1 pending

Run Dossier

Training run for MERGE

11 metrics · 18,912 train · 7 val

Chart Overlays

Compare this run against its peers

0/12
Add peer runs here to layer their train and val traces onto the current metric charts.
completed Feb 2, 24:21 Duration: 50m 3s Python 3.9.18 merge/val_dehaze_depth.py --pretrained_model_name_or_path=PixArt-alpha/PixArt-XL-2-512x512 --config=config/train_dehaze_depth.yaml --model_weights=/cluster/scratch/drothenpiele/MERGE/out/checkpoint-3000 --base_data_dir=/cluster/work/igp_psr/drothenpiele/data/ --log_run=2026-02-02_00-21-06_fd2a --runlog_dir=/cluster/work/igp_psr/drothenpiele/out/train/MERGE --resume_from_step=3000 --validation_steps=1000 --mixed_precision=fp16 --seed=42 --val_batch_size=1 --denoising_steps=20 --num_save_outputs=4 --num_layers=14 --num_attention_heads=16 --attention_head_dim=72 --ff_mult=1 --converter_init_type=pretrained --dataloader_num_workers=2
precision fp16 seed 42
ID 55906008 Job MERGE_d_d Part gpuhe.120h Node eu-g6-022 CPU 8 GPU nvidia_geforce_rtx_4090:1
Metrics 11/11
gpu 4
mem 3
total/gb
used/gb
util/pct
util/pct
dehaze/loss
depth/loss
loss
lr
psnr
ssim
total/loss
X-Axis
Y-Scale
Series

Loss

V
2026-02-02_00-21-06_fd2a val (smoothed)
2026-02-02_00-21-06_fd2a val (raw)

Dehaze Loss

T
2026-02-02_00-21-06_fd2a train (smoothed)
2026-02-02_00-21-06_fd2a train (raw)

Depth Loss

T
2026-02-02_00-21-06_fd2a train (smoothed)
2026-02-02_00-21-06_fd2a train (raw)

Gpu Mem Total Gb

T V
2026-02-02_00-21-06_fd2a train (smoothed)
2026-02-02_00-21-06_fd2a val (smoothed)
2026-02-02_00-21-06_fd2a train (raw)
2026-02-02_00-21-06_fd2a val (raw)

Gpu Mem Used Gb

T V
2026-02-02_00-21-06_fd2a train (smoothed)
2026-02-02_00-21-06_fd2a val (smoothed)
2026-02-02_00-21-06_fd2a train (raw)
2026-02-02_00-21-06_fd2a val (raw)

Gpu Mem Util Pct

T V
2026-02-02_00-21-06_fd2a train (smoothed)
2026-02-02_00-21-06_fd2a val (smoothed)
2026-02-02_00-21-06_fd2a train (raw)
2026-02-02_00-21-06_fd2a val (raw)

Gpu Util Pct

T V
2026-02-02_00-21-06_fd2a train (smoothed)
2026-02-02_00-21-06_fd2a val (smoothed)
2026-02-02_00-21-06_fd2a train (raw)
2026-02-02_00-21-06_fd2a val (raw)

Lr

T
2026-02-02_00-21-06_fd2a train (smoothed)
2026-02-02_00-21-06_fd2a train (raw)

Psnr

V
2026-02-02_00-21-06_fd2a val (smoothed)
2026-02-02_00-21-06_fd2a val (raw)

Ssim

V
2026-02-02_00-21-06_fd2a val (smoothed)
2026-02-02_00-21-06_fd2a val (raw)

Total Loss

T
2026-02-02_00-21-06_fd2a train (smoothed)
2026-02-02_00-21-06_fd2a train (raw)

No output snapshots found for this run.

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

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