train.py

import os
from tqdm.auto import tqdm
from opt import config_parser
import json, random
from renderer import *
from utils import *
from torch.utils.tensorboard import SummaryWriter
import datetime
from dataLoader import dataset_dict
import sys
from time import time
device = torch.device("cuda")

import logging
logging.getLogger("imageio_ffmpeg").setLevel(logging.ERROR)

# Hyper-Parameters
upsilon = 0.4
eta = 0
test_slimmed = [2, 4, 6, 8, 12, 16]

renderer = OctreeRender_trilinear_fast
train_log=open('train.log','a')
train_log.write(str(datetime.datetime.now())+'\n')
class SimpleSampler:
    def __init__(self, total, batch):
        self.total = total
        self.batch = batch
        self.curr = total
        self.ids = None
    def nextids(self):
        self.curr+=self.batch
        if self.curr + self.batch > self.total:
            self.ids = torch.LongTensor(np.random.permutation(self.total))
            self.curr = 0
        return self.ids[self.curr:self.curr+self.batch]
@torch.no_grad()
def export_mesh(args):
    ckpt = torch.load(args.ckpt, map_location=device)
    kwargs = ckpt['kwargs']
    kwargs.update({'device': device})
    kwargs['init_rank'] = args.init_rank
    kwargs['rank_inc'] = args.rank_inc
    tensorf = eval(args.model_name)(**kwargs)
    tensorf.load(ckpt)
    alpha,_ = tensorf.getDenseAlpha()
    convert_sdf_samples_to_ply(alpha.cpu(), f'{args.ckpt[:-3]}.ply',bbox=tensorf.aabb.cpu(), level=0.005)
    raise Exception('Mesh Exported!')
@torch.no_grad()
def render_test(args):
    dataset = dataset_dict[args.dataset_name]
    test_dataset = dataset(args.datadir, split='test', downsample=args.downsample_train, is_stack=True)
    white_bg = test_dataset.white_bg
    ndc_ray = args.ndc_ray
    if not os.path.exists(args.ckpt):
        print('the ckpt path does not exist')
        return
    ckpt = torch.load(args.ckpt, map_location=device)
    kwargs = ckpt['kwargs']
    kwargs.update({'device': device})
    kwargs['init_rank'] = args.init_rank
    kwargs['rank_inc'] = args.rank_inc
    tensorf = eval(args.model_name)(**kwargs)
    tensorf.load(ckpt)
    logfolder = os.path.dirname(args.ckpt)
    if args.render_train:
        os.makedirs(f'{logfolder}/imgs_train_all', exist_ok=True)
        train_dataset = dataset(args.datadir, split='train', downsample=args.downsample_train, is_stack=True)
        PSNRs_test = evaluation(train_dataset,tensorf, args, renderer, f'{logfolder}/imgs_train_all/',
                                N_vis=-1, N_samples=-1, white_bg = white_bg, ndc_ray=ndc_ray,device=device)
        print(f'======> {args.expname} train all psnr: {np.mean(PSNRs_test)} <========================')
    if args.render_test:
        os.makedirs(f'{logfolder}/{args.expname}/imgs_test_all', exist_ok=True)
        evaluation(test_dataset,tensorf, args, renderer, f'{logfolder}/{args.expname}/imgs_test_all/',
                                N_vis=-1, N_samples=-1, white_bg = white_bg, ndc_ray=ndc_ray,device=device)
    if args.render_path:
        c2ws = test_dataset.render_path
        os.makedirs(f'{logfolder}/{args.expname}/imgs_path_all', exist_ok=True)
        evaluation_path(test_dataset,tensorf, c2ws, renderer, f'{logfolder}/{args.expname}/imgs_path_all/',
                                N_vis=-1, N_samples=-1, white_bg = white_bg, ndc_ray=ndc_ray,device=device)
def reconstruction(args):
    dataset = dataset_dict[args.dataset_name]
    train_dataset = dataset(args.datadir, split='train', downsample=args.downsample_train, is_stack=False)
    test_dataset = dataset(args.datadir, split='test', downsample=args.downsample_train, is_stack=True)
    white_bg = train_dataset.white_bg
    near_far = train_dataset.near_far
    ndc_ray = args.ndc_ray
    upsamp_list = args.upsamp_list
    update_AlphaMask_list = args.update_AlphaMask_list
    n_lamb_sigma = args.n_lamb_sigma
    n_lamb_sh = args.n_lamb_sh
    if args.add_timestamp:
        logfolder = f'{args.basedir}/{args.expname}{datetime.datetime.now().strftime("-%Y%m%d-%H%M%S")}'
    else:
        logfolder = f'{args.basedir}/{args.expname}'
    os.makedirs(logfolder, exist_ok=True)
    os.makedirs(f'{logfolder}/imgs_vis', exist_ok=True)
    os.makedirs(f'{logfolder}/imgs_rgba', exist_ok=True)
    os.makedirs(f'{logfolder}/rgba', exist_ok=True)
    summary_writer = SummaryWriter(logfolder)
    aabb = train_dataset.scene_bbox.to(device)
    reso_cur = N_to_reso(args.N_voxel_init, aabb)
    nSamples = min(args.nSamples, cal_n_samples(reso_cur,args.step_ratio))
    if args.ckpt is not None:
        ckpt = torch.load(args.ckpt, map_location=device)
        kwargs = ckpt['kwargs']
        kwargs.update({'device':device})
        tensorf = eval(args.model_name)(init_rank=args.init_rank, rank_inc=args.init_rank, **kwargs)
        tensorf.load(ckpt)
    else:
        tensorf = eval(args.model_name)(aabb, reso_cur, device, init_rank, rank_inc,
                    density_n_comp=n_lamb_sigma, appearance_n_comp=n_lamb_sh, app_dim=args.data_dim_color, near_far=near_far,
                    shadingMode=args.shadingMode, alphaMask_thres=args.alpha_mask_thre, density_shift=args.density_shift, distance_scale=args.distance_scale,
                    pos_pe=args.pos_pe, view_pe=args.view_pe, fea_pe=args.fea_pe, featureC=args.featureC, step_ratio=args.step_ratio, fea2denseAct=args.fea2denseAct)
    grad_vars = tensorf.get_optparam_groups(args.lr_init, args.lr_basis)
    if args.lr_decay_iters > 0:
        lr_factor = args.lr_decay_target_ratio**(1/args.lr_decay_iters)
    else:
        args.lr_decay_iters = args.n_iters
        lr_factor = args.lr_decay_target_ratio**(1/args.n_iters)
    print("lr decay", args.lr_decay_target_ratio, args.lr_decay_iters)
    optimizer = torch.optim.Adam(grad_vars, betas=(0.9,0.99))
    N_voxel_list = (torch.round(torch.exp(torch.linspace(np.log(args.N_voxel_init), np.log(args.N_voxel_final), len(upsamp_list)+1))).long()).tolist()[1:]
    torch.cuda.empty_cache()
    PSNRs,PSNRs_test = [],[0]
    allrays, allrgbs = train_dataset.all_rays, train_dataset.all_rgbs
    if not args.ndc_ray:
        allrays, allrgbs = tensorf.filtering_rays(allrays, allrgbs, bbox_only=True)
    trainingSampler = SimpleSampler(allrays.shape[0], args.batch_size)
    Ortho_reg_weight = args.Ortho_weight
    print("initial Ortho_reg_weight", Ortho_reg_weight)
    L1_reg_weight = args.L1_weight_inital
    print("initial L1_reg_weight", L1_reg_weight)
    TV_weight_density, TV_weight_app = args.TV_weight_density, args.TV_weight_app
    tvreg = TVLoss()
    print(f"initial TV_weight density: {TV_weight_density} appearance: {TV_weight_app}")
    loss_rec=[float('inf')]
    pbar = tqdm(range(args.n_iters), miniters=args.progress_refresh_rate, file=sys.stdout)
    psnr_rec = -1
    last_rank_freeze = 0
    for iteration in pbar:
        ray_idx = trainingSampler.nextids()
        rays_train, rgb_train = allrays[ray_idx], allrgbs[ray_idx].to(device)
        rgb_map, alphas_map, depth_map, weights, uncertainty = renderer(rays_train, tensorf, chunk=args.batch_size,
                                N_samples=nSamples, white_bg = white_bg, ndc_ray=ndc_ray, device=device, is_train=True)
        loss = torch.mean((rgb_map - rgb_train) ** 2)
        loss_rec.append(loss)
        total_loss = loss
        if Ortho_reg_weight > 0:
            loss_reg = tensorf.vector_comp_diffs()
            total_loss += Ortho_reg_weight*loss_reg
            summary_writer.add_scalar('train/reg', loss_reg.detach().item(), global_step=iteration)
        if L1_reg_weight > 0:
            loss_reg_L1 = tensorf.density_L1()
            total_loss += L1_reg_weight*loss_reg_L1
            summary_writer.add_scalar('train/reg_l1', loss_reg_L1.detach().item(), global_step=iteration)
        if TV_weight_density>0:
            TV_weight_density *= lr_factor
            loss_tv = tensorf.TV_loss_density(tvreg) * TV_weight_density
            total_loss = total_loss + loss_tv
            summary_writer.add_scalar('train/reg_tv_density', loss_tv.detach().item(), global_step=iteration)
        if TV_weight_app>0:
            TV_weight_app *= lr_factor
            loss_tv = tensorf.TV_loss_app(tvreg)*TV_weight_app
            total_loss = total_loss + loss_tv
            summary_writer.add_scalar('train/reg_tv_app', loss_tv.detach().item(), global_step=iteration)
        optimizer.zero_grad()
        total_loss.backward()
        optimizer.step()
        loss = loss.detach().item()
        PSNRs.append(-10.0 * np.log(loss) / np.log(10.0))
        summary_writer.add_scalar('train/PSNR', PSNRs[-1], global_step=iteration)
        summary_writer.add_scalar('train/mse', loss, global_step=iteration)
        for param_group in optimizer.param_groups:
            param_group['lr'] = param_group['lr'] * lr_factor
        if iteration % args.progress_refresh_rate == 0:
            pbar.set_description(
                f'Iteration {iteration:05d}:'
                + f' train_psnr = {float(np.mean(PSNRs)):.2f}'
                + f' test_psnr = {float(np.mean(PSNRs_test)):.2f}'
                + f' mse = {loss:.6f}'
            )
            PSNRs = []
        if iteration % args.vis_every == args.vis_every - 1 and args.N_vis!=0:
            PSNRs_test = evaluation(test_dataset,tensorf, args, renderer, f'{logfolder}/imgs_vis/', N_vis=args.N_vis,
                                    prtx=f'{iteration:06d}_', N_samples=nSamples, white_bg = white_bg, ndc_ray=ndc_ray, compute_extra_metrics=False)
            temp = np.mean(PSNRs_test)
            summary_writer.add_scalar('test/psnr', np.mean(PSNRs_test), global_step=iteration)
            if temp<psnr_rec:
                print('early stop')
                break
            psnr_rec = temp
        if iteration in update_AlphaMask_list:
            if reso_cur[0] * reso_cur[1] * reso_cur[2]<256**3:
                reso_mask = reso_cur
            new_aabb = tensorf.updateAlphaMask(tuple(reso_mask))
            if iteration == update_AlphaMask_list[0]:
                tensorf.shrink(new_aabb)
                L1_reg_weight = args.L1_weight_rest
                print("continuing L1_reg_weight", L1_reg_weight)
            if not args.ndc_ray and iteration == update_AlphaMask_list[1]:
                allrays,allrgbs = tensorf.filtering_rays(allrays,allrgbs)
                trainingSampler = SimpleSampler(allrgbs.shape[0], args.batch_size)
        if iteration in upsamp_list:
            n_voxels = N_voxel_list.pop(0)
            reso_cur = N_to_reso(n_voxels, tensorf.aabb)
            nSamples = min(args.nSamples, cal_n_samples(reso_cur,args.step_ratio))
            tensorf.upsample_volume_grid(reso_cur,device)
            if args.lr_upsample_reset:
                print("reset lr to initial")
                lr_scale = 1
            else:
                lr_scale = args.lr_decay_target_ratio ** (iteration / args.n_iters)
            grad_vars = tensorf.get_optparam_groups(args.lr_init*lr_scale, args.lr_basis*lr_scale)
            optimizer = torch.optim.Adam(grad_vars, betas=(0.9, 0.99))
        if tensorf.current_rank[0]<n_lamb_sigma[0] and tensorf.current_rank[1]<n_lamb_sh[0] and abs(loss_rec[-2]-loss_rec[-1])/loss_rec[-1]>upsilon and iteration-last_rank_freeze>eta:
            print(iteration,tensorf.current_rank)
            tensorf.update_current_rank(device)
            last_rank_freeze = iteration
            print(iteration,tensorf.current_rank)
    tensorf.save(f'{logfolder}/{args.expname}{tensorf.current_rank}.th')
    if args.render_train:
        os.makedirs(f'{logfolder}/imgs_train_all', exist_ok=True)
        train_dataset = dataset(args.datadir, split='train', downsample=args.downsample_train, is_stack=True)
        PSNRs_test = evaluation(train_dataset,tensorf, args, renderer, f'{logfolder}/imgs_train_all/',
                                N_vis=-1, N_samples=-1, white_bg = white_bg, ndc_ray=ndc_ray,device=device)
        print(f'======> {args.expname} test all psnr: {np.mean(PSNRs_test)} <========================')
    if args.render_test:
        os.makedirs(f'{logfolder}/imgs_test_all', exist_ok=True)
        PSNRs_test = evaluation(test_dataset,tensorf, args, renderer, f'{logfolder}/imgs_test_all/',
                                N_vis=-1, N_samples=-1, white_bg = white_bg, ndc_ray=ndc_ray,device=device)
        summary_writer.add_scalar('test/psnr_all', np.mean(PSNRs_test), global_step=iteration)
        print(f'======> {args.expname} test all psnr: {np.mean(PSNRs_test)} <========================')
        while tensorf.current_rank[0]>0:
            tensorf.downdate_current_rank(device, 1)
            if tensorf.current_rank[0] in test_slimmed:
                tensorf.save(f'{logfolder}/{args.expname}{tensorf.current_rank}.th')
                print('='*20)
                print(tensorf.current_rank)
                os.makedirs(f'{logfolder}/imgs_test_all', exist_ok=True)
                PSNRs_test = evaluation(test_dataset,tensorf, args, renderer, f'{logfolder}/imgs_test_all/',
                                    N_vis=-1, N_samples=-1, white_bg = white_bg, ndc_ray=ndc_ray,device=device)
                summary_writer.add_scalar('test/psnr_all', np.mean(PSNRs_test), global_step=iteration)
                print(f'======> {args.expname} test all psnr: {np.mean(PSNRs_test)} <========================')
    if args.render_path:
        c2ws = test_dataset.render_path
        print('========>',c2ws.shape)
        os.makedirs(f'{logfolder}/imgs_path_all', exist_ok=True)
        evaluation_path(test_dataset,tensorf, c2ws, renderer, f'{logfolder}/imgs_path_all/',
                                N_vis=-1, N_samples=-1, white_bg = white_bg, ndc_ray=ndc_ray,device=device)
if __name__ == '__main__':
    args = config_parser()
    print(args)
    init_rank=args.init_rank
    rank_inc=args.rank_inc
    if args.export_mesh:
        export_mesh(args)
    if args.render_only and (args.render_test or args.render_path):
        render_test(args)
    else:
        reconstruction(args)