wanmeihuali / taichi_3d_gaussian_splatting

An unofficial implementation of paper 3D Gaussian Splatting for Real-Time Radiance Field Rendering by taichi lang.

Apache License 2.0

648 stars 61 forks source link

Update test_sagemaker.yaml #77

Closed wanmeihuali closed 1 year ago

github-actions[bot] commented 1 year ago

Running experiment on sagemaker with git sha 6af71cb54b73d39063ca053ea69260353e463ebf

github-actions[bot] commented 1 year ago

Training job densify-policy-6-6af71cb-230701-215322-tat-truck created

github-actions[bot] commented 1 year ago

Training job densify-policy-6-6af71cb-230701-215322-tat-truck completed.

Model url: s3://taichi-3d-gaussian-splatting-log/tat-truck/densify-policy-6-6af71cb-230701-215322-tat-truck/output/model.tar.gz,

tensorboard output path: s3://taichi-3d-gaussian-splatting-log/tat-truck/densify-policy-6-6af71cb-230701-215322-tat-truck/output/output.tar.gz

github-actions[bot] commented 1 year ago

Training job densify-policy-6-6af71cb-230701-215322-tat-truck final metrics:

Latest Metrics

val:psnr	val:ssim	train:num_valid_points	train:ssimloss	train:loss	train:l1loss	train:psnr	train:iteration	val:loss	train:ssim
22.929962158203125	0.7806919813156128	73078.0	0.18999195098876953	0.06536456942558289	0.03420772776007652	23.726150512695312	30000.0	0.07743681222200394	0.8100080490112305

Max Metrics

val:psnr	val:ssim	val:7kpsnr	train:7kpsnr	val:7kssim	val:5kpsnr	train:5kssim	train:psnr	train:5kpsnr	train:ssim	val:5kssim	train:7kssim
22.929962158203125	0.7806919813156128	22.370588302612305	22.95978355407715	0.7676122188568115	21.68787956237793	0.7536598443984985	23.726150512695312	22.24698257446289	0.8100080490112305	0.7457346320152283	0.7352254390716553

github-actions[bot] commented 1 year ago

Running experiment on sagemaker with git sha be24f78f142fb2447cdedc7541e9168895c0c411

github-actions[bot] commented 1 year ago

Training job densify-policy-6-be24f78-230701-233556-tat-truck created

github-actions[bot] commented 1 year ago

Training job densify-policy-6-be24f78-230701-233556-tat-truck completed.

Model url: s3://taichi-3d-gaussian-splatting-log/tat-truck/densify-policy-6-be24f78-230701-233556-tat-truck/output/model.tar.gz,

tensorboard output path: s3://taichi-3d-gaussian-splatting-log/tat-truck/densify-policy-6-be24f78-230701-233556-tat-truck/output/output.tar.gz

github-actions[bot] commented 1 year ago

Training job densify-policy-6-be24f78-230701-233556-tat-truck final metrics:

Latest Metrics

val:loss	val:ssim	train:psnr	val:psnr	train:num_valid_points	train:loss	train:iteration	train:ssimloss	train:ssim	train:l1loss
0.0778520256280899	0.7790136933326721	21.38400650024414	22.878053665161133	77158.0	0.08879497647285461	30000.0	0.2625211477279663	0.7374788522720337	0.04536343365907669

Max Metrics

train:5kssim	train:5kpsnr	train:psnr	val:ssim	val:psnr	val:7kssim	train:7kssim	val:7kpsnr	train:7kpsnr	val:5kssim	val:5kpsnr	train:ssim
0.77464759349823	23.1578369140625	21.38400650024414	0.7790136933326721	22.878053665161133	0.7652256488800049	0.7649277448654175	22.248048782348633	22.016603469848633	0.7451958060264587	21.656614303588867	0.7374788522720337

github-actions[bot] commented 1 year ago

Running experiment on sagemaker with git sha 3a19d21393167b94118fa483768b81a0a8cfb7a0

github-actions[bot] commented 1 year ago

Training job densify-policy-6-3a19d21-230702-012545-tat-truck created

github-actions[bot] commented 1 year ago

Training job densify-policy-6-3a19d21-230702-012545-tat-truck completed.

Model url: s3://taichi-3d-gaussian-splatting-log/tat-truck/densify-policy-6-3a19d21-230702-012545-tat-truck/output/model.tar.gz,

tensorboard output path: s3://taichi-3d-gaussian-splatting-log/tat-truck/densify-policy-6-3a19d21-230702-012545-tat-truck/output/output.tar.gz

github-actions[bot] commented 1 year ago

Training job densify-policy-6-3a19d21-230702-012545-tat-truck final metrics:

Latest Metrics

val:psnr	train:ssimloss	train:iteration	val:loss	train:l1loss	train:ssim	train:num_valid_points	train:psnr	train:loss	val:ssim
23.037145614624023	0.17705559730529785	30000.0	0.07469681650400162	0.03217865154147148	0.8229444026947021	142459.0	25.24092674255371	0.061154041439294815	0.7903799414634705

Max Metrics

train:5kssim	train:5kpsnr	val:psnr	val:7kssim	train:7kssim	val:7kpsnr	val:5kpsnr	val:5kssim	train:ssim	train:psnr	train:7kpsnr	val:ssim
0.7914466857910156	23.67293357849121	23.037145614624023	0.7760159373283386	0.8417671918869019	22.416505813598633	21.745046615600586	0.7542557120323181	0.8229444026947021	25.24092674255371	23.507719039916992	0.7903799414634705