Instant NGP Gets stuck when I start training
Opened this issue · 2 comments
I extracted the video data using COLMAP and stored it in a directory. The same data works very well for the model Nerfacto, but the whole system gets stuck when It comes to Instant NGP. No output is visible afterwards.
Here are my system details:
- AWS EC2 instant, currently accessed using SSH, using the Anaconda environment internally.
- Ubuntu 24.04
- Nvidia Cuda Toolkit version 12.4 (had to install this because of the GPU driver)
- Nvidia GPU: Tesla T4.
- CPU: 16 gigs RAM, 4 cores.
The only command I ran for training:
ns-train instant-ngp --data data/nerfstudio/video
According to some previous issues I read, I had to clear some cache from the ./cache folder, but that didn't work too.
Current stopping point (No output comes after this):
/home/atharavj873/.conda/envs/nerfstudio/lib/python3.11/site-packages/nerfstudio/field_components/activations.py:32: FutureWarning: torch.cuda.amp.custom_fwd(args...)is deprecated. Please usetorch.amp.custom_fwd(args..., device_type='cuda')instead. @custom_fwd(cast_inputs=torch.float32) /home/atharavj873/.conda/envs/nerfstudio/lib/python3.11/site-packages/nerfstudio/field_components/activations.py:38: FutureWarning:torch.cuda.amp.custom_bwd(args...)is deprecated. Please usetorch.amp.custom_bwd(args..., device_type='cuda')instead. @custom_bwd [05:20:10] Using --data alias for --data.pipeline.datamanager.data train.py:230 ──────────────────────────────────────────────────────── Config ──────────────────────────────────────────────────────── TrainerConfig( _target=<class 'nerfstudio.engine.trainer.Trainer'>, output_dir=PosixPath('outputs'), method_name='instant-ngp', experiment_name=None, project_name='nerfstudio-project', timestamp='2024-11-19_052010', machine=MachineConfig(seed=42, num_devices=1, num_machines=1, machine_rank=0, dist_url='auto', device_type='cuda'), logging=LoggingConfig( relative_log_dir=PosixPath('.'), steps_per_log=10, max_buffer_size=20, local_writer=LocalWriterConfig( _target=<class 'nerfstudio.utils.writer.LocalWriter'>, enable=True, stats_to_track=( <EventName.ITER_TRAIN_TIME: 'Train Iter (time)'>, <EventName.TRAIN_RAYS_PER_SEC: 'Train Rays / Sec'>, <EventName.CURR_TEST_PSNR: 'Test PSNR'>, <EventName.VIS_RAYS_PER_SEC: 'Vis Rays / Sec'>, <EventName.TEST_RAYS_PER_SEC: 'Test Rays / Sec'>, <EventName.ETA: 'ETA (time)'> ), max_log_size=10 ), profiler='basic' ), viewer=ViewerConfig( relative_log_filename='viewer_log_filename.txt', websocket_port=None, websocket_port_default=7007, websocket_host='0.0.0.0', num_rays_per_chunk=4096, max_num_display_images=512, quit_on_train_completion=False, image_format='jpeg', jpeg_quality=75, make_share_url=False, camera_frustum_scale=0.1, default_composite_depth=True ), pipeline=DynamicBatchPipelineConfig( _target=<class 'nerfstudio.pipelines.dynamic_batch.DynamicBatchPipeline'>, datamanager=VanillaDataManagerConfig( _target=<class 'nerfstudio.data.datamanagers.base_datamanager.VanillaDataManager'>, data=PosixPath('data/nerfstudio/video'), masks_on_gpu=False, images_on_gpu=False, dataparser=NerfstudioDataParserConfig( _target=<class 'nerfstudio.data.dataparsers.nerfstudio_dataparser.Nerfstudio'>, data=PosixPath('.'), scale_factor=1.0, downscale_factor=None, scene_scale=1.0, orientation_method='up', center_method='poses', auto_scale_poses=True, eval_mode='fraction', train_split_fraction=0.9, eval_interval=8, depth_unit_scale_factor=0.001, mask_color=None, load_3D_points=False ), train_num_rays_per_batch=4096, train_num_images_to_sample_from=-1, train_num_times_to_repeat_images=-1, eval_num_rays_per_batch=4096, eval_num_images_to_sample_from=-1, eval_num_times_to_repeat_images=-1, eval_image_indices=(0,), collate_fn=<function nerfstudio_collate at 0x779fc9b7c220>, camera_res_scale_factor=1.0, patch_size=1, camera_optimizer=None, pixel_sampler=PixelSamplerConfig( _target=<class 'nerfstudio.data.pixel_samplers.PixelSampler'>, num_rays_per_batch=4096, keep_full_image=False, is_equirectangular=False, ignore_mask=False, fisheye_crop_radius=None, rejection_sample_mask=True, max_num_iterations=100 ) ), model=InstantNGPModelConfig( _target=<class 'nerfstudio.models.instant_ngp.NGPModel'>, enable_collider=False, collider_params=None, loss_coefficients={'rgb_loss_coarse': 1.0, 'rgb_loss_fine': 1.0}, eval_num_rays_per_chunk=8192, prompt=None, grid_resolution=128, grid_levels=4, max_res=2048, log2_hashmap_size=19, alpha_thre=0.01, cone_angle=0.004, render_step_size=None, near_plane=0.05, far_plane=1000.0, use_gradient_scaling=False, use_appearance_embedding=False, background_color='random', disable_scene_contraction=False ), target_num_samples=262144, max_num_samples_per_ray=1024 ), optimizers={ 'fields': { 'optimizer': AdamOptimizerConfig( _target=<class 'torch.optim.adam.Adam'>, lr=0.01, eps=1e-15, max_norm=None, weight_decay=0 ), 'scheduler': ExponentialDecaySchedulerConfig( _target=<class 'nerfstudio.engine.schedulers.ExponentialDecayScheduler'>, lr_pre_warmup=1e-08, lr_final=0.0001, warmup_steps=0, max_steps=200000, ramp='cosine' ) } }, vis='viewer', data=PosixPath('data/nerfstudio/video'), prompt=None, relative_model_dir=PosixPath('nerfstudio_models'), load_scheduler=True, steps_per_save=2000, steps_per_eval_batch=500, steps_per_eval_image=500, steps_per_eval_all_images=25000, max_num_iterations=30000, mixed_precision=True, use_grad_scaler=False, save_only_latest_checkpoint=True, load_dir=None, load_step=None, load_config=None, load_checkpoint=None, log_gradients=False, gradient_accumulation_steps={}, start_paused=False ) ──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── Saving config to: outputs/video/instant-ngp/2024-11-19_052010/config.yml experiment_config.py:136 /home/atharavj873/.conda/envs/nerfstudio/lib/python3.11/site-packages/nerfstudio/engine/trainer.py:137: FutureWarning:torch.cuda.amp.GradScaler(args...)is deprecated. Please usetorch.amp.GradScaler('cuda', args...)instead. self.grad_scaler = GradScaler(enabled=self.use_grad_scaler) Saving checkpoints to: outputs/video/instant-ngp/2024-11-19_052010/nerfstudio_models trainer.py:142 Auto image downscale factor of 1 nerfstudio_dataparser.py:484 Setting up training dataset... Caching all 189 images. Setting up evaluation dataset... Caching all 21 images. /home/atharavj873/.conda/envs/nerfstudio/lib/python3.11/site-packages/torchmetrics/functional/image/lpips.py:323: FutureWarning: You are usingtorch.loadwithweights_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 forweights_onlywill be flipped toTrue. 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) ╭─────────────── viser ───────────────╮ │ ╷ │ │ HTTP │ http://0.0.0.0:7007 │ │ Websocket │ ws://0.0.0.0:7007 │ │ ╵ │ ╰─────────────────────────────────────╯ [NOTE] Not running eval iterations since only viewer is enabled. Use --vis {wandb, tensorboard, viewer+wandb, viewer+tensorboard} to run with eval. No Nerfstudio checkpoint to load, so training from scratch. Disabled comet/tensorboard/wandb event writers /home/atharavj873/.conda/envs/nerfstudio/lib/python3.11/site-packages/torch/utils/cpp_extension.py:1964: UserWarning: TORCH_CUDA_ARCH_LIST is not set, all archs for visible cards are included for compilation. If this is not desired, please set os.environ['TORCH_CUDA_ARCH_LIST']. warnings.warn( ( ●) NerfAcc: Setting up CUDA (This may take a few minutes the first time)
Have you been able to figure out your issue?
No, I wasn't able to. The last time I debugged the issue, the error was that instant ngp doesn't work on toolkit version 12.4 or greater. It needs to be a lower version.