search

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
662 stars 62 forks source link

support different learning rate for high order color #96

Closed wanmeihuali closed 1 year ago

github-actions[bot] commented 1 year ago

Running experiment on sagemaker with git sha dee1758b00432e7ad0b757672dc8fd50b6d569f8

github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-dee1758-230713-063115-tat-train-baseline created

github-actions[bot] commented 1 year ago

Running experiment on sagemaker with git sha dee1758b00432e7ad0b757672dc8fd50b6d569f8

github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-dee1758-230713-063154-tat-truck-baseline created

github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-dee1758-230713-063154-tat-truck-baseline completed.

Model url: s3://taichi-3d-gaussian-splatting-log/tat-truck-baseline/diff-lr-high-order-dee1758-230713-063154-tat-truck-baseline/output/model.tar.gz,

tensorboard output path: s3://taichi-3d-gaussian-splatting-log/tat-truck-baseline/diff-lr-high-order-dee1758-230713-063154-tat-truck-baseline/output/output.tar.gz

github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-dee1758-230713-063154-tat-truck-baseline final metrics:

Latest Metrics

train:iteration train:l1loss train:loss train:num_valid_points train:psnr train:ssim train:ssimloss val:loss val:psnr val:ssim
30000.0 0.02999657206237316 0.048126913607120514 446722.0 25.758638381958008 0.8793517351150513 0.12064826488494873 0.053786009550094604 25.057065963745117 0.8642633557319641

Max Metrics

train:5kpsnr train:5kssim train:7kpsnr train:7kssim train:psnr train:ssim val:5kpsnr val:5kssim val:7kpsnr val:7kssim val:psnr val:ssim
23.58283805847168 0.8447558283805847 25.95098876953125 0.863956093788147 25.758638381958008 0.8793517351150513 23.0758113861084 0.8114733099937439 23.748981475830078 0.8333640098571777 25.057065963745117 0.8642633557319641
github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-dee1758-230713-063115-tat-train-baseline completed.

Model url: s3://taichi-3d-gaussian-splatting-log/tat-train-baseline/diff-lr-high-order-dee1758-230713-063115-tat-train-baseline/output/model.tar.gz,

tensorboard output path: s3://taichi-3d-gaussian-splatting-log/tat-train-baseline/diff-lr-high-order-dee1758-230713-063115-tat-train-baseline/output/output.tar.gz

github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-dee1758-230713-063115-tat-train-baseline final metrics:

Latest Metrics

train:iteration train:l1loss train:loss train:num_valid_points train:psnr train:ssim train:ssimloss val:loss val:psnr val:ssim
30000.0 0.03364260122179985 0.058240197598934174 641553.0 25.809932708740234 0.8433694243431091 0.15663057565689087 0.10425259172916412 20.67436408996582 0.7819371223449707

Max Metrics

train:5kpsnr train:5kssim train:7kpsnr train:7kssim train:psnr train:ssim val:5kpsnr val:5kssim val:7kpsnr val:7kssim val:psnr val:ssim
23.583200454711914 0.8213088512420654 23.027008056640625 0.827880859375 25.809932708740234 0.8433694243431091 19.095823287963867 0.7244452238082886 19.643190383911133 0.7477101683616638 20.67436408996582 0.7819371223449707
github-actions[bot] commented 1 year ago

Running experiment on sagemaker with git sha 0d59d71ed0fab1c2aef8c83dbf028e0fe4b6c66f

github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-0d59d71-230713-095434-tat-truck-baseline created

github-actions[bot] commented 1 year ago

Running experiment on sagemaker with git sha 0d59d71ed0fab1c2aef8c83dbf028e0fe4b6c66f

github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-0d59d71-230713-095451-tat-train-baseline created

github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-0d59d71-230713-095434-tat-truck-baseline completed.

Model url: s3://taichi-3d-gaussian-splatting-log/tat-truck-baseline/diff-lr-high-order-0d59d71-230713-095434-tat-truck-baseline/output/model.tar.gz,

tensorboard output path: s3://taichi-3d-gaussian-splatting-log/tat-truck-baseline/diff-lr-high-order-0d59d71-230713-095434-tat-truck-baseline/output/output.tar.gz

github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-0d59d71-230713-095434-tat-truck-baseline final metrics:

Latest Metrics

train:iteration train:l1loss train:loss train:num_valid_points train:psnr train:ssim train:ssimloss val:loss val:psnr val:ssim
30000.0 0.02908407151699066 0.047144077718257904 439696.0 25.56326675415039 0.8806158900260925 0.11938410997390747 0.05415225401520729 25.00387191772461 0.8637361526489258

Max Metrics

train:5kpsnr train:5kssim train:7kpsnr train:7kssim train:psnr train:ssim val:5kpsnr val:5kssim val:7kpsnr val:7kssim val:psnr val:ssim
23.10543441772461 0.794529139995575 22.79907989501953 0.8084046244621277 25.56326675415039 0.8806158900260925 22.63949203491211 0.8051724433898926 23.659250259399414 0.8327409625053406 25.00387191772461 0.8637361526489258
github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-0d59d71-230713-095451-tat-train-baseline completed.

Model url: s3://taichi-3d-gaussian-splatting-log/tat-train-baseline/diff-lr-high-order-0d59d71-230713-095451-tat-train-baseline/output/model.tar.gz,

tensorboard output path: s3://taichi-3d-gaussian-splatting-log/tat-train-baseline/diff-lr-high-order-0d59d71-230713-095451-tat-train-baseline/output/output.tar.gz

github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-0d59d71-230713-095451-tat-train-baseline final metrics:

Latest Metrics

train:iteration train:l1loss train:loss train:num_valid_points train:psnr train:ssim train:ssimloss val:loss val:psnr val:ssim
30000.0 0.04156315699219704 0.05734596401453018 634695.0 25.112308502197266 0.8795228004455566 0.12047719955444336 0.10744719952344894 20.46383285522461 0.7787225842475891

Max Metrics

train:5kpsnr train:5kssim train:7kpsnr train:7kssim train:psnr train:ssim val:5kpsnr val:5kssim val:7kpsnr val:7kssim val:psnr val:ssim
19.573684692382812 0.6696200370788574 22.1914119720459 0.8305708765983582 25.112308502197266 0.8795228004455566 18.986038208007812 0.7217147350311279 19.548891067504883 0.7458884119987488 20.46383285522461 0.7787225842475891
github-actions[bot] commented 1 year ago

Running experiment on sagemaker with git sha 634ee23f36eea64814ff5c87bb850e9ba01de177

github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-634ee23-230713-210958-tat-train-baseline created

github-actions[bot] commented 1 year ago

Running experiment on sagemaker with git sha 634ee23f36eea64814ff5c87bb850e9ba01de177

github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-634ee23-230713-211025-tat-truck-baseline created

github-actions[bot] commented 1 year ago

Running experiment on sagemaker with git sha bdbdeff42cdd9660d9bb33b1b990f7385b7d1799

github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-bdbdeff-230713-212339-tat-train-baseline created

github-actions[bot] commented 1 year ago

Running experiment on sagemaker with git sha bdbdeff42cdd9660d9bb33b1b990f7385b7d1799

github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-bdbdeff-230713-212451-tat-truck-baseline created

github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-634ee23-230713-211025-tat-truck-baseline completed.

Model url: s3://taichi-3d-gaussian-splatting-log/tat-truck-baseline/diff-lr-high-order-634ee23-230713-211025-tat-truck-baseline/output/model.tar.gz,

tensorboard output path: s3://taichi-3d-gaussian-splatting-log/tat-truck-baseline/diff-lr-high-order-634ee23-230713-211025-tat-truck-baseline/output/output.tar.gz

github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-634ee23-230713-211025-tat-truck-baseline final metrics:

Latest Metrics

train:iteration train:l1loss train:loss train:num_valid_points train:psnr train:ssim train:ssimloss val:loss val:psnr val:ssim
30000.0 0.029478367418050766 0.04761563241481781 425253.0 25.607559204101562 0.879835307598114 0.12016469240188599 0.0541822612285614 25.15340232849121 0.864093005657196

Max Metrics

train:5kpsnr train:5kssim train:7kpsnr train:7kssim train:psnr train:ssim val:5kpsnr val:5kssim val:7kpsnr val:7kssim val:psnr val:ssim
22.76749038696289 0.8022288084030151 24.07423210144043 0.8696603775024414 25.607559204101562 0.879835307598114 23.01710319519043 0.8127131462097168 23.70611572265625 0.8350174427032471 25.15340232849121 0.864093005657196
github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-bdbdeff-230713-212451-tat-truck-baseline completed.

Model url: s3://taichi-3d-gaussian-splatting-log/tat-truck-baseline/diff-lr-high-order-bdbdeff-230713-212451-tat-truck-baseline/output/model.tar.gz,

tensorboard output path: s3://taichi-3d-gaussian-splatting-log/tat-truck-baseline/diff-lr-high-order-bdbdeff-230713-212451-tat-truck-baseline/output/output.tar.gz

github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-bdbdeff-230713-212451-tat-truck-baseline final metrics:

Latest Metrics

train:iteration train:l1loss train:loss train:num_valid_points train:psnr train:ssim train:ssimloss val:loss val:psnr val:ssim
30000.0 0.02352140098810196 0.03890781104564667 431558.0 28.2199764251709 0.8995465636253357 0.1004534363746643 0.05371525511145592 25.166696548461914 0.865128755569458

Max Metrics

train:5kpsnr train:5kssim train:7kpsnr train:7kssim train:psnr train:ssim val:5kpsnr val:5kssim val:7kpsnr val:7kssim val:psnr val:ssim
22.378013610839844 0.80736243724823 23.708072662353516 0.8396358489990234 28.2199764251709 0.8995465636253357 23.012554168701172 0.8131798505783081 23.764801025390625 0.8363999724388123 25.166696548461914 0.865128755569458
github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-634ee23-230713-210958-tat-train-baseline completed.

Model url: s3://taichi-3d-gaussian-splatting-log/tat-train-baseline/diff-lr-high-order-634ee23-230713-210958-tat-train-baseline/output/model.tar.gz,

tensorboard output path: s3://taichi-3d-gaussian-splatting-log/tat-train-baseline/diff-lr-high-order-634ee23-230713-210958-tat-train-baseline/output/output.tar.gz

github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-634ee23-230713-210958-tat-train-baseline final metrics:

Latest Metrics

train:iteration train:l1loss train:loss train:num_valid_points train:psnr train:ssim train:ssimloss val:loss val:psnr val:ssim
30000.0 0.03289628028869629 0.04381106048822403 633554.0 27.057884216308594 0.9125298261642456 0.0874701738357544 0.09862954914569855 21.049579620361328 0.8022857904434204

Max Metrics

train:5kpsnr train:5kssim train:7kpsnr train:7kssim train:psnr train:ssim val:5kpsnr val:5kssim val:7kpsnr val:7kssim val:psnr val:ssim
19.338516235351562 0.8447942137718201 21.15665054321289 0.792857825756073 27.057884216308594 0.9125298261642456 19.244234085083008 0.7334745526313782 19.815261840820312 0.7583078145980835 21.049579620361328 0.8022857904434204
github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-bdbdeff-230713-212339-tat-train-baseline completed.

Model url: s3://taichi-3d-gaussian-splatting-log/tat-train-baseline/diff-lr-high-order-bdbdeff-230713-212339-tat-train-baseline/output/model.tar.gz,

tensorboard output path: s3://taichi-3d-gaussian-splatting-log/tat-train-baseline/diff-lr-high-order-bdbdeff-230713-212339-tat-train-baseline/output/output.tar.gz

github-actions[bot] commented 1 year ago

Training job diff-lr-high-order-bdbdeff-230713-212339-tat-train-baseline final metrics:

Latest Metrics

train:iteration train:l1loss train:loss train:num_valid_points train:psnr train:ssim train:ssimloss val:loss val:psnr val:ssim
30000.0 0.03855619952082634 0.05696183443069458 641886.0 25.024015426635742 0.8694156408309937 0.13058435916900635 0.09821027517318726 21.12907600402832 0.8022344708442688

Max Metrics

train:5kpsnr train:5kssim train:7kpsnr train:7kssim train:psnr train:ssim val:5kpsnr val:5kssim val:7kpsnr val:7kssim val:psnr val:ssim
17.87548828125 0.6419569849967957 28.271541595458984 0.8978681564331055 25.024015426635742 0.8694156408309937 19.22608184814453 0.7309048771858215 19.845035552978516 0.7598333954811096 21.12907600402832 0.8022344708442688