train on llff flower scene

[that-liu]

!/bin/bash

mkdir ./runs/ python -m main_nerf /data3/liuyu/nerf_llff_data/flower/\ --mode blender \ --upsample_steps 128 \ --num_rays 1024 \ --max_ray_batch 1024 \ --mode llff \ --downsample_val 8 \ --reg_ramp_start_step 3000 \ --reg_ramp_finish_step 8000 \ --iters 10000 \ --min_near 0.2 \ --spread_loss_strength 1.5e-5 \ --num_train 3 \ --use_frustum_reg \ --patch_weight_start 0.2 \ --patch_weight_finish 0.2 \ --normalise_diffusion_losses \ --patch_regulariser_path models/rgbd-patch-diffusion.pt \ --frustum_regularise_patches \ --apply_geom_reg_to_patches \ --downsample_train 8 \ --patch_reg_start_step 0 \ --workspace ./runs/train/3_poses/flower/ I used this command and the following error occurred in the flower scene of training llff.: ==> Start Training Epoch 1, lr=0.010000 ... 0% 0/201 [00:00<?, ?it/s]Traceback (most recent call last): File "/data1/liuyu/miniconda3/envs/diffusionerf/lib/python3.9/runpy.py", line 197, in _run_module_as_main return _run_code(code, main_globals, None, File "/data1/liuyu/miniconda3/envs/diffusionerf/lib/python3.9/runpy.py", line 87, in _run_code exec(code, run_globals) File "/data1/liuyu/MPI/diffusionerf/main_nerf.py", line 145, in run(opt) File "/data1/liuyu/MPI/diffusionerf/main_nerf.py", line 122, in run trainer.train(train_loader, valid_loader, max_epoch) File "/data1/liuyu/MPI/diffusionerf/nerf/utils.py", line 657, in train self.train_one_epoch(train_loader) File "/data1/liuyu/MPI/diffusionerf/nerf/utils.py", line 776, in train_one_epoch preds, truths, loss, extra_outputs = self.train_step(data) File "/data1/liuyu/MPI/diffusionerf/nerf/utils.py", line 412, in train_step outputs = self.model.render(rays_o, rays_d, staged=False, bg_color=bg_color, perturb=True, force_all_rays=False, vars(self.opt)) File "/data1/liuyu/MPI/diffusionerf/nerf/renderer.py", line 370, in render results = _run(rays_o, rays_d, kwargs) File "/data1/liuyu/MPI/diffusionerf/nerf/renderer.py", line 129, in run z_vals = perturb_z_vals(z_vals) File "/data1/liuyu/MPI/diffusionerf/nerf/rendering_helpers.py", line 31, in perturb_z_vals assert (deltas >= 0).all() AssertionError 0% 0/201 [00:00<?, ?it/s]