1 pending

Run Dossier

Training run for SD 2.1. Mari

16 metrics · 4,928 train · 0 val

Chart Overlays

Compare this run against its peers

0/48
Add peer runs here to layer their train and val traces onto the current metric charts.
external assumed crashed Mar 19, 10:51 Duration: 21m 58s Python 3.11.7 /cluster/home/drothenpiele/.cache/venv/mari/bin/euler-eval /cluster/home/drothenpiele/models/stable_diffusion_mari/mari/.euler-launches/euler_launch_8c149318-f376-41f8-a987-7b0a63e65f61/config.json --rgb-fid-backend clean-fid --mask-sky -v
lr 3e-5batch 1epochs 200precision fp16model stable-diffusion-2-1 seed 42wd 0.01000grad accum 16warmup 0.05000ema nograd clip 1
ID 60828693 Job mari-train-3 Part gpuhe.120h Node eu-g6-008 CPU 8 GPU 1
mari/runs/2026-03-19_10-51-21_48d4
Metrics 16/16
gpu 4
mem 3
total/gb
used/gb
util/pct
util/pct
visibility 7
aux 2
loss
scale
loss
mean
rank/loss
std
tv/loss
dehaze/loss
depth/loss
loss
lr
preserve/loss
X-Axis
Y-Scale
Series

Loss

T
2026-03-19_10-51-21_48d4 train (smoothed)
2026-03-19_10-51-21_48d4 train (raw)

Dehaze Loss

T
2026-03-19_10-51-21_48d4 train (smoothed)
2026-03-19_10-51-21_48d4 train (raw)

Depth Loss

T
2026-03-19_10-51-21_48d4 train (smoothed)
2026-03-19_10-51-21_48d4 train (raw)

Gpu Mem Total Gb

T
2026-03-19_10-51-21_48d4 train (smoothed)
2026-03-19_10-51-21_48d4 train (raw)

Gpu Mem Used Gb

T
2026-03-19_10-51-21_48d4 train (smoothed)
2026-03-19_10-51-21_48d4 train (raw)

Gpu Mem Util Pct

T
2026-03-19_10-51-21_48d4 train (smoothed)
2026-03-19_10-51-21_48d4 train (raw)

Gpu Util Pct

T
2026-03-19_10-51-21_48d4 train (smoothed)
2026-03-19_10-51-21_48d4 train (raw)

Lr

T
2026-03-19_10-51-21_48d4 train (smoothed)
2026-03-19_10-51-21_48d4 train (raw)

Preserve Loss

T
2026-03-19_10-51-21_48d4 train (smoothed)
2026-03-19_10-51-21_48d4 train (raw)

Visibility Aux Loss

T
2026-03-19_10-51-21_48d4 train (smoothed)
2026-03-19_10-51-21_48d4 train (raw)

Visibility Aux Scale

T
2026-03-19_10-51-21_48d4 train (smoothed)
2026-03-19_10-51-21_48d4 train (raw)

Visibility Loss

T
2026-03-19_10-51-21_48d4 train (smoothed)
2026-03-19_10-51-21_48d4 train (raw)

Visibility Mean

T
2026-03-19_10-51-21_48d4 train (smoothed)
2026-03-19_10-51-21_48d4 train (raw)

Visibility Rank Loss

T
2026-03-19_10-51-21_48d4 train (smoothed)
2026-03-19_10-51-21_48d4 train (raw)

Visibility Std

T
2026-03-19_10-51-21_48d4 train (smoothed)
2026-03-19_10-51-21_48d4 train (raw)

Visibility Tv Loss

T
2026-03-19_10-51-21_48d4 train (smoothed)
2026-03-19_10-51-21_48d4 train (raw)

No output snapshots found for this run.

Outputs are generated during training and saved to outputs/epoch_N_step_M/ directories.

epoch 9 / step 880 Checkpoint #95
3/19/2026, 10:02:44 PM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-depth-vkitti/checkpoint-9
epoch 19 / step 1760 Checkpoint #96
3/19/2026, 10:02:44 PM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-depth-vkitti/checkpoint-19
epoch 29 / step 2640 Checkpoint #97
3/19/2026, 10:02:44 PM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-depth-vkitti/checkpoint-29
epoch 39 / step 3520 Checkpoint #98
3/19/2026, 10:02:44 PM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-depth-vkitti/checkpoint-39
epoch 49 / step 4400 Checkpoint #99
3/19/2026, 10:02:44 PM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-depth-vkitti/checkpoint-49
epoch 59 / step 5280 Checkpoint #100
3/19/2026, 10:02:44 PM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-depth-vkitti/checkpoint-59
epoch 69 / step 6160 Checkpoint #101
3/19/2026, 10:02:44 PM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-depth-vkitti/checkpoint-69
epoch 79 / step 7040 Checkpoint #102
3/19/2026, 10:02:44 PM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-depth-vkitti/checkpoint-79
epoch 89 / step 7920 Checkpoint #193
3/20/2026, 10:28:36 AM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-depth-vkitti/checkpoint-89
epoch 99 / step 8800 Checkpoint #194
3/20/2026, 10:28:36 AM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-depth-vkitti/checkpoint-99
epoch 109 / step 9680 Checkpoint #195
3/20/2026, 10:28:36 AM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-depth-vkitti/checkpoint-109
epoch 119 / step 10560 Checkpoint #196
3/20/2026, 10:28:36 AM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-depth-vkitti/checkpoint-119
epoch 129 / step 11440 Checkpoint #197
3/20/2026, 10:28:36 AM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-depth-vkitti/checkpoint-129
epoch 139 / step 12320 Checkpoint #198
3/20/2026, 10:28:36 AM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-depth-vkitti/checkpoint-139
epoch 149 / step 13200 Checkpoint #199
3/20/2026, 10:28:36 AM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-depth-vkitti/checkpoint-149
epoch 159 / step 14080 Checkpoint #200
3/20/2026, 10:28:36 AM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-depth-vkitti/checkpoint-159
epoch 169 / step 14960 Checkpoint #201
3/20/2026, 10:28:36 AM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-depth-vkitti/checkpoint-169
epoch 179 / step 15840 Checkpoint #301
3/22/2026, 9:54:53 AM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-depth-vkitti/checkpoint-179
epoch 189 / step 16720 Checkpoint #302
3/22/2026, 9:54:53 AM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-depth-vkitti/checkpoint-189
epoch 199 / step 17600 Checkpoint #303
3/22/2026, 9:54:53 AM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-depth-vkitti/checkpoint-199
Producer launch exports are available. Manage launch-owned exports here for quick reference.
Open Launch

Inherited Launch Exports

These exports are published by the run's producer launch.

Published

The producer launch does not publish any exports yet.

Run-Owned Exports

Publish direct filesystem paths here.

Published

No run-owned exports are published yet.

Raw Artifacts

Run-owned exports are typically direct paths, so there are no captured artifacts to publish from here.

Euler View - ML Experiment Monitor