STATE — Tahoe (X_hvg)

State Transition (ST) models using X_hvg inputs on Tahoe. This repository contains 1 fewshot and 1 zeroshot canonical models, with their saved configurations, loading files and available CellEval v1 outputs.

Paired representation · All 36 canonical models · STATE code and setup

Model index

One row is one trained model. Additional saved checkpoints are snapshots of that run, not additional models. Fewshot uses the original best.ckpt; the replaced zeroshot bundles use final.ckpt.

Regime Test context(s) Canonical checkpoint Saved evaluation
fewshot C32, HOP62, HepG2-C3A, Hs 766T, PANC-1 fewshot/state_generalization_X_hvg/checkpoints/best.ckpt CSV + H5AD bundle
zeroshot C32, HOP62, HepG2-C3A, Hs 766T, PANC-1 zeroshot/state_generalization_zeroshot_X_hvg/checkpoints/final.ckpt CSV + H5AD bundle

X_hvg versus X_state

Representation comparison

Values are equal-weight means of context means, using paired perturbations with finite values for both models. DE-count Spearman is computed across perturbations within each context. The tables and figure use the same 5 contexts in both regimes.

Fewshot

Metric X_hvg X_state
AUPRC 0.6775 0.6386
Pearson delta 0.9240 0.9147
DE-count Spearman 0.9083 0.9113
DE overlap at N 0.5254 0.5156
LFC Spearman 0.6029 0.5811
PDS L1 0.9926 0.9911

Zeroshot

Metric X_hvg X_state
AUPRC 0.3004 0.5395
Pearson delta 0.3821 0.5965
DE-count Spearman 0.4534 0.7681
DE overlap at N 0.1513 0.4433
LFC Spearman 0.4324 0.2870
PDS L1 0.7349 0.9129

Across the three dataset comparisons, the zeroshot reruns favor X_state on AUPRC, Pearson delta and DE-count Spearman. Fewshot results are closer for Tahoe and Parse; Replogle fewshot favors X_hvg. The advantage is metric-dependent: zeroshot Replogle overlap is lower with X_state, and zeroshot Tahoe/Parse LFC Spearman is lower.

These are descriptive comparisons of saved experiments, not a controlled estimate of representation effects across regimes. The original fewshot and rerun zeroshot recipes differ. All selected zeroshot runs trained for 120k steps; Tahoe uses the 768-width, sample-batch pair. For example, the original Tahoe fewshot horizons were 200k (X_hvg) and 80k (X_state). See each run’s unchanged config.yaml for architecture, batching and training settings.

Full precision: paired context scores, all available contexts, same-context summary. The reproduction script downloads only the pinned per-perturbation CSVs listed with SHA256 hashes in model_index.json. These are saved CellEval v1 scores, not CellEval2 rescoring.

Download and use

Download one complete run to retain its checkpoint, vocabulary maps, gene ordering and config. The command below excludes the large saved generations. Remove the exclusion to include adata_pred.h5ad and adata_real.h5ad.

hf download arcinstitute/ST-HVG-Tahoe \
  --revision 18118ca49c1da2a5c631ef5024448a85e0978f08 \
  --include "zeroshot/state_generalization_zeroshot_X_hvg/*" \
  --exclude "*.h5ad" --local-dir ./state-model

Install the STATE package and prepare matching input features (X_hvg), gene order, perturbation labels and context metadata. Configs retain their original training-time filesystem paths; provide a TOML with paths to your local data and the intended split rather than relying on those absolute paths.

state tx predict \
  --output-dir ./state-model/zeroshot/state_generalization_zeroshot_X_hvg \
  --checkpoint final.ckpt --toml /path/to/local-data-split.toml

For a retained fewshot run, use its directory and --checkpoint best.ckpt. Saved CSVs and generations can be inspected without rerunning prediction. The refresh preserves the saved evaluation outputs; it does not claim numerical parity from a different software version.

Collection and provenance

Dataset X_hvg X_state Canonical models
Parse ST-HVG-Parse ST-SE-Parse 16
Replogle ST-HVG-Replogle ST-SE-Replogle 16
Tahoe ST-HVG-Tahoe ST-SE-Tahoe 4

Updated 2026-11-12: 18 zeroshot bundles replaced across the collection, 18 fewshot models retained, and Plasmablast removed from the default branches. Complete selected rerun bundles include checkpoints, configs, loading files, training logs, CellEval CSVs, PR arrays and predicted/real H5ADs. Previous artifacts remain available under the revision tag before-zeroshot-refresh-20261112; historical revisions are not part of the canonical 36-model index.

Use is governed by the model license and acceptable use policy.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Collection including arcinstitute/ST-HVG-Tahoe