Try add depth cov loss to alpha

[github-actions[bot]]

Running experiment on sagemaker with git sha 066021aa123644daa3062c60638f9f8b2beaad53

[github-actions[bot]]

Training job depth-cov-alpha-066021a-231010-043747-tat-truck-baseline created

[github-actions[bot]]

Running experiment on sagemaker with git sha 066021aa123644daa3062c60638f9f8b2beaad53

[github-actions[bot]]

Training job depth-cov-alpha-066021a-231010-043832-tat-train-baseline created

[github-actions[bot]]

Training job depth-cov-alpha-066021a-231010-043747-tat-truck-baseline completed.

Model url: s3://taichi-3d-gaussian-splatting-log/tat-truck-baseline/depth-cov-alpha-066021a-231010-043747-tat-truck-baseline/output/model.tar.gz,

tensorboard output path: s3://taichi-3d-gaussian-splatting-log/tat-truck-baseline/depth-cov-alpha-066021a-231010-043747-tat-truck-baseline/output/output.tar.gz

[github-actions[bot]]

Training job depth-cov-alpha-066021a-231010-043747-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	0.16899815201759338	0.22982710599899292	1460290.0	13.249738693237305	0.5268570780754089	0.47314292192459106	0.2583858370780945	12.574592590332031	0.4547976851463318

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
13.387032508850098	0.4607965350151062	12.851665496826172	0.501745879650116	13.249738693237305	0.5268570780754089	13.144658088684082	0.461590051651001	13.1847562789917	0.4601612985134125	12.574592590332031	0.4547976851463318

[github-actions[bot]]

Training job depth-cov-alpha-066021a-231010-043832-tat-train-baseline completed.

Model url: s3://taichi-3d-gaussian-splatting-log/tat-train-baseline/depth-cov-alpha-066021a-231010-043832-tat-train-baseline/output/model.tar.gz,

tensorboard output path: s3://taichi-3d-gaussian-splatting-log/tat-train-baseline/depth-cov-alpha-066021a-231010-043832-tat-train-baseline/output/output.tar.gz

[github-actions[bot]]

Training job depth-cov-alpha-066021a-231010-043832-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	0.22788961231708527	0.28201058506965637	1926860.0	10.292320251464844	0.5015056133270264	0.49849438667297363	0.2991228401660919	11.33243179321289	0.40195339918136597

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
11.000405311584473	0.40921854972839355	13.565939903259277	0.49352020025253296	10.292320251464844	0.5015056133270264	12.242801666259766	0.4126058518886566	12.171857833862305	0.4141305387020111	11.33243179321289	0.40195339918136597

[github-actions[bot]]

Running experiment on sagemaker with git sha ef2a620359d70ff4ffbfcb8c50fa1218d6a43863

[github-actions[bot]]

Training job depth-cov-alpha-ef2a620-231010-070706-tat-truck-baseline created

[github-actions[bot]]

Running experiment on sagemaker with git sha ef2a620359d70ff4ffbfcb8c50fa1218d6a43863

[github-actions[bot]]

Training job depth-cov-alpha-ef2a620-231010-070800-tat-train-baseline created

[github-actions[bot]]

Training job depth-cov-alpha-ef2a620-231010-070706-tat-truck-baseline completed.

Model url: s3://taichi-3d-gaussian-splatting-log/tat-truck-baseline/depth-cov-alpha-ef2a620-231010-070706-tat-truck-baseline/output/model.tar.gz,

tensorboard output path: s3://taichi-3d-gaussian-splatting-log/tat-truck-baseline/depth-cov-alpha-ef2a620-231010-070706-tat-truck-baseline/output/output.tar.gz

[github-actions[bot]]

Training job depth-cov-alpha-ef2a620-231010-070706-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	0.18834741413593292	0.2695528268814087	523669.0	12.855175971984863	0.40562543272972107	0.5943745374679565	0.2713155746459961	11.975367546081543	0.4527238607406616

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
11.305298805236816	0.40053802728652954	11.623887062072754	0.4389086961746216	12.855175971984863	0.40562543272972107	12.24751091003418	0.4557172656059265	11.689866065979004	0.4469371736049652	11.975367546081543	0.4527238607406616

[github-actions[bot]]

Training job depth-cov-alpha-ef2a620-231010-070800-tat-train-baseline completed.

Model url: s3://taichi-3d-gaussian-splatting-log/tat-train-baseline/depth-cov-alpha-ef2a620-231010-070800-tat-train-baseline/output/model.tar.gz,

tensorboard output path: s3://taichi-3d-gaussian-splatting-log/tat-train-baseline/depth-cov-alpha-ef2a620-231010-070800-tat-train-baseline/output/output.tar.gz

[github-actions[bot]]

Training job depth-cov-alpha-ef2a620-231010-070800-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	0.18673770129680634	0.23802334070205688	754624.0	12.492565155029297	0.5568341016769409	0.4431658983230591	0.3074747920036316	10.969109535217285	0.39942070841789246

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
10.823578834533691	0.3788817822933197	10.171375274658203	0.42297717928886414	12.492565155029297	0.5568341016769409	11.443517684936523	0.41093453764915466	11.217612266540527	0.4085210859775543	10.969109535217285	0.39942070841789246