onescience_manifest_version: "0.1"
resource_type: dataset
resource:
  id: OneScience/oc20
  name: OC20
  domain: matchem
  domain_tags:
    - matchem
    - catalysis
    - oc20
    - atomistic_dataset
  task: mlip_training_data
  task_tags:
    - mlip
    - energy_prediction
    - force_prediction
    - extxyz_training
  modalities:
    - atomistic_structure
  input_formats:
    - extxyz
  output_formats: []
  summary: 面向 OneScience/UMA 微调和训练评测示例的 Open Catalyst 2020 (OC20) S2EF extxyz 子集。

platform_resource:
  primary:
    platform: modelscope
    repo_id: OneScience/oc20
    repo_type: dataset
    url: https://modelscope.cn/datasets/OneScience/oc20
    revision: main
    readme_path: README.md
    manifest_path: manifest.yaml
  mirrors: []
  access:
    visibility: public
    license: Apache License 2.0

website_integration:
  enabled: true
  click_target:
    platform: modelscope
    resource_url: https://modelscope.cn/datasets/OneScience/oc20
  llm_handoff:
    readme_required: true
    manifest_required: true
    download_readme_first: true
    resolve_related_models: true
    default_run_goal: minimal_validation

runtime:
  enabled: true
  onescience_domain: matchem
  min_onescience_version: null
  supported_execution:
    - local
    - local_slurm
    - remote_slurm
    - scnet_mcp

onescience:
  repo: https://gitee.com/onescience-ai/onescience
  official_links:
    gitee:
      doc: https://gitee.com/onescience-ai/onescience-doc
      onescience: https://gitee.com/onescience-ai/onescience
      skills: https://gitee.com/onescience-ai/oneskills
    github:
      doc: https://github.com/onescience-ai/OneScience-doc
      onescience: https://github.com/onescience-ai/OneScience
      skills: https://github.com/onescience-ai/oneskills
  install:
    required_by_default: false
    command: bash install.sh matchem
  source_paths:
    - matchem/uma/oc20
  compatibility:
    examples_path: examples/matchem/uma
    status: examples_compatible

runtime_package:
  kind: standard_runtime_package
  package_root: .
  standard_layout:
    workdir: .
    config_dir: metadata
    checkpoint_dir: null
    data_dir: data
    output_dir: outputs
  apply_policy:
    mode: direct_use
    target: session_workdir
    overwrite: true
    protect_installed_onescience: true
  entry_files:
    - scripts/validate_oc20.py
    - metadata/schema.yaml
    - metadata/sha256_manifest.txt
  entrypoints:
    preflight: scripts/validate_oc20.py
    inference: null
    train: null
    finetune: null
    evaluate: null
    visualize: null
    deploy: null

files:
  model_files: []
  dataset_files:
    - id: oc20_s2ef_200k_uncompressed
      path: data/oc20/s2ef_200k_uncompressed
      role: train_split
      description_zh: OC20 S2EF 200k 训练子集 extxyz 文件，当前包含 2 个分片、共 10000 帧。
      required: true
      required_for: [preflight, train]
      format: extxyz
      source:
        platform: modelscope
        repo_id: OneScience/oc20
        repo_type: dataset
        path: data/oc20/s2ef_200k_uncompressed
        revision: main
        url: null
        download_method: command_ref
        command_ref: commands.download.download_dataset_package
      mirrors: []
      local_path: session_workdir/data/oc20/s2ef_200k_uncompressed
      size: 84M
      checksum: metadata/sha256_manifest.txt
      large_file: true
    - id: oc20_s2ef_val_id_uncompressed
      path: data/oc20/s2ef_val_id_uncompressed
      role: validation_split
      description_zh: OC20 S2EF val_id 验证子集 extxyz 文件，当前包含 4 个分片、共 20000 帧。
      required: true
      required_for: [preflight, evaluate]
      format: extxyz
      source:
        platform: modelscope
        repo_id: OneScience/oc20
        repo_type: dataset
        path: data/oc20/s2ef_val_id_uncompressed
        revision: main
        url: null
        download_method: command_ref
        command_ref: commands.download.download_dataset_package
      mirrors: []
      local_path: session_workdir/data/oc20/s2ef_val_id_uncompressed
      size: 167M
      checksum: metadata/sha256_manifest.txt
      large_file: true
  config_files:
    - id: schema
      path: metadata/schema.yaml
      role: schema
      description_zh: OC20 extxyz 数据结构、字段名、shape、单位、split 和 UMA 配置键说明。
      required: true
      required_for: [preflight, train, evaluate]
      format: yaml
      source:
        platform: modelscope
        repo_id: OneScience/oc20
        repo_type: dataset
        path: metadata/schema.yaml
        revision: main
        url: null
        download_method: command_ref
        command_ref: commands.download.download_dataset_package
      mirrors: []
      local_path: session_workdir/metadata/schema.yaml
      size: null
      checksum: null
      large_file: false
    - id: sha256_manifest
      path: metadata/sha256_manifest.txt
      role: checksum
      description_zh: OC20 数据文件 SHA256 清单，用于校验整理后数据与原始数据一致。
      required: true
      required_for: [preflight]
      format: txt
      source:
        platform: modelscope
        repo_id: OneScience/oc20
        repo_type: dataset
        path: metadata/sha256_manifest.txt
        revision: main
        url: null
        download_method: command_ref
        command_ref: commands.download.download_dataset_package
      mirrors: []
      local_path: session_workdir/metadata/sha256_manifest.txt
      size: null
      checksum: null
      large_file: false
  sample_files:
    - id: validate_script
      path: scripts/validate_oc20.py
      role: script
      description_zh: OC20 数据读取验证脚本，检查 extxyz 字段、shape、帧数、元素集合和 SHA256。
      required: true
      required_for: [preflight]
      format: py
      source:
        platform: modelscope
        repo_id: OneScience/oc20
        repo_type: dataset
        path: scripts/validate_oc20.py
        revision: main
        url: null
        download_method: command_ref
        command_ref: commands.download.download_dataset_package
      mirrors: []
      local_path: session_workdir/scripts/validate_oc20.py
      size: null
      checksum: null
      large_file: false

dataset:
  format:
    - extxyz
  sample_unit: 单个催化剂表面吸附构型，包含元素、坐标、energy、free_energy 和 forces。
  schema:
    path: metadata/schema.yaml
  splits:
    train:
      path: data/oc20/s2ef_200k_uncompressed
      frames: 10000
    validation:
      path: data/oc20/s2ef_val_id_uncompressed
      frames: 20000
    test:
      path: null
      note_zh: "当前未提供独立测试集；如需测试可使用 s2ef_val_id_uncompressed 或从 OC20 官网下载完整 split。"

relations:
  required_datasets: []
  optional_datasets: []
  compatible_models:
    - id: OneScience/UMA
      role: finetune_data
      required_for: [preflight, train, evaluate]
      resource_ref:
        platform: modelscope
        repo_id: OneScience/UMA
        repo_type: model
        url: https://modelscope.cn/models/OneScience/UMA
        revision: main
        readme_path: README.md
        manifest_path: manifest.yaml
      expected_local_path: session_workdir/data/oc20
      adapter: null

run_matrix:
  scenarios:
    - name: validate_dataset
      default: true
      capability: preflight
      required_datasets:
        - id: OneScience/oc20
          role: train_data
          local_path: data/oc20
      required_model_files: []
      required_data_files:
        - data/oc20/s2ef_200k_uncompressed
        - data/oc20/s2ef_val_id_uncompressed
        - metadata/sha256_manifest.txt
      preconditions:
        - 已下载数据集仓库 OneScience/oc20。
      command_refs:
        - commands.preflight.validate_dataset
      outputs:
        - path: stdout
          description_zh: 数据字段、帧数、元素集合和 SHA256 检查结果。

capabilities:
  inference: false
  train: false
  finetune: false
  evaluate: false
  visualize: false
  deploy: false

commands:
  download:
    - name: download_dataset_package
      target: dataset
      repo_id: OneScience/oc20
      source: .
      local_path: session_workdir/
      required: true
      required_for: [preflight, train, evaluate]
      cwd: .
      command: modelscope download --dataset OneScience/oc20 --local_dir session_workdir
      expected_outputs:
        - path: session_workdir/data/oc20/s2ef_200k_uncompressed
          type: file
      success_criteria:
        - exit_code == 0
        - output_path_exists
  preflight:
    - name: validate_dataset
      description: 验证 OC20 extxyz 文件结构、字段名、帧数和 SHA256。
      cwd: .
      command: python scripts/validate_oc20.py --dataset-root data/oc20 --checksum-manifest metadata/sha256_manifest.txt
      required_files:
        - data/oc20/s2ef_200k_uncompressed
        - data/oc20/s2ef_val_id_uncompressed
        - metadata/sha256_manifest.txt
        - scripts/validate_oc20.py
      required_env:
        - python
      expected_outputs:
        - path: stdout
          type: text
      success_criteria:
        - exit_code == 0
  prepare: []
  inference: []
  train: []
  finetune: []
  evaluate: []
  visualize: []
  deploy: []

expected_outputs:
  - path: stdout
    type: text
    description_zh: 验证脚本输出每个 split、字段和 SHA256 的 OK 结果。

diagnostics:
  - error_pattern: "缺少文件"
    likely_cause: 数据集下载不完整或运行 cwd 不在数据集仓库根目录。
    action: 重新执行 modelscope download --dataset OneScience/oc20，并在下载后的数据集根目录运行验证命令。
  - error_pattern: "SHA256 不一致"
    likely_cause: 文件被截断、损坏或被替换。
    action: 删除损坏文件后重新下载数据集。
  - error_pattern: "energy"
    likely_cause: extxyz comment 中缺少 UMA OC20 配置要求的能量字段。
    action: 检查数据是否为本仓库 OC20 数据，或同步修改模型配置 energy_key。
  - error_pattern: "forces"
    likely_cause: extxyz Properties 中缺少 UMA OC20 配置要求的力字段。
    action: 检查数据是否为本仓库 OC20 数据，或同步修改模型配置 forces_key。

domain_extension:
  matchem:
    chemical_system:
      elements:
        - Ag
        - Al
        - As
        - Au
        - B
        - Bi
        - C
        - Ca
        - Cd
        - Cl
        - Co
        - Cr
        - Cs
        - Cu
        - Fe
        - Ga
        - Ge
        - H
        - Hf
        - Hg
        - In
        - Ir
        - K
        - Mn
        - Mo
        - N
        - Na
        - Nb
        - Ni
        - O
        - Os
        - P
        - Pb
        - Pd
        - Pt
        - Rb
        - Re
        - Rh
        - Ru
        - S
        - Sb
        - Sc
        - Se
        - Si
        - Sn
        - Sr
        - Ta
        - Tc
        - Te
        - Ti
        - Tl
        - V
        - W
        - Y
        - Zn
        - Zr
    properties:
      - energy
      - forces
    units:
      energy: eV
      forces: eV/Angstrom
    structure_format:
      - extxyz

large_resource:
  enabled: true
  total_size: 251M
  file_count: 6
  download_policy: full
  default_download: full
  minimal_files:
    - data/oc20/s2ef_200k_uncompressed
    - data/oc20/s2ef_val_id_uncompressed

reproducibility:
  level: minimal
  goal: 完成一次 OC20 extxyz 数据读取和 SHA256 校验。
  command_ref: commands.preflight.validate_dataset
  expected_runtime: 2-5 minutes
  expected_outputs:
    - stdout