Benchmark

ATOM3D

ATOM3D provides molecular-structure datasets and utilities for task-specific evaluation.

Sources

drorlab/atom3d official source · Pinned README: Overview; dataset access; supported formats and splitting/filtering utilities

57 evaluations · 57 results

Overview

Datasets

Three-dimensional molecular data with associated labels/metadata; ligand-binding affinity is one documented example.

Sources

drorlab/atom3d official source · Pinned README: Overview; dataset access; supported formats and splitting/filtering utilities

Metrics

SMP: MAE; PIP/MSP/LEP: classification AUROC; RES: accuracy; LBA: RMSE and correlations; PSR/RSR: correlations of predicted structure quality with reference GDT_TS/RMSD. Metrics and aggregation belong to each task.

Sources

atom3d primary benchmark evidence · Sections 3.1–3.8, 4–5; Appendix D–F; Table 8

Allowed inputs

Three-dimensional molecular coordinates and task labels; supported formats include PDB, SDF and XYZ.

Sources

drorlab/atom3d official source · Pinned README: Overview; dataset access; supported formats and splitting/filtering utilities

Evaluation procedure diagram

How it worksEvaluation procedure

Conceptual procedure. Task variants and protocol versions retain their separate scoring conditions.

Sources (2)

drorlab/atom3d official source; atom3d primary benchmark evidence · Pinned README: Overview; dataset access; supported formats and splitting/filtering utilities; Sections 3.1–3.8, 4–5; Appendix D–F; Table 8

Source reviewed · Automated source review, 2026-09-16. All specifications and missing details

Results

Each comparison retains its reviewed evaluation scope, dataset and metric. Results are shown without a pooled ranking.

Find a protocol, dataset or metric

Evaluation scope and datasetMetric and reported setup

ATOM3D LBA-RMSE: Ligand binding affinity, root mean squared error

rmse (error) · Lower values are better.

ATOM3D LBA-RMSE: Ligand binding affinity, root mean squared error · ATOM3D LBA (ATOM3D split)

Evidence origin: Author-reported evaluation.

ATOM3D: Tasks On Molecules in Three Dimensions · Table 5, row(LBA RMSE)

The paper marks some runs with an asterisk to say their training data differed, though the splitting criteria were the same.

Comparison details and limitations

Every method ATOM3D reports on Ligand binding affinity, root mean squared error, scored with RMSE on ATOM3D LBA.

Author-reported numbers, source checked but not independently reproduced.
MAE and RMSE are errors, better when lower. The other metrics are better when higher.
Comparison methods are named by the citation the table prints; the paper's text says which method each is.

Automated source review: 2026-09-18. Numerical source review does not establish independent reproduction.

Dots show point estimates. Whiskers show only explicitly defined uncertainty (standard deviation, standard error or a labelled interval); their definitions remain in Table. Unresolved uncertainty is not plotted. Differences do not establish statistical significance.

Find a tested model or method

Showing 5 of 5 matching rows.

Zoom to observed range

Tested configuration

1.391.521.651.791.92

Reported score

DeepAffinity (unified RNN/RNN-CNN; DSSP-derived SPS) (cited as [Karimi et al., 2019])
1.89
GNN
1.6
ENN
1.57
DeepDTA (ATOM3D baseline) (cited as [Öztürk et al., 2018])
1.56
3DCNN
1.42

Methods and evaluation design

Procedure, tasks and evaluated configurations

How it works

Evaluation methodology

ATOM3D evaluates predictions from molecular structures using eight separate tasks. It supplies task-specific reference labels and partitions, ranging from random small molecules to held-out protein families and future structure-prediction targets. Each task has its own metric and representation-matched baseline; the suite is not a single universal structure score.

Sources

atom3d primary benchmark evidence · Sections 3.1–3.8, 4–5; Appendix D–F; Table 8

Evaluation design

Benchmarks bring together tasks and protocols. A task describes the biological question; a protocol defines a particular test.

Tasks

These source-backed links do not make different protocols or scores interchangeable.

Baseline coverage

Reference methods help show what a model adds beyond simple controls. We track a null control and a conventional method for each protocol.

No concrete protocols are explicitly linked to this suite. Protocol identification and baseline selection are outstanding.

Protocol coverage CSV · Model evaluation matrix · Source table · Release and checksums

Coverage is derived from release 2026-09-29-06401fd5b220. Source citations describe the original records; they do not validate an unreviewed baseline proposal. No results have been generated by this audit.

Run this benchmark

Download and load an ATOM3D task

Install the library and download one of the eight three-dimensional molecular tasks scored on this page.

Generate predictions and evaluate them. This recipe does not establish reproduction of a particular published score.

Dataset access: Downloaded by the library into a local LMDB dataset.
Model and weights: None. The reported networks are trained from scratch.
Licences: Project licence: MIT. Upstream data licences are separate and unreported here.
Software: Python with the atom3d package from PyPI.
Hardware: Not stated in the cited section. Several of these steps expect a GPU.

Required inputs and expected outputs

Inputs

A model over three-dimensional atomic structure.

Outputs

An LMDB dataset for the chosen task and split.

Execution steps

1. Install (Command line)
Source reviewed; these instructions have not been executed by rewire.
```
pip install atom3d
```
ATOM3D: repository README · README.md at 4c2f3b7e, Installation, lines 24-24
2. Download a dataset (Python)
Source reviewed; these instructions have not been executed by rewire.
```
import atom3d.datasets as da
da.download_dataset('lba', PATH_TO_DATASET) # Download LBA dataset.
```
ATOM3D: repository README · README.md at 4c2f3b7e, Downloading a dataset, lines 42-43

3. Load a dataset (Python)

Source reviewed; these instructions have not been executed by rewire.

import atom3d.datasets as da
dataset = da.load_dataset(PATH_TO_DATASET, {'lmdb','pdb','silent','sdf','xyz','xyz-gdb'})
print(len(dataset))  # Print length
print(dataset[0].keys())  # Print keys

ATOM3D: repository README · README.md at 4c2f3b7e, Loading a dataset, lines 52-55

Use your own model

Run your model locally and return predictions keyed by the input IDs. The evaluator supplies biological inputs without test labels and owns scoring. This interface is not a sandbox for model code.

Pass your existing prediction function into this adapter. Its output direction must match the selected protocol.

class MyModelAdapter:
    def __init__(self, score):
        self.score = score

    def predict(self, inputs):
        return {row["id"]: float(self.score(row)) for row in inputs}

# adapter = MyModelAdapter(your_prediction_function)
# report = rewirebench.run(prepared, adapter, output="runs/my-model")

Alternatively, generate a keyed prediction file in your existing model environment and use the score-only recipe. Your model code and weights do not need to be shared.

ATOM3D: repository README · README.md at 4c2f3b7e

Scope and limitations

Quoted from the project's README and not executed by rewire, so the commands are evidence of what the project documents rather than a verified run.
The project may have changed since the pinned commit.
Two of the metrics on this page are errors where lower is better.
The comparison methods appear in the paper only as citations.

Contribute a result for review. The library can submit an exported evaluation for private review when intake is open. Check the contribution page for access and sign-in.

Original repository instructions

Run this benchmark

Official installation, dataset-download and LMDB loading examples are available. These sections do not supply a complete task-specific model-training and scoring run; select the task and follow its example/documentation before reporting a benchmark result.

A maintained rewire runner has not been verified for this benchmark. Check data access, weights, licences, dependencies and hardware in the linked official documentation; requirements have not been fully extracted.

drorlab/atom3d / README.md · README.md lines 19–70 (Installation and Usage)

Strengths, limitations and unresolved questions

Strengths and limitations

Strengths supported by sources

Common data loaders allow different 3D representations to use the same dataset definitions.
Sources
drorlab/atom3d official source · Pinned README: Overview; dataset access; supported formats and splitting/filtering utilities

Limitations and conditions

Comparisons must retain the task, split and available structural information. In particular, the two LBA identity thresholds test different generalization regimes.
Sources
atom3d primary benchmark evidence · Sections 3.1–3.8, 4–5; Appendix D–F; Table 8

Profile review details

Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction.

Stable record: discovery-benchmark-atom3d

Specifications

Inputs, training, access and other details

Explanatory profile: source reviewed · Automated source review, 2026-09-16. Review applies to the cited claims; unresolved fields are listed below. Numerical results retain their own review status.

Data, procedure and scoring
Property	Description and evidence
Datasets	Three-dimensional molecular data with associated labels/metadata; ligand-binding affinity is one documented example. Sources drorlab/atom3d official source · Pinned README: Overview; dataset access; supported formats and splitting/filtering utilities
Splits	Task-specific: random molecules for SMP; 30% protein sequence identity for PIP, MSP and the strict LBA split; CATH topology groups for RES; protein targets for LEP; competition years for PSR and RSR. LBA also provides a less restrictive 60% identity split. Sources atom3d primary benchmark evidence · Sections 3.1–3.8, 4–5; Appendix D–F; Table 8
Metrics	SMP: MAE; PIP/MSP/LEP: classification AUROC; RES: accuracy; LBA: RMSE and correlations; PSR/RSR: correlations of predicted structure quality with reference GDT_TS/RMSD. Metrics and aggregation belong to each task. Sources atom3d primary benchmark evidence · Sections 3.1–3.8, 4–5; Appendix D–F; Table 8
Baselines	3D CNNs, graph networks and equivariant networks are compared with task-specific 1D/2D methods; structure-ranking tasks also use established 3D methods. Appendix F documents the per-task comparators. Sources atom3d primary benchmark evidence · Sections 3.1–3.8, 4–5; Appendix D–F; Table 8
Leakage controls	Protein-sequence, topology, target and temporal partitions reduce task-specific overlap; PIP prunes DIPS proteins against DB5. SMP uses a random molecular split, so it is not a scaffold-held-out test. Sources atom3d primary benchmark evidence · Sections 3.1–3.8, 4–5; Appendix D–F; Table 8
Uncertainty	Table 8 reports standard deviations over three replicates. The paper distinguishes this run variation from the choice of molecular dataset and split. Sources atom3d primary benchmark evidence · Sections 3.1–3.8, 4–5; Appendix D–F; Table 8
Entity type	Molecular dataset and evaluation suite. Sources drorlab/atom3d official source · Pinned README: Overview; dataset access; supported formats and splitting/filtering utilities
Organisms	No single organism defines this suite of molecular structure datasets. · Not applicable Sources drorlab/atom3d official source · Pinned README: Overview; dataset access; supported formats and splitting/filtering utilities
Assays	Task-specific structural and molecular-property labels. Sources drorlab/atom3d official source · Pinned README: Overview; dataset access; supported formats and splitting/filtering utilities
Allowed inputs	Three-dimensional molecular coordinates and task labels; supported formats include PDB, SDF and XYZ. Sources drorlab/atom3d official source · Pinned README: Overview; dataset access; supported formats and splitting/filtering utilities
Adaptation	Supervised task evaluation; task-specific training configurations are linked separately. Sources drorlab/atom3d official source · Pinned README: Overview; dataset access; supported formats and splitting/filtering utilities

Evidence

Source checking verifies the cited claim or transcription. It does not establish independent reproduction.

Papers and result coverage

Last literature check: 2026-09-17. Dated primary-source discovery and protocol/table screening. Source checking does not mean experimental reproduction. Only separately extracted and independently reviewed numeric batches are publishable.

Paper or primary resource	Version	Reference
ATOM3D: Tasks On Molecules in Three Dimensions	Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256	Read source

Historical gaps recorded on 2026-09-17

The catalogue now holds 57 result rows for this benchmark. A note below about pending extraction describes the state on 2026-09-17 and may since have been answered by a later batch. The result rows and their sources are the current record.

complete comparable numeric result batch: Specific candidate tables and protocol boundaries are documented; no graph values or incomplete winner-only selection are converted into publishable rows.

Search and extraction details

primary protocol reviewed

Searches

ATOM3D benchmark paper Table results 2012.04035

Evidence locations

Tables 3–8; Table 8 reports all task metrics and replicate standard deviations

Evidence table

Inspect claims, sources and review details

Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.

One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.

29 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-09-29-06401fd5b220
Property and statement	Original source and location	Review and provenance
Diagram caption Conceptual procedure. Task variants and protocol versions retain their separate scoring conditions. Individual claims	atom3d primary benchmark evidence Original source ↗ Pinned README: Overview; dataset access; supported formats and splitting/filtering utilities; Sections 3.1–3.8, 4–5; Appendix D–F; Table 8 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: arXiv:2012.04035v4 (15 January 2022) Retrieved: 2026-09-16T21:05:49.738972+00:00	source checked automated source review · 2026-09-16 Audit details Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction. Field: `attributes.profile.diagram.caption` Source artifact SHA-256: `92656c20a15311c32bed9edc7f465bb26eb30338f40324fe43bec4b1fc6a7890` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Diagram caption Conceptual procedure. Task variants and protocol versions retain their separate scoring conditions. Individual claims	drorlab/atom3d official source Original source ↗ Pinned README: Overview; dataset access; supported formats and splitting/filtering utilities; Sections 3.1–3.8, 4–5; Appendix D–F; Table 8 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 4c2f3b7e9efe128791b83f03b2e8cae91e78b018 Retrieved: 2026-09-16T10:31:32.435757+00:00	source checked automated source review · 2026-09-16 Audit details Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction. Field: `attributes.profile.diagram.caption` Source artifact SHA-256: `6e404699412cb687bd737c4423c568f6ece1bad9f62a13e15ac1785062ccccd1` Hash scope: Hash scope not separately documented; inspect source record
Diagram steps Allowed inputs: Three-dimensional molecular coordinates and task labels; supported formats include PDB, SDF and XYZ. Splits: Task-specific: random molecules for SMP; 30% protein sequence identity for PIP, MSP and the strict LBA split; CATH topology groups for RES; protein targets for LEP; competition years for PSR and RSR. LBA also provides a less restrictive 60% identity split. Metrics: SMP: MAE; PIP/MSP/LEP: classification AUROC; RES: accuracy; LBA: RMSE and correlations; PSR/RSR: correlations of predicted structure quality with reference GDT_TS/RMSD. Metrics and aggregation belong to each task. Individual claims	atom3d primary benchmark evidence Original source ↗ Pinned README: Overview; dataset access; supported formats and splitting/filtering utilities; Sections 3.1–3.8, 4–5; Appendix D–F; Table 8 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: arXiv:2012.04035v4 (15 January 2022) Retrieved: 2026-09-16T21:05:49.738972+00:00	source checked automated source review · 2026-09-16 Audit details Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction. Field: `attributes.profile.diagram.steps` Source artifact SHA-256: `92656c20a15311c32bed9edc7f465bb26eb30338f40324fe43bec4b1fc6a7890` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Diagram steps Allowed inputs: Three-dimensional molecular coordinates and task labels; supported formats include PDB, SDF and XYZ. Splits: Task-specific: random molecules for SMP; 30% protein sequence identity for PIP, MSP and the strict LBA split; CATH topology groups for RES; protein targets for LEP; competition years for PSR and RSR. LBA also provides a less restrictive 60% identity split. Metrics: SMP: MAE; PIP/MSP/LEP: classification AUROC; RES: accuracy; LBA: RMSE and correlations; PSR/RSR: correlations of predicted structure quality with reference GDT_TS/RMSD. Metrics and aggregation belong to each task. Individual claims	drorlab/atom3d official source Original source ↗ Pinned README: Overview; dataset access; supported formats and splitting/filtering utilities; Sections 3.1–3.8, 4–5; Appendix D–F; Table 8 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 4c2f3b7e9efe128791b83f03b2e8cae91e78b018 Retrieved: 2026-09-16T10:31:32.435757+00:00	source checked automated source review · 2026-09-16 Audit details Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction. Field: `attributes.profile.diagram.steps` Source artifact SHA-256: `6e404699412cb687bd737c4423c568f6ece1bad9f62a13e15ac1785062ccccd1` Hash scope: Hash scope not separately documented; inspect source record
Diagram title Evaluation procedure Individual claims	atom3d primary benchmark evidence Original source ↗ Pinned README: Overview; dataset access; supported formats and splitting/filtering utilities; Sections 3.1–3.8, 4–5; Appendix D–F; Table 8 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: arXiv:2012.04035v4 (15 January 2022) Retrieved: 2026-09-16T21:05:49.738972+00:00	source checked automated source review · 2026-09-16 Audit details Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction. Field: `attributes.profile.diagram.title` Source artifact SHA-256: `92656c20a15311c32bed9edc7f465bb26eb30338f40324fe43bec4b1fc6a7890` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Diagram title Evaluation procedure Individual claims	drorlab/atom3d official source Original source ↗ Pinned README: Overview; dataset access; supported formats and splitting/filtering utilities; Sections 3.1–3.8, 4–5; Appendix D–F; Table 8 Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 4c2f3b7e9efe128791b83f03b2e8cae91e78b018 Retrieved: 2026-09-16T10:31:32.435757+00:00	source checked automated source review · 2026-09-16 Audit details Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction. Field: `attributes.profile.diagram.title` Source artifact SHA-256: `6e404699412cb687bd737c4423c568f6ece1bad9f62a13e15ac1785062ccccd1` Hash scope: Hash scope not separately documented; inspect source record
Datasets Three-dimensional molecular data with associated labels/metadata; ligand-binding affinity is one documented example. Individual claims	drorlab/atom3d official source Original source ↗ Pinned README: Overview; dataset access; supported formats and splitting/filtering utilities Version: 4c2f3b7e9efe128791b83f03b2e8cae91e78b018 Retrieved: 2026-09-16T10:31:32.435757+00:00	source checked automated source review · 2026-09-16 Audit details Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction. Field: `attributes.profile.facts.0.value` Source artifact SHA-256: `6e404699412cb687bd737c4423c568f6ece1bad9f62a13e15ac1785062ccccd1` Hash scope: Hash scope not separately documented; inspect source record
Splits Task-specific: random molecules for SMP; 30% protein sequence identity for PIP, MSP and the strict LBA split; CATH topology groups for RES; protein targets for LEP; competition years for PSR and RSR. LBA also provides a less restrictive 60% identity split. Individual claims	atom3d primary benchmark evidence Original source ↗ Sections 3.1–3.8, 4–5; Appendix D–F; Table 8 Version: arXiv:2012.04035v4 (15 January 2022) Retrieved: 2026-09-16T21:05:49.738972+00:00	source checked automated source review · 2026-09-16 Audit details Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction. Field: `attributes.profile.facts.1.value` Source artifact SHA-256: `92656c20a15311c32bed9edc7f465bb26eb30338f40324fe43bec4b1fc6a7890` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Adaptation Supervised task evaluation; task-specific training configurations are linked separately. Individual claims	drorlab/atom3d official source Original source ↗ Pinned README: Overview; dataset access; supported formats and splitting/filtering utilities Version: 4c2f3b7e9efe128791b83f03b2e8cae91e78b018 Retrieved: 2026-09-16T10:31:32.435757+00:00	source checked automated source review · 2026-09-16 Audit details Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction. Field: `attributes.profile.facts.10.value` Source artifact SHA-256: `6e404699412cb687bd737c4423c568f6ece1bad9f62a13e15ac1785062ccccd1` Hash scope: Hash scope not separately documented; inspect source record
Metrics SMP: MAE; PIP/MSP/LEP: classification AUROC; RES: accuracy; LBA: RMSE and correlations; PSR/RSR: correlations of predicted structure quality with reference GDT_TS/RMSD. Metrics and aggregation belong to each task. Individual claims	atom3d primary benchmark evidence Original source ↗ Sections 3.1–3.8, 4–5; Appendix D–F; Table 8 Version: arXiv:2012.04035v4 (15 January 2022) Retrieved: 2026-09-16T21:05:49.738972+00:00	source checked automated source review · 2026-09-16 Audit details Primary paper and/or task implementation reviewed for the explicitly cited methodology claims. Scope-limited absence is recorded only after the documented source search; no model runs or independent reproduction. Field: `attributes.profile.facts.2.value` Source artifact SHA-256: `92656c20a15311c32bed9edc7f465bb26eb30338f40324fe43bec4b1fc6a7890` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact

All evidence (CSV)All evidence (JSONL)How to read the evidence

Sources and history

View linked audit checks and correction history

Release 2026-09-29-06401fd5b220 · Record review: discovered

5 source records and release history

drorlab/atom3d official source · Original source · 4c2f3b7e9efe128791b83f03b2e8cae91e78b018
ATOM3D: Tasks On Molecules in Three Dimensions · Original source · Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256
drorlab/atom3d / README.md · Original source · Git commit 4c2f3b7e9efe128791b83f03b2e8cae91e78b018
ATOM3D: repository README · Original source · 4c2f3b7e9efe128791b83f03b2e8cae91e78b018
atom3d primary benchmark evidence · Original source · arXiv:2012.04035v4 (15 January 2022)

Download this release

Technical metadata and extraction receipts

Stable ID: discovery-benchmark-atom3d

areas: molecular-interactions
entity level: suite
scope note: Specialist molecular or omics evaluation; protocol details require review before numerical comparison.
task: Three-dimensional molecular learning tasks
version: Not reported
benchmark research: review date: 2026-09-17; status: primary_protocol_reviewed; primary sources: evidence-expansion-atom3d-92656c20; inspected locators: Tables 3–8; Table 8 reports all task metrics and replicate standard deviations; searched queries: ATOM3D benchmark paper Table results 2012.04035; gaps: complete comparable numeric result batch: Specific candidate tables and protocol boundaries are documented; no graph values or incomplete winner-only selection are converted into publishable rows.; claim scope: Dated primary-source discovery and protocol/table screening. Source checking does not mean experimental reproduction. Only separately extracted and independently reviewed numeric batches are publishable.
historical missing metadata: dataset release: unextracted; metric implementation: unextracted; split manifest: unextracted; version: unextracted
metadata review scope: historical_missing_metadata preserves the original discovery state. Current descriptive evidence and missingness are recorded in profile.facts; numerical-result review is separate.
entity classification: review date: 2026-09-17; rationale: The cited profile describes a collection of evaluation tasks or protocols; retain it as the top-level benchmark suite. Its datasets and individual protocols remain separate records.; source ids: src-discovery-drorlab-atom3d; source locator: Pinned README: Overview; dataset access; supported formats and splitting/filtering utilities; ambiguities: None recorded
run documentation: record id: discovery-benchmark-atom3d; source ids: run-doc-atom3d-readme-md-4c2f3b7e; status: official_documentation_linked; summary: Official installation, dataset-download and LMDB loading examples are available. These sections do not supply a complete task-specific model-training and scoring run; select the task and follow its example/documentation before reporting a benchmark result.; source locator: README.md lines 19–70 (Installation and Usage)
run recipes: id: atom3d-official; protocol id: discovery-benchmark-atom3d; version: 4c2f3b7e9efe128791b83f03b2e8cae91e78b018; title: Download and load an ATOM3D task; purpose: generate_and_evaluate; summary: Install the library and download one of the eight three-dimensional molecular tasks scored on this page.; inputs: A model over three-dimensional atomic structure.; outputs: An LMDB dataset for the chosen task and split.; requirements: data: Downloaded by the library into a local LMDB dataset.; weights: None. The reported networks are trained from scratch.; licence: Project licence: MIT. Upstream data licences are separate and unreported here.; software: Python with the atom3d package from PyPI.; hardware: Not stated in the cited section. Several of these steps expect a GPU.; instructions: runtime: command_line; title: Install; code: pip install atom3d; status: source_reviewed_not_executed; source ids: project-recipe-atom3d-4c2f3b7e; source locator: README.md at 4c2f3b7e, Installation, lines 24-24; runtime: python; title: Download a dataset; code: import atom3d.datasets as da da.download_dataset('lba', PATH_TO_DATASET) # Download LBA dataset.; status: source_reviewed_not_executed; source ids: project-recipe-atom3d-4c2f3b7e; source locator: README.md at 4c2f3b7e, Downloading a dataset, lines 42-43; runtime: python; title: Load a dataset; code: import atom3d.datasets as da dataset = da.load_dataset(PATH_TO_DATASET, {'lmdb','pdb','silent','sdf','xyz','xyz-gdb'}) print(len(dataset)) # Print length print(dataset[0].keys()) # Print keys; status: source_reviewed_not_executed; source ids: project-recipe-atom3d-4c2f3b7e; source locator: README.md at 4c2f3b7e, Loading a dataset, lines 52-55; limitations: Quoted from the project's README and not executed by rewire, so the commands are evidence of what the project documents rather than a verified run.; The project may have changed since the pinned commit.; Two of the metrics on this page are errors where lower is better.; The comparison methods appear in the paper only as citations.; source ids: project-recipe-atom3d-4c2f3b7e; source locator: README.md at 4c2f3b7e

Related records

Suggest a correction

Overview

Datasets

Metrics

Allowed inputs

Results

Methods and evaluation design

How it works

Evaluation design

Tasks

Baseline coverage

Run this benchmark

Download and load an ATOM3D task

Inputs

Outputs

Execution steps

1. Install (Command line)

2. Download a dataset (Python)

3. Load a dataset (Python)

Run this benchmark

Strengths and limitations

Strengths supported by sources

Limitations and conditions

Specifications

Evidence

Papers and result coverage

Searches

Evidence locations

Evidence table

Sources and history