Datasets:

aditijc
/

smooth-hierarchy-ucf101-canary-v1

Modalities:

Formats:

Size:

Tags:

smooth-hierarchy

Libraries:

Dataset card Data Studio Files Files and versions

Dataset Viewer

Auto-converted to Parquet Duplicate

Split (1)

train · 1 rows

experiment_slug string	job_id string	cluster string	node string	gpu string	elapsed string	exit_code int64	global_steps int64	max_epochs int64	limit_train_batches int64	image_size int64	batch_size int64	lr float64	dataset string	data_root string	model_arch string	lambda1 float64	lambda2 float64	lambda3 float64	lambda4 float64	skip_beta_power float64	recon_all_betas bool	frames_per_clip int64	step_between_clips int64	delta_t int64	val_fraction float64	checkpoint_every_n_train_steps int64	wandb_log_images_every int64	wandb_run_url string	loss_components_logged string	beta_image_panels string	artifact_status string	canary bool	script_name string	sbatch string	repo_head string	model string	description string
smooth-hierarchy-ucf101	6876770	torch	gh116	NVIDIA H200	00:06:28	0	400	2	200	64	128	0.0001	ucf101	/scratch/ajc10180/data/ucf101	FiLM-conditioned U-Net	1	1	1	1	1	false	32	8	2	0.05	100	50	https://wandb.ai/aditijc5-new-york-university/smooth-hierarchy/runs/9yngjsbe	l1,l2,l3,l4,total_loss	0,0.1,0.5,0.9,1.0	final	true	train.py	canary.sbatch	348c9cf	FiLM-UNet	Canary run for smooth-hierarchy-ucf101: 2 epochs x 200 batches on UCF101 @ 64x64, verifying pipeline before 50-epoch main run.

smooth-hierarchy-ucf101 — Canary v1

Canary run artifacts for the smooth-hierarchy-ucf101 experiment. This is a pipeline-verification run, not the main training run. It establishes that:

FFmpeg 7.1 + torchcodec 0.11.0 + torch 2.11.0+cu130 work on NYU torch H200 nodes
The 4-loss FiLM U-Net (L1 recon @ β=1, L2 collapse @ β=0, L3 abstraction-margin, L4 temporal-margin) trains end-to-end
Checkpoints save every 100 train steps + every epoch
W&B logs all 4 loss components and reconstruction panels at β ∈ {0, 0.1, 0.5, 0.9, 1.0}

Provenance

Job: torch:6876770 on gh116 (NVIDIA H200, 143 GB VRAM)
Elapsed: 6m28s, exit 0
Global steps: 400 (2 epochs × 200 --limit_train_batches)
W&B: canary-6876770
Repo head: 348c9cf

Files

File	Purpose
`last.ckpt`	Most recent checkpoint (47 MB) — rolling
`smooth-hierarchy-epoch=01-step=00000400.ckpt`	Final step-indexed checkpoint
`smooth-hierarchy-epoch=epoch=01.ckpt`	Epoch-1 end checkpoint
`canary.sbatch`	The sbatch script used for this run
`canary-6876770.out` / `canary-6876770.err`	Full job logs
`canary_metadata.parquet`	Run config + hyperparams for reproducibility
`uv.lock`	Exact Python dep resolution for bit-exact env reconstruction

Hyperparameters (canary)

Dataset: UCF101 (101 classes, 13,320 videos) pulled from quchenyuan/UCF101-ZIP HF mirror (CRCV source was flaky)
Image size: 64 × 64
Batch size: 128
Max epochs: 2
--limit_train_batches: 200
LR: 1e-4
λ₁ = λ₂ = λ₃ = λ₄ = 1.0
--skip_beta_power: 1.0
--recon_all_betas: False
--frames_per_clip: 32, --step_between_clips: 8, --delta_t: 2

How to reproduce

Obtain the smooth-hierarchy training repo at head 348c9cf from its source of record (private repo; see experiment owner).
On an HPC with an H100/H200/A100 GPU, run setup.sbatch (fetches UCF101 via HF mirror, builds venv via uv using the uv.lock in this dataset).
sbatch canary.sbatch with WANDB_API_KEY + HF_TOKEN available in ~/.raca/keys.env.

Exact dependency set is pinned in uv.lock in this dataset.

Downloads last month: 61

Number of rows:

Total file size: