkh2/WorldMem_Repro/experiments/exp_base.py:74: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  ckpt = torch.load(checkpoint_path, map_location=torch.device('cpu'))
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lightning/pytorch/plugins/precision/amp.py:54: `torch.cuda.amp.GradScaler(args...)` is deprecated. Please use `torch.amp.GradScaler('cuda', args...)` instead.
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lightning/pytorch/plugins/precision/amp.py:54: `torch.cuda.amp.GradScaler(args...)` is deprecated. Please use `torch.amp.GradScaler('cuda', args...)` instead.
/proj/cvl/users/x_fahkh2/WorldMem_Repro/experiments/exp_base.py:74: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  ckpt = torch.load(checkpoint_path, map_location=torch.device('cpu'))
/proj/cvl/users/x_fahkh2/WorldMem_Repro/experiments/exp_base.py:74: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  ckpt = torch.load(checkpoint_path, map_location=torch.device('cpu'))
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lightning/pytorch/plugins/precision/amp.py:54: `torch.cuda.amp.GradScaler(args...)` is deprecated. Please use `torch.amp.GradScaler('cuda', args...)` instead.
/proj/cvl/users/x_fahkh2/WorldMem_Repro/experiments/exp_base.py:74: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  ckpt = torch.load(checkpoint_path, map_location=torch.device('cpu'))
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lightning/pytorch/plugins/precision/amp.py:54: `torch.cuda.amp.GradScaler(args...)` is deprecated. Please use `torch.amp.GradScaler('cuda', args...)` instead.
Outputs will be saved to: /proj/cvl/users/x_fahkh2/WorldMem_Repro/checkpoints/bimamba_stage_b_joint_ckpt_40k_fixed_v2
Will load checkpoint from /proj/cvl/users/x_fahkh2/WorldMem_Repro/checkpoints/bimamba_stage_a_40k_epoch3_step40000_oasis500m_ditreset_noxattn.ckpt
Executing task: training out of ['training']
[2026-04-22 23:27:30,977][pytorch_lightning.utilities.rank_zero][INFO] - Using 16bit Automatic Mixed Precision (AMP)
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lightning/pytorch/plugins/precision/amp.py:54: `torch.cuda.amp.GradScaler(args...)` is deprecated. Please use `torch.amp.GradScaler('cuda', args...)` instead.
/proj/cvl/users/x_fahkh2/WorldMem_Repro/experiments/exp_base.py:74: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  ckpt = torch.load(checkpoint_path, map_location=torch.device('cpu'))
[2026-04-22 23:27:31,109][pytorch_lightning.utilities.rank_zero][INFO] - GPU available: True (cuda), used: True
[2026-04-22 23:27:31,110][pytorch_lightning.utilities.rank_zero][INFO] - TPU available: False, using: 0 TPU cores
[2026-04-22 23:27:31,110][pytorch_lightning.utilities.rank_zero][INFO] - IPU available: False, using: 0 IPUs
[2026-04-22 23:27:31,110][pytorch_lightning.utilities.rank_zero][INFO] - HPU available: False, using: 0 HPUs
[2026-04-22 23:27:31,111][pytorch_lightning.utilities.rank_zero][INFO] - `Trainer(limit_val_batches=1)` was configured so 1 batch will be used.
/proj/cvl/users/x_fahkh2/WorldMem_Repro/experiments/exp_base.py:74: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  ckpt = torch.load(checkpoint_path, map_location=torch.device('cpu'))
/proj/cvl/users/x_fahkh2/envs/worldmem/lib/python3.10/site-packages/lightning/pytorch/plugins/precision/amp.py:54: `torch.cuda.amp.GradScaler(args...)` is deprecated. Please use `torch.amp.GradScaler('cuda', args...)` instead.
/proj/cvl/users/x_fahkh2/WorldMem_Repro/experiments/exp_base.py:74: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
  ckpt = torch.load(checkpoint_path, map_location=torch.device('cpu'))
[2026-04-22 23:27:45,984][pytorch_lightning.utilities.rank_zero][INFO] - Model weights loaded.
INFO: Initializing distributed: GLOBAL_RANK: 4, MEMBER: 5/8
Outputs will be saved to: /proj/cvl/users/x_fahkh2/WorldMem_Repro/checkpoints/bimamba_stage_b_joint_ckpt_40k_fixed_v2
Will load checkpoint from /proj/cvl/users/x_fahkh2/WorldMem_Repro/checkpoints/bimamba_stage_a_40k_epoch3_step40000_oasis500m_ditreset_noxattn.ckpt
Executing task: training out of ['training']
[2026-04-22 23:27:52,184][lightning.fabric.utilities.distributed][INFO] - Initializing distributed: GLOBAL_RANK: 4, MEMBER: 5/8
Outputs will be saved to: /proj/cvl/users/x_fahkh2/WorldMem_Repro/checkpoints/bimamba_stage_b_joint_ckpt_40k_fixed_v2
Will load checkpoint from /proj/cvl/users/x_fahkh2/WorldMem_Repro/checkpoints/bimamba_stage_a_40k_epoch3_step40000_oasis500m_ditreset_noxattn.ckpt
Executing task: training out of ['training']
[2026-04-22 23:27:52,652][lightning.fabric.utilities.distributed][INFO] - Initializing distributed: GLOBAL_RANK: 5, MEMBER: 6/8
INFO: Initializing distributed: GLOBAL_RANK: 5, MEMBER: 6/8
Outputs will be saved to: /proj/cvl/users/x_fahkh2/WorldMem_Repro/checkpoints/bimamba_stage_b_joint_ckpt_40k_fixed_v2
Will load checkpoint from /proj/cvl/users/x_fahkh2/WorldMem_Repro/checkpoints/bimamba_stage_a_40k_epoch3_step40000_oasis500m_ditreset_noxattn.ckpt
Executing task: training out of ['training']
[2026-04-22 23:27:52,692][lightning.fabric.utilities.distributed][INFO] - Initializing distributed: GLOBAL_RANK: 7, MEMBER: 8/8
INFO: Initializing distributed: GLOBAL_RANK: 7, MEMBER: 8/8
Outputs will be saved to: /proj/cvl/users/x_fahkh2/WorldMem_Repro/checkpoints/bimamba_stage_b_joint_ckpt_40k_fixed_v2
Will load checkpoint from /proj/cvl/users/x_fahkh2/WorldMem_Repro/checkpoints/bimamba_stage_a_40k_epoch3_step40000_oasis500m_ditreset_noxattn.ckpt
Executing task: training out of ['training']
[2026-04-22 23:27:52,714][lightning.fabric.utilities.distributed][INFO] - Initializing distributed: GLOBAL_RANK: 6, MEMBER: 7/8
INFO: Initializing distributed: GLOBAL_RANK: 6, MEMBER: 7/8
INFO: Initializing distributed: GLOBAL_RANK: 0, MEMBER: 1/8
[2026-04-22 23:27:52,747][lightning.fabric.utilities.distributed][INFO] - Initializing distributed: GLOBAL_RANK: 0, MEMBER: 1/8
INFO: Initializing distributed: GLOBAL_RANK: 3, MEMBER: 4/8
Outputs will be saved to: /proj/cvl/users/x_fahkh2/WorldMem_Repro/checkpoints/bimamba_stage_b_joint_ckpt_40k_fixed_v2
Will load checkpoint from /proj/cvl/users/x_fahkh2/WorldMem_Repro/checkpoints/bimamba_stage_a_40k_epoch3_step40000_oasis500m_ditreset_noxattn.ckpt
Executing task: training out of ['training']
[2026-04-22 23:27:52,747][lightning.fabric.utilities.distributed][INFO] - Initializing distributed: GLOBAL_RANK: 3, MEMBER: 4/8
INFO: Initializing distributed: GLOBAL_RANK: 2, MEMBER: 3/8
Outputs will be saved to: /proj/cvl/users/x_fahkh2/WorldMem_Repro/checkpoints/bimamba_stage_b_joint_ckpt_40k_fixed_v2
Will load checkpoint from /proj/cvl/users/x_fahkh2/WorldMem_Repro/checkpoints/bimamba_stage_a_40k_epoch3_step40000_oasis500m_ditreset_noxattn.ckpt
Executing task: training out of ['training']
[2026-04-22 23:27:52,971][lightning.fabric.utilities.distributed][INFO] - Initializing distributed: GLOBAL_RANK: 2, MEMBER: 3/8
INFO: Initializing distributed: GLOBAL_RANK: 1, MEMBER: 2/8
Outputs will be saved to: /proj/cvl/users/x_fahkh2/WorldMem_Repro/checkpoints/bimamba_stage_b_joint_ckpt_40k_fixed_v2
Will load checkpoint from /proj/cvl/users/x_fahkh2/WorldMem_Repro/checkpoints/bimamba_stage_a_40k_epoch3_step40000_oasis500m_ditreset_noxattn.ckpt
Executing task: training out of ['training']
[2026-04-22 23:27:53,603][lightning.fabric.utilities.distributed][INFO] - Initializing distributed: GLOBAL_RANK: 1, MEMBER: 2/8
[2026-04-22 23:27:58,583][pytorch_lightning.utilities.rank_zero][INFO] - ----------------------------------------------------------------------------------------------------
distributed_backend=nccl
All distributed processes registered. Starting with 8 processes
----------------------------------------------------------------------------------------------------

wandb: WARNING `resume` will be ignored since W&B syncing is set to `offline`. Starting a new run with run id 6th79dex.
wandb: Tracking run with wandb version 0.17.9
wandb: W&B syncing is set to `offline` in this directory.  
wandb: Run `wandb online` or set WANDB_MODE=online to enable cloud syncing.
INFO: LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0,1,2,3,4,5,6,7]
[2026-04-22 23:28:17,001][lightning.pytorch.accelerators.cuda][INFO] - LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0,1,2,3,4,5,6,7]
[2026-04-22 23:28:17,001][lightning.pytorch.accelerators.cuda][INFO] - LOCAL_RANK: 2 - CUDA_VISIBLE_DEVICES: [0,1,2,3,4,5,6,7]
INFO: LOCAL_RANK: 2 - CUDA_VISIBLE_DEVICES: [0,1,2,3,4,5,6,7]
[2026-04-22 23:28:17,001][lightning.pytorch.accelerators.cuda][INFO] - LOCAL_RANK: 3 - CUDA_VISIBLE_DEVICES: [0,1,2,3,4,5,6,7]
INFO: LOCAL_RANK: 3 - CUDA_VISIBLE_DEVICES: [0,1,2,3,4,5,6,7]
[2026-04-22 23:28:17,001][lightning.pytorch.accelerators.cuda][INFO] - LOCAL_RANK: 4 - CUDA_VISIBLE_DEVICES: [0,1,2,3,4,5,6,7]
INFO: LOCAL_RANK: 4 - CUDA_VISIBLE_DEVICES: [0,1,2,3,4,5,6,7]
[2026-04-22 23:28:17,001][lightning.pytorch.accelerators.cuda][INFO] - LOCAL_RANK: 5 - CUDA_VISIBLE_DEVICES: [0,1,2,3,4,5,6,7]
INFO: LOCAL_RANK: 5 - CUDA_VISIBLE_DEVICES: [0,1,2,3,4,5,6,7]
INFO: LOCAL_RANK: 7 - CUDA_VISIBLE_DEVICES: [0,1,2,3,4,5,6,7]
[2026-04-22 23:28:17,001][lightning.pytorch.accelerators.cuda][INFO] - LOCAL_RANK: 7 - CUDA_VISIBLE_DEVICES: [0,1,2,3,4,5,6,7]
[2026-04-22 23:28:17,001][lightning.pytorch.accelerators.cuda][INFO] - LOCAL_RANK: 6 - CUDA_VISIBLE_DEVICES: [0,1,2,3,4,5,6,7]
INFO: LOCAL_RANK: 6 - CUDA_VISIBLE_DEVICES: [0,1,2,3,4,5,6,7]
[2026-04-22 23:28:17,002][lightning.pytorch.accelerators.cuda][INFO] - LOCAL_RANK: 1 - CUDA_VISIBLE_DEVICES: [0,1,2,3,4,5,6,7]
INFO: LOCAL_RANK: 1 - CUDA_VISIBLE_DEVICES: [0,1,2,3,4,5,6,7]
INFO: 
  | Name                   | Type                                  | Params
---------------------------------------------------------------------------------
0 | diffusion_model        | DiffusionMamba                        | 609 M 
1 | validation_lpips_model | LearnedPerceptualImagePatchSimilarity | 2.5 M 
2 | vae                    | AutoencoderKL                         | 229 M 
3 | mamba_memory           | BiMambaMemory                         | 4.5 M 
---------------------------------------------------------------------------------
614 M     Trainable params
231 M     Non-trainable params
845 M     Total params
3,383.355 Total estimated model params size (MB)
[2026-04-22 23:28:20,561][lightning.pytorch.callbacks.model_summary][INFO] - 
  | Name                   | Type                                  | Params
---------------------------------------------------------------------------------
0 | diffusion_model        | DiffusionMamba                        | 609 M 
1 | validation_lpips_model | LearnedPerceptualImagePatchSimilarity | 2.5 M 
2 | vae                    | AutoencoderKL                         | 229 M 
3 | mamba_memory           | BiMambaMemory                         | 4.5 M 
---------------------------------------------------------------------------------
614 M     Trainable params
231 M     Non-trainable params
845 M     Total params
3,383.355 Total estimated model params size (MB)
INFO: SLURM auto-requeueing enabled. Setting signal handlers.
[2026-04-22 23:28:21,287][lightning.pytorch.trainer.connectors.signal_connector][INFO] - SLURM auto-requeueing enabled. Setting signal handlers.
INFO: SLURM auto-requeueing enabled. Setting signal handlers.
[2026-04-22 23:28:21,288][lightning.pytorch.trainer.connectors.signal_connector][INFO] - SLURM auto-requeueing enabled. Setting signal handlers.
[2026-04-22 23:28:21,288][lightning.pytorch.trainer.connectors.signal_connector][INFO] - SLURM auto-requeueing enabled. Setting signal handlers.
INFO: SLURM auto-requeueing enabled. Setting signal handlers.
[2026-04-22 23:28:21,288][lightning.pytorch.trainer.connectors.signal_connector][INFO] - SLURM auto-requeueing enabled. Setting signal handlers.
INFO: SLURM auto-requeueing enabled. Setting signal handlers.
[2026-04-22 23:28:21,288][lightning.pytorch.trainer.connectors.signal_connector][INFO] - SLURM auto-requeueing enabled. Setting signal handlers.
INFO: SLURM auto-requeueing enabled. Setting signal handlers.
INFO: SLURM auto-requeueing enabled. Setting signal handlers.
[2026-04-22 23:28:21,288][lightning.pytorch.trainer.connectors.signal_connector][INFO] - SLURM auto-requeueing enabled. Setting signal handlers.
[2026-04-22 23:28:21,288][lightning.pytorch.trainer.connectors.signal_connector][INFO] - SLURM auto-requeueing enabled. Setting signal handlers.
INFO: SLURM auto-requeueing enabled. Setting signal handlers.
[2026-04-22 23:28:21,288][lightning.pytorch.trainer.connectors.signal_connector][INFO] - SLURM auto-requeueing enabled. Setting signal handlers.
INFO: SLURM auto-requeueing enabled. Setting signal handlers.
/proj/cvl/users/x_fahkh2/WorldMem_Repro/algorithms/worldmem/models/mamba_memory.py:173: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with torch.cuda.amp.autocast(enabled=False):
/proj/cvl/users/x_fahkh2/WorldMem_Repro/algorithms/worldmem/models/mamba_memory.py:173: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with torch.cuda.amp.autocast(enabled=False):
/proj/cvl/users/x_fahkh2/WorldMem_Repro/algorithms/worldmem/models/mamba_memory.py:173: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with torch.cuda.amp.autocast(enabled=False):
/proj/cvl/users/x_fahkh2/WorldMem_Repro/algorithms/worldmem/models/mamba_memory.py:173: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with torch.cuda.amp.autocast(enabled=False):
/proj/cvl/users/x_fahkh2/WorldMem_Repro/algorithms/worldmem/models/mamba_memory.py:173: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with torch.cuda.amp.autocast(enabled=False):
/proj/cvl/users/x_fahkh2/WorldMem_Repro/algorithms/worldmem/models/mamba_memory.py:173: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with torch.cuda.amp.autocast(enabled=False):
/proj/cvl/users/x_fahkh2/WorldMem_Repro/algorithms/worldmem/models/mamba_memory.py:173: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with torch.cuda.amp.autocast(enabled=False):
/proj/cvl/users/x_fahkh2/WorldMem_Repro/algorithms/worldmem/models/mamba_memory.py:173: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
  with torch.cuda.amp.autocast(enabled=False):

Training: |          | 0/? [00:00<?, ?it/s]
Training:   0%|          | 0/203307 [00:00<?, ?it/s]
Epoch 0:   0%|          | 0/203307 [00:00<?, ?it/s] 
Epoch 0:   0%|          | 1/203307 [00:20<1131:12:38,  0.05it/s]
Epoch 0:   0%|          | 1/203307 [00:20<1131:58:24,  0.05it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 2/203307 [00:23<672:30:30,  0.08it/s, v_num=9dex, training/loss=0.186] 
Epoch 0:   0%|          | 2/203307 [00:23<672:51:51,  0.08it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 3/203307 [00:27<518:10:12,  0.11it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 3/203307 [00:27<518:27:08,  0.11it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 4/203307 [00:31<444:35:40,  0.13it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 4/203307 [00:31<444:37:36,  0.13it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 5/203307 [00:35<400:40:23,  0.14it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 5/203307 [00:35<400:41:38,  0.14it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 6/203307 [00:39<372:52:33,  0.15it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 6/203307 [00:39<372:59:02,  0.15it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 7/203307 [00:45<363:09:40,  0.16it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 7/203307 [00:45<363:11:19,  0.16it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 8/203307 [00:49<346:08:33,  0.16it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 8/203307 [00:49<346:08:53,  0.16it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 9/203307 [00:53<333:13:59,  0.17it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 9/203307 [00:53<333:14:19,  0.17it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 10/203307 [00:57<322:39:13,  0.18it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 10/203307 [00:57<322:39:28,  0.18it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 11/203307 [01:00<312:45:30,  0.18it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 11/203307 [01:00<312:49:31,  0.18it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 12/203307 [01:04<305:18:34,  0.18it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 12/203307 [01:04<305:18:54,  0.18it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 13/203307 [01:08<299:11:13,  0.19it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 13/203307 [01:08<299:11:29,  0.19it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 14/203307 [01:13<297:36:04,  0.19it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 14/203307 [01:13<297:36:20,  0.19it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 15/203307 [01:18<296:10:18,  0.19it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 15/203307 [01:18<296:11:05,  0.19it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 16/203307 [01:22<291:18:34,  0.19it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 16/203307 [01:22<291:18:46,  0.19it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 17/203307 [01:26<287:04:05,  0.20it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 17/203307 [01:26<287:04:17,  0.20it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 18/203307 [01:30<282:55:46,  0.20it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 18/203307 [01:30<282:55:54,  0.20it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 19/203307 [01:34<279:33:37,  0.20it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 19/203307 [01:34<279:33:47,  0.20it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 20/203307 [01:38<276:42:18,  0.20it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 20/203307 [01:38<276:42:33,  0.20it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 21/203307 [01:41<273:52:26,  0.21it/s, v_num=9dex, training/loss=0.186]
Epoch 0:   0%|          | 21/203307 [01:41<273:52:35,  0.21it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 22/203307 [01:46<273:44:43,  0.21it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 22/203307 [01:46<273:44:50,  0.21it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 23/203307 [01:51<273:38:41,  0.21it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 23/203307 [01:51<273:38:47,  0.21it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 24/203307 [01:55<271:26:05,  0.21it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 24/203307 [01:55<271:26:14,  0.21it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 25/203307 [01:59<269:40:34,  0.21it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 25/203307 [01:59<269:40:44,  0.21it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 26/203307 [02:03<268:05:04,  0.21it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 26/203307 [02:03<268:05:11,  0.21it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 27/203307 [02:07<266:12:20,  0.21it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 27/203307 [02:07<266:12:27,  0.21it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 28/203307 [02:11<264:43:14,  0.21it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 28/203307 [02:11<264:43:21,  0.21it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 29/203307 [02:15<263:12:40,  0.21it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 29/203307 [02:15<263:12:48,  0.21it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 30/203307 [02:19<261:54:37,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 30/203307 [02:19<261:56:05,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 31/203307 [02:23<262:12:41,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 31/203307 [02:23<262:13:20,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 32/203307 [02:28<261:12:37,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 32/203307 [02:28<261:12:44,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 33/203307 [02:31<259:46:25,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 33/203307 [02:31<259:47:37,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 34/203307 [02:35<258:35:24,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 34/203307 [02:35<258:36:39,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 35/203307 [02:39<257:28:55,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 35/203307 [02:39<257:30:14,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 36/203307 [02:43<256:40:31,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 36/203307 [02:43<256:41:41,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 37/203307 [02:47<255:39:48,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 37/203307 [02:47<255:40:51,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 38/203307 [02:51<255:07:21,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 38/203307 [02:51<255:08:18,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 39/203307 [02:56<255:16:19,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 39/203307 [02:56<255:17:17,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 40/203307 [03:01<255:41:07,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 40/203307 [03:01<255:42:05,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 41/203307 [03:05<255:09:38,  0.22it/s, v_num=9dex, training/loss=0.174]
Epoch 0:   0%|          | 41/203307 [03:05<255:09:42,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 42/203307 [03:09<254:42:42,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 42/203307 [03:09<254:43:32,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 43/203307 [03:13<253:52:16,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 43/203307 [03:13<253:53:13,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 44/203307 [03:17<253:16:16,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 44/203307 [03:17<253:17:12,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 45/203307 [03:21<252:37:44,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 45/203307 [03:21<252:38:30,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 46/203307 [03:25<252:32:30,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 46/203307 [03:25<252:33:21,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 47/203307 [03:30<252:42:16,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 47/203307 [03:30<252:43:10,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 48/203307 [03:35<253:17:15,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 48/203307 [03:35<253:17:18,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 49/203307 [03:39<252:48:18,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 49/203307 [03:39<252:48:59,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 50/203307 [03:43<252:35:18,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 50/203307 [03:43<252:35:45,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 51/203307 [03:47<251:54:54,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 51/203307 [03:47<251:55:37,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 52/203307 [03:51<251:32:30,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 52/203307 [03:51<251:32:34,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 53/203307 [03:55<250:53:55,  0.23it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 53/203307 [03:55<250:53:58,  0.23it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 54/203307 [03:59<250:50:10,  0.23it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 54/203307 [03:59<250:50:14,  0.23it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 55/203307 [04:04<251:02:35,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 55/203307 [04:04<251:02:40,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 56/203307 [04:09<251:34:12,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 56/203307 [04:09<251:34:21,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 57/203307 [04:13<251:03:27,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 57/203307 [04:13<251:04:03,  0.22it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 58/203307 [04:17<250:44:48,  0.23it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 58/203307 [04:17<250:44:51,  0.23it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 59/203307 [04:21<250:11:56,  0.23it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 59/203307 [04:21<250:12:17,  0.23it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 60/203307 [04:25<250:00:23,  0.23it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 60/203307 [04:25<250:00:47,  0.23it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 61/203307 [04:29<249:38:01,  0.23it/s, v_num=9dex, training/loss=0.160]
Epoch 0:   0%|          | 61/203307 [04:29<249:38:33,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 62/203307 [04:33<249:19:00,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 62/203307 [04:33<249:19:38,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 63/203307 [04:37<248:53:54,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 63/203307 [04:37<248:53:57,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 64/203307 [04:42<249:27:50,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 64/203307 [04:42<249:27:52,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 65/203307 [04:47<249:33:08,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 65/203307 [04:47<249:33:11,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 66/203307 [04:51<249:17:19,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 66/203307 [04:51<249:17:21,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 67/203307 [04:55<248:50:36,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 67/203307 [04:55<248:51:13,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 68/203307 [04:59<248:34:30,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 68/203307 [04:59<248:34:37,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 69/203307 [05:03<248:18:09,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 69/203307 [05:03<248:18:38,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 70/203307 [05:07<248:04:18,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 70/203307 [05:07<248:04:45,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 71/203307 [05:11<247:42:32,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 71/203307 [05:11<247:42:51,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 72/203307 [05:16<248:14:52,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 72/203307 [05:16<248:14:54,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 73/203307 [05:21<248:22:06,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 73/203307 [05:21<248:22:35,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 74/203307 [05:25<248:07:57,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 74/203307 [05:25<248:07:59,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 75/203307 [05:29<247:55:17,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 75/203307 [05:29<247:55:19,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 76/203307 [05:33<247:43:41,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 76/203307 [05:33<247:43:43,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 77/203307 [05:37<247:31:21,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 77/203307 [05:37<247:31:34,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 78/203307 [05:41<247:16:34,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 78/203307 [05:41<247:17:00,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 79/203307 [05:45<247:00:45,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 79/203307 [05:45<247:01:09,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 80/203307 [05:50<247:34:11,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 80/203307 [05:50<247:34:17,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 81/203307 [05:55<247:45:11,  0.23it/s, v_num=9dex, training/loss=0.168]
Epoch 0:   0%|          | 81/203307 [05:55<247:45:39,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 82/203307 [06:00<248:00:06,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 82/203307 [06:00<248:00:09,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 83/203307 [06:04<247:43:58,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 83/203307 [06:04<247:44:00,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 84/203307 [06:08<247:41:01,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 84/203307 [06:08<247:41:04,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 85/203307 [06:12<247:30:38,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 85/203307 [06:12<247:30:41,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 86/203307 [06:16<247:26:10,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 86/203307 [06:16<247:26:15,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 87/203307 [06:21<247:14:54,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 87/203307 [06:21<247:15:17,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 88/203307 [06:25<247:12:29,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 88/203307 [06:25<247:12:51,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 89/203307 [06:30<247:38:58,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 89/203307 [06:30<247:39:21,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 90/203307 [06:35<247:55:21,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 90/203307 [06:35<247:55:45,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 91/203307 [06:39<247:39:45,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 91/203307 [06:39<247:40:10,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 92/203307 [06:43<247:38:28,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 92/203307 [06:43<247:38:50,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 93/203307 [06:47<247:29:11,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 93/203307 [06:47<247:29:33,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 94/203307 [06:51<247:23:36,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 94/203307 [06:51<247:23:57,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 95/203307 [06:55<247:08:57,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 95/203307 [06:55<247:09:22,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 96/203307 [07:00<247:05:37,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 96/203307 [07:00<247:05:58,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 97/203307 [07:05<247:26:36,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 97/203307 [07:05<247:26:56,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 98/203307 [07:09<247:24:25,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 98/203307 [07:09<247:24:44,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 99/203307 [07:14<247:32:13,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 99/203307 [07:14<247:32:39,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 100/203307 [07:18<247:29:48,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 100/203307 [07:18<247:29:55,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 101/203307 [07:22<247:25:44,  0.23it/s, v_num=9dex, training/loss=0.154]
Epoch 0:   0%|          | 101/203307 [07:22<247:25:45,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 102/203307 [07:26<247:21:37,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 102/203307 [07:26<247:21:38,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 103/203307 [07:31<247:12:47,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 103/203307 [07:31<247:13:11,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 104/203307 [07:35<247:13:07,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 104/203307 [07:35<247:13:26,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 105/203307 [07:40<247:33:31,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 105/203307 [07:40<247:33:53,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 106/203307 [07:44<247:28:54,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 106/203307 [07:44<247:29:13,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 107/203307 [07:49<247:36:31,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 107/203307 [07:49<247:36:48,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 108/203307 [07:53<247:36:43,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 108/203307 [07:53<247:37:04,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 109/203307 [07:57<247:23:27,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 109/203307 [07:57<247:23:29,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 110/203307 [08:01<247:13:40,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 110/203307 [08:01<247:14:00,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 111/203307 [08:05<247:00:15,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 111/203307 [08:05<247:00:37,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 112/203307 [08:09<246:51:12,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 112/203307 [08:09<246:51:34,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 113/203307 [08:14<247:11:49,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 113/203307 [08:14<247:12:09,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 114/203307 [08:19<247:25:37,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 114/203307 [08:19<247:25:39,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 115/203307 [08:23<247:13:23,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 115/203307 [08:23<247:13:25,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 116/203307 [08:28<247:23:38,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 116/203307 [08:28<247:23:55,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 117/203307 [08:32<247:13:16,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 117/203307 [08:32<247:13:29,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 118/203307 [08:36<247:03:23,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 118/203307 [08:36<247:03:40,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 119/203307 [08:40<246:49:21,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 119/203307 [08:40<246:49:41,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 120/203307 [08:44<246:46:38,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 120/203307 [08:44<246:46:56,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 121/203307 [08:49<247:02:12,  0.23it/s, v_num=9dex, training/loss=0.153]
Epoch 0:   0%|          | 121/203307 [08:49<247:02:27,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 122/203307 [08:54<247:09:33,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 122/203307 [08:54<247:09:50,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 123/203307 [08:58<246:59:57,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 123/203307 [08:58<247:00:14,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 124/203307 [09:03<247:11:19,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 124/203307 [09:03<247:11:35,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 125/203307 [09:07<247:01:19,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 125/203307 [09:07<247:01:36,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 126/203307 [09:11<246:58:38,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 126/203307 [09:11<246:58:57,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 127/203307 [09:15<246:48:45,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 127/203307 [09:15<246:49:03,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 128/203307 [09:19<246:46:14,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 128/203307 [09:19<246:46:32,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 129/203307 [09:24<247:03:54,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 129/203307 [09:24<247:04:12,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 130/203307 [09:28<247:00:21,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 130/203307 [09:28<247:00:51,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 131/203307 [09:33<247:06:43,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 131/203307 [09:33<247:07:00,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 132/203307 [09:37<247:05:34,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 132/203307 [09:37<247:05:51,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 133/203307 [09:42<247:12:51,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 133/203307 [09:42<247:13:06,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 134/203307 [09:46<247:09:42,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 134/203307 [09:46<247:09:59,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 135/203307 [09:50<247:03:06,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 135/203307 [09:50<247:03:29,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 136/203307 [09:55<246:56:56,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 136/203307 [09:55<246:57:11,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 137/203307 [09:59<246:46:31,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 137/203307 [09:59<246:46:46,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 138/203307 [10:04<247:04:08,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 138/203307 [10:04<247:04:24,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 139/203307 [10:08<247:09:45,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 139/203307 [10:08<247:09:59,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 140/203307 [10:13<247:06:30,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 140/203307 [10:13<247:06:46,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 141/203307 [10:17<247:15:30,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 141/203307 [10:17<247:15:47,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 142/203307 [10:22<247:15:18,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 142/203307 [10:22<247:15:35,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 143/203307 [10:26<247:08:16,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 143/203307 [10:26<247:08:33,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 144/203307 [10:30<247:08:17,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 144/203307 [10:30<247:08:31,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 145/203307 [10:34<246:59:58,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 145/203307 [10:34<247:00:13,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 146/203307 [10:39<247:14:44,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 146/203307 [10:39<247:15:00,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 147/203307 [10:44<247:20:39,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 147/203307 [10:44<247:20:55,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 148/203307 [10:48<247:20:31,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 148/203307 [10:48<247:20:45,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 149/203307 [10:52<247:16:30,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 149/203307 [10:52<247:16:43,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 150/203307 [10:57<247:25:08,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 150/203307 [10:57<247:25:21,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 151/203307 [11:01<247:18:52,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 151/203307 [11:01<247:19:05,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 152/203307 [11:05<247:15:06,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 152/203307 [11:05<247:15:18,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 153/203307 [11:10<247:10:41,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 153/203307 [11:10<247:10:56,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 154/203307 [11:15<247:26:58,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 154/203307 [11:15<247:27:00,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 155/203307 [11:19<247:31:55,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 155/203307 [11:19<247:32:10,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 156/203307 [11:24<247:29:39,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 156/203307 [11:24<247:29:53,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 157/203307 [11:28<247:23:55,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 157/203307 [11:28<247:24:08,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 158/203307 [11:33<247:33:06,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 158/203307 [11:33<247:33:18,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 159/203307 [11:37<247:24:57,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 159/203307 [11:37<247:25:09,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 160/203307 [11:41<247:23:08,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 160/203307 [11:41<247:23:22,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 161/203307 [11:45<247:17:23,  0.23it/s, v_num=9dex, training/loss=0.145]
Epoch 0:   0%|          | 161/203307 [11:45<247:17:36,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 162/203307 [11:50<247:25:52,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 162/203307 [11:50<247:26:07,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 163/203307 [11:55<247:37:55,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 163/203307 [11:55<247:38:09,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 164/203307 [12:00<247:45:09,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 164/203307 [12:00<247:45:22,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 165/203307 [12:04<247:38:27,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 165/203307 [12:04<247:38:40,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 166/203307 [12:08<247:34:23,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 166/203307 [12:08<247:34:36,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 167/203307 [12:12<247:39:15,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 167/203307 [12:12<247:39:29,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 168/203307 [12:17<247:36:51,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 168/203307 [12:17<247:37:03,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 169/203307 [12:21<247:31:57,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 169/203307 [12:21<247:32:08,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 170/203307 [12:26<247:38:41,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 170/203307 [12:26<247:38:54,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 171/203307 [12:30<247:43:48,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 171/203307 [12:30<247:43:58,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 172/203307 [12:35<247:48:29,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 172/203307 [12:35<247:48:42,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 173/203307 [12:39<247:48:21,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 173/203307 [12:39<247:48:34,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 174/203307 [12:43<247:44:37,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 174/203307 [12:43<247:44:49,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 175/203307 [12:48<247:50:41,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 175/203307 [12:48<247:50:53,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 176/203307 [12:52<247:46:06,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 176/203307 [12:52<247:46:15,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 177/203307 [12:56<247:38:55,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 177/203307 [12:56<247:39:07,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 178/203307 [13:01<247:43:01,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 178/203307 [13:01<247:43:13,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 179/203307 [13:05<247:37:42,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 179/203307 [13:05<247:37:55,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 180/203307 [13:10<247:52:57,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 180/203307 [13:10<247:53:09,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 181/203307 [13:15<247:57:39,  0.23it/s, v_num=9dex, training/loss=0.142]
Epoch 0:   0%|          | 181/203307 [13:15<247:57:51,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 182/203307 [13:19<247:52:21,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 182/203307 [13:19<247:52:33,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 183/203307 [13:23<247:44:11,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 183/203307 [13:23<247:44:24,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 184/203307 [13:28<247:51:16,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 184/203307 [13:28<247:51:28,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 185/203307 [13:32<247:46:14,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 185/203307 [13:32<247:46:27,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 186/203307 [13:36<247:41:42,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 186/203307 [13:36<247:41:53,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 187/203307 [13:40<247:39:43,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 187/203307 [13:40<247:39:54,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 188/203307 [13:45<247:48:47,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 188/203307 [13:45<247:48:49,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 189/203307 [13:50<247:52:21,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 189/203307 [13:50<247:52:32,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 190/203307 [13:54<247:48:34,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 190/203307 [13:54<247:48:45,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 191/203307 [13:58<247:42:16,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 191/203307 [13:58<247:42:27,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 192/203307 [14:03<247:48:51,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 192/203307 [14:03<247:49:03,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 193/203307 [14:07<247:44:18,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 193/203307 [14:07<247:44:27,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 194/203307 [14:11<247:39:04,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 194/203307 [14:11<247:39:14,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 195/203307 [14:16<247:40:43,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 195/203307 [14:16<247:40:55,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 196/203307 [14:20<247:35:18,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 196/203307 [14:20<247:35:28,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 197/203307 [14:25<247:48:04,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 197/203307 [14:25<247:48:14,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 198/203307 [14:29<247:52:51,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 198/203307 [14:29<247:53:03,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 199/203307 [14:33<247:46:16,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 199/203307 [14:33<247:46:26,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 200/203307 [14:38<247:40:40,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 200/203307 [14:38<247:40:46,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 201/203307 [14:42<247:45:03,  0.23it/s, v_num=9dex, training/loss=0.157]
Epoch 0:   0%|          | 201/203307 [14:42<247:45:13,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 202/203307 [14:46<247:39:29,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 202/203307 [14:46<247:39:38,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 203/203307 [14:51<247:38:43,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 203/203307 [14:51<247:38:52,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 204/203307 [14:55<247:32:42,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 204/203307 [14:55<247:32:53,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 205/203307 [15:00<247:41:52,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 205/203307 [15:00<247:42:01,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 206/203307 [15:04<247:46:51,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 206/203307 [15:04<247:47:00,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 207/203307 [15:08<247:40:07,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 207/203307 [15:08<247:40:17,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 208/203307 [15:13<247:43:38,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 208/203307 [15:13<247:43:52,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 209/203307 [15:17<247:35:24,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 209/203307 [15:17<247:35:35,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 210/203307 [15:21<247:32:12,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 210/203307 [15:21<247:32:22,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 211/203307 [15:25<247:29:58,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 211/203307 [15:25<247:30:09,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 212/203307 [15:29<247:26:12,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 212/203307 [15:29<247:26:21,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 213/203307 [15:34<247:27:45,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 213/203307 [15:34<247:27:55,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 214/203307 [15:39<247:39:14,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 214/203307 [15:39<247:39:24,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 215/203307 [15:43<247:41:53,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 215/203307 [15:44<247:42:03,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 216/203307 [15:48<247:36:34,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 216/203307 [15:48<247:36:45,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 217/203307 [15:52<247:31:48,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 217/203307 [15:52<247:31:59,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 218/203307 [15:56<247:34:02,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 218/203307 [15:56<247:34:13,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 219/203307 [16:01<247:35:10,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 219/203307 [16:01<247:35:20,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 220/203307 [16:05<247:30:59,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 220/203307 [16:05<247:31:10,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 221/203307 [16:09<247:25:21,  0.23it/s, v_num=9dex, training/loss=0.144]
Epoch 0:   0%|          | 221/203307 [16:09<247:25:22,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 222/203307 [16:14<247:31:54,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 222/203307 [16:14<247:32:04,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 223/203307 [16:18<247:35:48,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 223/203307 [16:18<247:35:58,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 224/203307 [16:22<247:29:22,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 224/203307 [16:22<247:29:31,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 225/203307 [16:27<247:31:53,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 225/203307 [16:27<247:32:03,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 226/203307 [16:31<247:28:28,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 226/203307 [16:31<247:28:38,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 227/203307 [16:35<247:21:10,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 227/203307 [16:35<247:21:20,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 228/203307 [16:39<247:22:39,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 228/203307 [16:39<247:22:50,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 229/203307 [16:44<247:24:50,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 229/203307 [16:44<247:25:00,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 230/203307 [16:48<247:18:01,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 230/203307 [16:48<247:18:09,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 231/203307 [16:53<247:26:16,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 231/203307 [16:53<247:26:25,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 232/203307 [16:57<247:27:25,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 232/203307 [16:57<247:27:36,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 233/203307 [17:01<247:22:25,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 233/203307 [17:01<247:22:34,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 234/203307 [17:06<247:19:58,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 234/203307 [17:06<247:20:08,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 235/203307 [17:10<247:19:01,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 235/203307 [17:10<247:19:09,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 236/203307 [17:14<247:21:09,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 236/203307 [17:14<247:21:20,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 237/203307 [17:18<247:16:17,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 237/203307 [17:18<247:16:26,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 238/203307 [17:22<247:11:05,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 238/203307 [17:22<247:11:12,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 239/203307 [17:27<247:17:12,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 239/203307 [17:27<247:17:21,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 240/203307 [17:32<247:20:07,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 240/203307 [17:32<247:20:17,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 241/203307 [17:36<247:16:39,  0.23it/s, v_num=9dex, training/loss=0.137]
Epoch 0:   0%|          | 241/203307 [17:36<247:16:48,  0.23it/s, v_num=9dex, training/loss=0.130]
Epoch 0:   0%|          | 242/203307 [17:41<247:20:28,  0.23it/s, v_num=9dex, training/loss=0.130]
Epoch 0:   0%|          | 242/203307 [17:41<247:20:37,  0.23it/s, v_num=9dex, training/loss=0.130]
Epoch 0:   0%|          | 243/203307 [17:45<247:16:10,  0.23it/s, v_num=9dex, training/loss=0.130]
Epoch 0:   0%|          | 243/203307 [17:45<247:16:19,  0.23it/s, v_num=9dex, training/loss=0.130]
Epoch 0:   0%|          | 244/203307 [17:49<247:20:10,  0.23it/s, v_num=9dex, training/loss=0.130]
Epoch 0: