-
Notifications
You must be signed in to change notification settings - Fork 5
/
Copy pathtrain.py
344 lines (290 loc) · 16.6 KB
/
train.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
#
# Copyright (C) 2023, Inria
# GRAPHDECO research group, https://team.inria.fr/graphdeco
# All rights reserved.
#
# This software is free for non-commercial, research and evaluation use
# under the terms of the LICENSE.md file.
#
# For inquiries contact [email protected]
#
import os
import numpy as np
import open3d as o3d
import cv2
import torch
import torchvision
import random
import imageio
from random import randint
from utils.loss_utils import l1_loss, ssim
from gaussian_renderer import render, network_gui
import sys
from scene import Scene, GaussianModel
from utils.general_utils import safe_state
import uuid
from tqdm import tqdm
from utils.image_utils import psnr
from argparse import ArgumentParser, Namespace
from arguments import ModelParams, PipelineParams, OptimizationParams
from utils.vis_utils import get_grayscale_image
from utils.system_utils import load_config
from omegaconf import OmegaConf
try:
from torch.utils.tensorboard import SummaryWriter
TENSORBOARD_FOUND = True
except ImportError:
TENSORBOARD_FOUND = False
def training(config, testing_iterations, saving_iterations, checkpoint_iterations, checkpoint, debug_from):
first_iter = 0
dataset = config.dataset
opt = config.optimizer
gs_model = config.gs_model
pipe = config.pipeline
tb_writer = prepare_output_and_logger(config)
gaussians = GaussianModel(gs_model)
scene = Scene(config, gaussians)
gaussians.training_setup(opt)
if checkpoint:
(model_params, first_iter) = torch.load(checkpoint)
gaussians.restore(model_params, opt)
bg_color = [1, 1, 1] if dataset.white_background else [0, 0, 0]
background = torch.tensor(bg_color, dtype=torch.float32, device="cuda")
iter_start = torch.cuda.Event(enable_timing = True)
iter_end = torch.cuda.Event(enable_timing = True)
trainCameras = scene.getTrainCameras().copy()
testCameras = scene.getTestCameras().copy()
for idx, camera in enumerate(scene.getTrainCameras() + scene.getTestCameras()):
camera.idx = idx
# highresolution index
highresolution_index = []
for index, camera in enumerate(trainCameras):
if camera.image_width >= 800:
highresolution_index.append(index)
viewpoint_stack = None
ema_loss_for_log = 0.0
ema_dist_for_log = 0.0
ema_normal_for_log = 0.0
progress_bar = tqdm(range(first_iter, opt.iterations), desc="Training progress")
first_iter += 1
reset_opacity_densify_delay = 0
for iteration in range(first_iter, opt.iterations + 1):
iter_start.record()
gaussians.update_learning_rate(iteration)
# Every 1000 its we increase the levels of SH up to a maximum degree
if iteration % 1000 == 0:
gaussians.oneupSHdegree()
# Pick a random Camera
if not viewpoint_stack:
viewpoint_stack = scene.getTrainCameras().copy()
viewpoint_cam = viewpoint_stack.pop(randint(0, len(viewpoint_stack)-1))
# Render
if (iteration - 1) == debug_from:
pipe.debug = True
gt_image = viewpoint_cam.get_image.cuda()
render_pkg = render(viewpoint_cam, gaussians, pipe, opt, background, kernel_size=dataset.kernel_size, stop_z_gradient = False)
rendering, viewspace_point_tensor, visibility_filter, radii = render_pkg["render"], render_pkg["viewspace_points"], render_pkg["visibility_filter"], render_pkg["radii"]
image = rendering[:3, :, :]
lambda_normal = opt.lambda_normal if iteration > opt.normal_from_iter else 0.0
lambda_dist = opt.lambda_dist if iteration > opt.dist_from_iter else 0.0
lambda_curv_dist = opt.lambda_curv_dist if iteration > opt.curv_from_iter else 0.0
lambda_curv_flat = opt.lambda_curv_flat if iteration > opt.curv_from_iter else 0.0
# Depth and curvature distortion regularization. The curvature distortion and flatten loss performs poorly.
rend_dist = render_pkg["render_dist"]
render_curvature_dist = render_pkg["render_curvature_dist"]
dist_loss = lambda_dist * (rend_dist).mean()
# curvature_dist_loss = lambda_curv_dist * (render_curvature_dist).mean()
curvature_dist_loss = torch.tensor(0, device = image.device)
rend_normal = render_pkg['render_normal']
surf_normal = render_pkg['surf_normal']
num_curvature = torch.abs(torch.nan_to_num(render_pkg["render_curvature"],0.00001))
rend_curvature_log = torch.clamp_max(torch.log(torch.clamp_min(num_curvature, 0.00001)), opt.curvature_clamp_threshold)
curv_mask = 1 - torch.sigmoid(rend_curvature_log)
curv_flat_mask = num_curvature < 2.0
# curv_flat_loss = lambda_curv_flat * render_pkg["render_curvature"][curv_flat_mask].mean()
curv_flat_loss = torch.tensor(0, device = image.device)
alpha_map = viewpoint_cam.get_mask
normal_error = (alpha_map * curv_mask.detach() * (1 - (rend_normal * surf_normal))).sum(dim=0)[None]
normal_loss = lambda_normal * (normal_error).mean()
alpha_loss = torch.tensor(0, device = image.device)
if dataset.use_alpha:
alpha_loss = 0.05 * torch.abs(alpha_map.permute(1,2,0) - render_pkg["render_alpha"].permute(1,2,0)).mean()
Ll1 = l1_loss(image, gt_image)
rgb_loss = (1.0 - opt.lambda_dssim) * Ll1 + opt.lambda_dssim * (1.0 - ssim(image, gt_image))
loss = rgb_loss + dist_loss + normal_loss + curvature_dist_loss + curv_flat_loss + alpha_loss
loss.backward()
iter_end.record()
radii = render_pkg["radii"]
# Output the images from the training process.
with torch.no_grad():
image_write = torch.clamp(render_pkg["render"], 0.0, 1.0).permute(1,2,0)
depth_write = render_pkg["render_depth"].permute(1,2,0).squeeze()
alpha_write = render_pkg["render_alpha"].permute(1,2,0).squeeze()
surf_normal_write = render_pkg["surf_normal"].permute(1,2,0)
surf_normal_write = surf_normal_write * 0.5 + 0.5
render_normal_write = render_pkg["render_normal"].permute(1,2,0)
render_normal_write = render_normal_write * 0.5 + 0.5
render_curvature_write = render_pkg["render_curvature"].permute(1,2,0)
render_curvature_log_write = torch.log(render_curvature_write + render_curvature_write.min() + 1e-7)
image_write = (image_write.detach().cpu().numpy()*255).astype(np.uint8)
depth_write = get_grayscale_image(depth_write, data_range=None,cmap='jet')
alpha_write = get_grayscale_image(alpha_write, data_range=[0,1],cmap='jet')
render_normal_write = (render_normal_write.detach().cpu().numpy()*255).astype(np.uint8)
surf_normal_write = (surf_normal_write.detach().cpu().numpy()*255).astype(np.uint8)
render_curvature_write = get_grayscale_image(render_curvature_write, data_range=None,cmap='jet')
render_curvature_log_write = get_grayscale_image(render_curvature_log_write, data_range=None,cmap='jet')
output_list = {
"rgb": image_write,
"depth": depth_write,
"alpha_map":alpha_write,
"render_normal": render_normal_write,
"surf_normal": surf_normal_write,
"curvature": render_curvature_write,
"curvature_log": render_curvature_log_write,
}
if (iteration) % 50 == 0:
prefix = f"grad_threshold={opt.densify_grad_threshold}_lambda_dist={opt.lambda_dist}"
for case in output_list:
output_dir = os.path.join(dataset.model_path, prefix, case)
os.makedirs(output_dir,exist_ok=True)
imageio.imwrite(os.path.join(output_dir,f"{iteration}_{case}.jpg"), output_list[case])
with torch.no_grad():
# Progress bar
ema_loss_for_log = 0.4 * rgb_loss.item() + 0.6 * ema_loss_for_log
ema_curv_dist_for_log = 0.4 * curvature_dist_loss.item() + 0.6 * ema_dist_for_log
ema_dist_for_log = 0.4 * dist_loss.item() + 0.6 * ema_dist_for_log
ema_normal_for_log = 0.4 * normal_loss.item() + 0.6 * ema_normal_for_log
if iteration % 10 == 0:
progress_bar.set_postfix({"Loss": f"{ema_loss_for_log:.{5}f}",
"distort": f"{ema_dist_for_log:.{5}f}",
# "distcurv": f"{ema_curv_dist_for_log:.{5}f}",
# "normal": f"{ema_normal_for_log:.{5}f}",
"radiiMax":f"{radii.max().item():.{5}f}",
"Points":f"{len(gaussians.get_xyz)}",
# "s":f"{s.item():.{5}f}",
})
progress_bar.update(10)
if iteration == opt.iterations:
progress_bar.close()
# Log and save
training_report(tb_writer, iteration, Ll1, loss, l1_loss, iter_start.elapsed_time(iter_end), testing_iterations, scene, render, (pipe, opt, background, dataset.kernel_size), dataset)
if (iteration in saving_iterations):
print("\n[ITER {}] Saving Gaussians".format(iteration))
scene.save(iteration)
# Densification
if iteration < opt.densify_until_iter:
# Keep track of max radii in image-space for pruning
gaussians.max_radii2D[visibility_filter] = torch.max(gaussians.max_radii2D[visibility_filter], radii[visibility_filter])
gaussians.add_densification_stats(viewspace_point_tensor, radii, visibility_filter)
if iteration > opt.densify_from_iter and iteration % opt.densification_interval == 0:
if reset_opacity_densify_delay <= 0:
size_threshold = 20 if iteration > opt.opacity_reset_interval else None
gaussians.densify_and_prune(opt.densify_grad_threshold, 0.05, scene.cameras_extent, size_threshold)
else:
reset_opacity_densify_delay -= 1
if iteration % opt.opacity_reset_interval == 0 or (dataset.white_background and iteration == opt.densify_from_iter):
gaussians.reset_opacity()
# For scenes with thousands of images, after each reset of opacity,
# the Gaussians in the scene need to be sufficiently optimized before pruning.
reset_opacity_densify_delay = opt.reset_opacity_densify_delay
# Optimizer step
if iteration < opt.iterations:
gaussians.optimizer.step()
gaussians.optimizer.zero_grad(set_to_none = True)
if (iteration in checkpoint_iterations):
print("\n[ITER {}] Saving Checkpoint".format(iteration))
torch.save((gaussians.capture(), iteration), scene.model_path + "/chkpnt" + str(iteration) + ".pth")
def prepare_output_and_logger(args):
if not args.model_path:
if os.getenv('OAR_JOB_ID'):
unique_str=os.getenv('OAR_JOB_ID')
else:
unique_str = str(uuid.uuid4())
args.model_path = os.path.join("./exp/", unique_str[0:10])
# Set up output folder
print("Output folder: {}".format(args.model_path))
os.makedirs(args.model_path, exist_ok = True)
OmegaConf.save(args,os.path.join(args.model_path, 'config.yaml'))
# Create Tensorboard writer
tb_writer = None
if TENSORBOARD_FOUND:
tb_writer = SummaryWriter(args.model_path)
else:
print("Tensorboard not available: not logging progress")
return tb_writer
def training_report(tb_writer, iteration, Ll1, loss, l1_loss, elapsed, testing_iterations, scene : Scene, renderFunc, renderArgs, dataset):
if tb_writer:
tb_writer.add_scalar('train_loss_patches/l1_loss', Ll1.item(), iteration)
tb_writer.add_scalar('train_loss_patches/total_loss', loss.item(), iteration)
tb_writer.add_scalar('iter_time', elapsed, iteration)
tb_writer.add_scalar('total_points', scene.gaussians.get_xyz.shape[0], iteration)
# Report test and samples of training set
if iteration in testing_iterations:
torch.cuda.empty_cache()
validation_configs = ({'name': 'test', 'cameras' : scene.getTestCameras()},
{'name': 'train', 'cameras' : [scene.getTrainCameras()[idx % len(scene.getTrainCameras())] for idx in range(5, 30, 5)]})
for config in validation_configs:
if config['cameras'] and len(config['cameras']) > 0:
l1_test = 0.0
psnr_test = 0.0
for idx, viewpoint in enumerate(config['cameras']):
rendering = renderFunc(viewpoint, scene.gaussians, *renderArgs)["render"]
image = rendering[:3, :, :]
image = torch.clamp(image, 0.0, 1.0)
gt_image = torch.clamp(viewpoint.get_image.to("cuda"), 0.0, 1.0)
if tb_writer and (idx < 5):
tb_writer.add_images(config['name'] + "_view_{}/render".format(viewpoint.image_name), image[None], global_step=iteration)
if iteration == testing_iterations[0]:
tb_writer.add_images(config['name'] + "_view_{}/ground_truth".format(viewpoint.image_name), gt_image[None], global_step=iteration)
l1_test += l1_loss(image, gt_image).mean().double()
psnr_test += psnr(image, gt_image).mean().double()
output_dir = os.path.join(dataset.model_path, "test")
output_dir = os.path.join(output_dir,f"{iteration}")
os.makedirs(output_dir,exist_ok=True)
imageio.imwrite(os.path.join(output_dir,f"{config['name']}_{idx}_rgb.jpg"),(image.permute(1,2,0).cpu().numpy()*255).astype(np.uint8))
psnr_test /= len(config['cameras'])
l1_test /= len(config['cameras'])
print("\n[ITER {}] Evaluating {}: L1 {} PSNR {}".format(iteration, config['name'], l1_test, psnr_test))
if tb_writer:
tb_writer.add_scalar(config['name'] + '/loss_viewpoint - l1_loss', l1_test, iteration)
tb_writer.add_scalar(config['name'] + '/loss_viewpoint - psnr', psnr_test, iteration)
if tb_writer:
tb_writer.add_histogram("scene/opacity_histogram", scene.gaussians.get_opacity, iteration)
torch.cuda.empty_cache()
if __name__ == "__main__":
# Set up command line argument parser
parser = ArgumentParser(description="Training script parameters")
lp = ModelParams(parser)
op = OptimizationParams(parser)
pp = PipelineParams(parser)
parser.add_argument('--ip', type=str, default="127.0.0.1")
parser.add_argument('--port', type=int, default=6009)
parser.add_argument('--debug_from', type=int, default=-1)
parser.add_argument('--detect_anomaly', action='store_true', default=False)
parser.add_argument("--test_iterations", nargs="+", type=int, default= [i * 500 for i in range(0, 60)])
parser.add_argument("--save_iterations", nargs="+", type=int, default=[7_000, 15_000, 30_000, 45_000, 60_000])
parser.add_argument("--quiet", action="store_true")
parser.add_argument("--checkpoint_iterations", nargs="+", type=int, default=[])
parser.add_argument("--start_checkpoint", type=str, default = None)
parser.add_argument('--conf_path',default='./config/base.yaml')
args, extras = parser.parse_known_args()
args.save_iterations.append(args.iterations)
config_base = OmegaConf.load("./config/base.yaml")
config_case = OmegaConf.load(args.conf_path)
config = OmegaConf.merge(config_base, config_case)
config.model_path = config.model_path.replace(" ", "")
config.model_path = config.model_path.replace("\n", "")
# config = load_config(args.conf_path, cli_args=extras)
print("Optimizing " + args.model_path)
# Initialize system state (RNG)
# safe_state(args.quiet)
random.seed(0)
np.random.seed(0)
torch.manual_seed(0)
torch.cuda.set_device(torch.device("cuda:0"))
# # Start GUI server, configure and run training
# network_gui.init(args.ip, args.port)
torch.autograd.set_detect_anomaly(args.detect_anomaly)
training(config, args.test_iterations, args.save_iterations, args.checkpoint_iterations, args.start_checkpoint, args.debug_from)
# All done
print("\nTraining complete.")