yerfor
RTX4090
@tailangjun
我用的docker，搭建环境运行代码老是报错，难道docker和实体机有区别吗
(geneface) root@autodl-container-b37a11a83c-cb82e0e6:~/autodl-tmp/GeneFace-1.1.0# bash scripts/infer_lm3d_radnerf.sh
/root/miniconda3/envs/geneface/lib/python3.9/site-packages/torchvision/io/image.py:13: UserWarning: Failed to load image Python extension: '/root/miniconda3/envs/geneface/lib/python3.9/site-packages/torchvision/image.so: undefined symbol: _ZNK3c107SymBool10guard_boolEPKcl'If you don't plan on using image functionality from torchvision.io, you can ignore this warning. Otherwise, there might be something wrong with your environment. Did you have libjpeg or libpng installed before building torchvision from source?
warn(
/root/miniconda3/envs/geneface/lib/python3.9/site-packages/torch/amp/autocast_mode.py:198: UserWarning: User provided device_type of 'cuda', but CUDA is not available. Disabling
warnings.warn('User provided device_type of 'cuda', but CUDA is not available. Disabling')
| Unknow hparams: []
| Hparams chains: ['egs/egs_bases/radnerf/base.yaml', 'egs/egs_bases/radnerf/lm3d_radnerf.yaml', 'checkpoints/May/lm3d_radnerf_torso/config.yaml']
| Hparams:
accumulate_grad_batches: 1, ambient_out_dim: 2, amp: True, base_config: ['egs/egs_bases/radnerf/lm3d_radnerf.yaml'], binary_data_dir: data/binary/videos,
bound: 1, camera_offset: [0, 0, 0], camera_scale: 4.0, clip_grad_norm: 0, clip_grad_value: 0,
cond_out_dim: 64, cond_type: idexp_lm3d_normalized, cond_win_size: 1, cuda_ray: True, debug: False,
density_thresh: 10, density_thresh_torso: 0.01, desired_resolution: 2048, dt_gamma: 0.00390625, eval_max_batches: 100,
exp_name: , far: 0.9, finetune_lips: True, finetune_lips_start_iter: 200000, geo_feat_dim: 128,
grid_interpolation_type: linear, grid_size: 128, grid_type: tiledgrid, gui_fovy: 21.24, gui_h: 512,
gui_max_spp: 1, gui_radius: 3.35, gui_w: 512, head_model_dir: checkpoints/May/lm3d_radnerf, hidden_dim_ambient: 128,
hidden_dim_color: 128, hidden_dim_sigma: 128, individual_embedding_dim: 4, individual_embedding_num: 13000, infer: True,
infer_audio_source_name: data/raw/val_wavs/zozo.wav, infer_bg_img_fname: , infer_c2w_name: , infer_cond_name: infer_out/May/pred_lm3d/zozo.npy, infer_lm3d_clamp_std: 2.5,
infer_lm3d_lle_percent: 0.0, infer_lm3d_smooth_sigma: 0.0, infer_out_video_name: infer_out/May/pred_video/zozo_radnerf_torso_smo.mp4, infer_scale_factor: 1.0, infer_smo_std: 0.0,
infer_smooth_camera_path: True, infer_smooth_camera_path_kernel_size: 7, lambda_ambient: 0.1, lambda_lpips_loss: 0.01, lambda_weights_entropy: 0.0001,
load_ckpt: , load_imgs_to_memory: False, log2_hashmap_size: 16, lr: 0.0005, max_ray_batch: 4096,
max_steps: 16, max_updates: 250000, min_near: 0.05, n_rays: 65536, near: 0.3,
num_ckpt_keep: 1, num_layers_ambient: 3, num_layers_color: 2, num_layers_sigma: 3, num_sanity_val_steps: 2,
num_steps: 16, num_valid_plots: 5, optimizer_adam_beta1: 0.9, optimizer_adam_beta2: 0.999, print_nan_grads: False,
processed_data_dir: data/processed/videos, raw_data_dir: data/raw/videos, resume_from_checkpoint: 0, save_best: True, save_codes: ['tasks', 'modules', 'egs'],
save_gt: True, scheduler: exponential, seed: 9999, smo_win_size: 5, smooth_lips: False,
task_cls: tasks.radnerfs.radnerf_torso.RADNeRFTorsoTask, tb_log_interval: 100, torso_head_aware: False, torso_individual_embedding_dim: 8, torso_shrink: 0.8,
torso_train_mode: 1, update_extra_interval: 16, upsample_steps: 0, use_window_cond: True, val_check_interval: 2000,
valid_infer_interval: 10000, valid_monitor_key: val_loss, valid_monitor_mode: min, validate: False, video_id: May,
warmup_updates: 0, weight_decay: 0, with_att: True, work_dir: checkpoints/May/lm3d_radnerf_torso,
08/30 05:50:22 PM GPU available: False, GPU used: [1]
trainval: Smooth head trajectory (rotation and translation) with a window size of 7
Traceback (most recent call last):
File "/root/autodl-tmp/GeneFace-1.1.0/inference/nerfs/lm3d_radnerf_infer.py", line 99, in
User provided device_type of 'cuda', but CUDA is not available. Disabling 貌似你的 CUDA没安装好
我用的 CUDA11.7 + python3.9 + pytorch1.13.1
How fast is inference on a 4090?
用4090生成速度怎么样?
推理还挺快的，比方说这个 6296帧480x480的视频，时长 00:04:11.84，耗时 03:35，可以做到实时的。
NeRF is rendering frames...: 100%|█████████████████████████████████████████████████████████████████████████████████| 6296/6296 [03:35<00:00, 29.21it/s]
How fast is inference on a 4090? 用4090生成速度怎么样?
推理还挺快的，比方说这个 6296帧480x480的视频，时长 00:04:11.84，耗时 03:35，可以做到实时的。
NeRF is rendering frames...: 100%|█████████████████████████████████████████████████████████████████████████████████| 6296/6296 [03:35<00:00, 29.21it/s]
朋友，你有“~/.cache/torch_extensions/py39_cu118/_grid_encoder/_grid_encoder.so”这个文件吗，我这里一直因为没有这个文件走不完demo最后一步，不知道怎么解决，也是4090
