Euler View
No jobs

Run Dossier

Training run for SD 2.1. Mari

42 metrics · 53,526 train · 292 val

Chart Overlays

Compare this run against its peers

0/105
Add peer runs here to layer their train and val traces onto the current metric charts.
crashed Jul 20, 14:34 Duration: 17h 0m Python 3.11.7 train_joint.py --config /cluster/home/drothenpiele/models/stable_diffusion_mari/mari/.euler-launches/euler_launch_f6d64967-2276-428d-8dba-f576f2b10ca6/joint_vkitti.yaml --resume /cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-metric-2026-07-19-17-44-22-88f6/checkpoint-11
lr 4e-5batch 16epochs 200precision fp16model stable-diffusion-2-1 seed 42wd 0grad accum 1warmup 0.05000ema nograd clip 1
ID 7852386
mari/runs/2026-07-20_14-34-09_8844
Error
BrokenPipeError: [Errno 32] Broken pipe
Metrics 42/42
depth.train 21
diag 15
depth_abs_rel
depth_delta1
depth_mae
valset_muses_absrel
valset_muses_delta1
valset_muses_delta2
valset_muses_delta3
valset_muses_mae
valset_muses_num_samples
valset_muses_projected_pixels
valset_muses_rmse
valset_muses_rmse_log
valset_muses_silog
valset_muses_sqrel
valset_muses_valid_pixels
loss 6
depth
depth_head
depth_tv
edge_smoothness
multiscale_smoothness
smoothness_aux
rgb.train 11
diag 2
image_psnr
image_ssim
loss 7
dehaze
preserve
total
visibility
visibility_aux
visibility_rank
visibility_tv
stat 2
visibility_mean
visibility_std
sys.train 6
gpu_mem_total_gb
gpu_mem_used_gb
gpu_mem_util_pct
gpu_util_pct
lr
visibility_aux_scale
Legacy 4
data/camera 3
enabled/any/source
fields/emitted
rays/emitted
train/camera/valid/fraction
42 selected
X-Axis
Y-Scale
Series
Group
Namespace

Depth Abs Rel

↓ kind=diag
V
2026-07-20_14-34-09_8844 val (smoothed) —
2026-07-20_14-34-09_8844 val (raw) —

Depth Delta1

↑ kind=diag
V
2026-07-20_14-34-09_8844 val (smoothed) —
2026-07-20_14-34-09_8844 val (raw) —

Depth Mae

kind=diag
V
2026-07-20_14-34-09_8844 val (smoothed) —
2026-07-20_14-34-09_8844 val (raw) —

Valset Muses Absrel

kind=diag
V
2026-07-20_14-34-09_8844 val (smoothed) —
2026-07-20_14-34-09_8844 val (raw) —

Valset Muses Delta1

kind=diag
V
2026-07-20_14-34-09_8844 val (smoothed) —
2026-07-20_14-34-09_8844 val (raw) —

Valset Muses Delta2

kind=diag
V
2026-07-20_14-34-09_8844 val (smoothed) —
2026-07-20_14-34-09_8844 val (raw) —

Valset Muses Delta3

kind=diag
V
2026-07-20_14-34-09_8844 val (smoothed) —
2026-07-20_14-34-09_8844 val (raw) —

Valset Muses Mae

kind=diag
V
2026-07-20_14-34-09_8844 val (smoothed) —
2026-07-20_14-34-09_8844 val (raw) —

Valset Muses Num Samples

kind=diag
V
2026-07-20_14-34-09_8844 val (smoothed) —
2026-07-20_14-34-09_8844 val (raw) —

Valset Muses Projected Pixels

kind=diag
V
2026-07-20_14-34-09_8844 val (smoothed) —
2026-07-20_14-34-09_8844 val (raw) —

Valset Muses Rmse

kind=diag
V
2026-07-20_14-34-09_8844 val (smoothed) —
2026-07-20_14-34-09_8844 val (raw) —

Valset Muses Rmse Log

kind=diag
V
2026-07-20_14-34-09_8844 val (smoothed) —
2026-07-20_14-34-09_8844 val (raw) —

Valset Muses Silog

kind=diag
V
2026-07-20_14-34-09_8844 val (smoothed) —
2026-07-20_14-34-09_8844 val (raw) —

Valset Muses Sqrel

kind=diag
V
2026-07-20_14-34-09_8844 val (smoothed) —
2026-07-20_14-34-09_8844 val (raw) —

Valset Muses Valid Pixels

kind=diag
V
2026-07-20_14-34-09_8844 val (smoothed) —
2026-07-20_14-34-09_8844 val (raw) —

Depth

↓ kind=loss
T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

Depth Head

kind=loss
T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

Depth Tv

↓ kind=loss
T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

Edge Smoothness

↓ kind=loss
T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

Multiscale Smoothness

↓ kind=loss
T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

Smoothness Aux

↓ kind=loss
T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

Image Psnr

↑ dB kind=diag
V
2026-07-20_14-34-09_8844 val (smoothed) —
2026-07-20_14-34-09_8844 val (raw) —

Image Ssim

↑ kind=diag
V
2026-07-20_14-34-09_8844 val (smoothed) —
2026-07-20_14-34-09_8844 val (raw) —

Dehaze

↓ kind=loss
T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

Preserve

↓ kind=loss
T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

Total

↓ kind=loss
T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

Visibility

↓ kind=loss
T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

Visibility Aux

↓ kind=loss
T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

Visibility Rank

↓ kind=loss
T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

Visibility Tv

↓ kind=loss
T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

Visibility Mean

kind=stat
T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

Visibility Std

kind=stat
T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

Gpu Mem Total Gb

T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

Gpu Mem Used Gb

T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

Gpu Mem Util Pct

T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

Gpu Util Pct

T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

Lr

T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

Visibility Aux Scale

T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

Data/Camera Enabled Any Source

V
2026-07-20_14-34-09_8844 val (smoothed) —
2026-07-20_14-34-09_8844 val (raw) —

Data/Camera Fields Emitted

V
2026-07-20_14-34-09_8844 val (smoothed) —
2026-07-20_14-34-09_8844 val (raw) —

Data/Camera Rays Emitted

V
2026-07-20_14-34-09_8844 val (smoothed) —
2026-07-20_14-34-09_8844 val (raw) —

Train/Camera Valid Fraction

T
2026-07-20_14-34-09_8844 train (smoothed) —
2026-07-20_14-34-09_8844 train (raw) —

No output snapshots found for this run.

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

epoch 13 / step 22666 Checkpoint #4021
7/20/2026, 2:44:20 PM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-metric-2026-07-20-14-34-09-8844/checkpoint-13
epoch 15 / step 25904 Checkpoint #4024
7/20/2026, 6:56:41 PM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-metric-2026-07-20-14-34-09-8844/checkpoint-15
epoch 17 / step 29142 Checkpoint #4025
7/20/2026, 6:56:41 PM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-metric-2026-07-20-14-34-09-8844/checkpoint-17
epoch 19 / step 32380 Checkpoint #4029
7/20/2026, 9:04:06 PM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-metric-2026-07-20-14-34-09-8844/checkpoint-19
epoch 21 / step 35618 Checkpoint #4034
7/21/2026, 1:35:37 AM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-metric-2026-07-20-14-34-09-8844/checkpoint-21
epoch 23 / step 38856 Checkpoint #4035
7/21/2026, 1:35:37 AM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-metric-2026-07-20-14-34-09-8844/checkpoint-23
epoch 25 / step 42094 Checkpoint #4042
7/21/2026, 8:59:40 AM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-metric-2026-07-20-14-34-09-8844/checkpoint-25
epoch 27 / step 45332 Checkpoint #4043
7/21/2026, 8:59:40 AM
/cluster/scratch/drothenpiele/euler_train/mari/checkpoints/joint-dehaze-metric-2026-07-20-14-34-09-8844/checkpoint-27
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