onescience_manifest_version: "0.1"
resource_type: dataset

resource:
  id: OneScience/pdenneval
  name: pdenneval
  domain: cfd
  domain_tags: [cfd, pde, pdebench, hdf5]
  task: pdenneval_train_eval_dataset
  task_tags: [train_data, eval_data, dataset_validation]
  modalities: [gridded_field, pde_solution]
  input_formats: [hdf5]
  output_formats: [hdf5]
  summary: PDENNEval 配套 HDF5 数据集，包含 2D Darcy Flow、1D Burgers 和 1D Advection 数据文件。

platform_resource:
  primary:
    platform: modelscope
    repo_id: OneScience/pdenneval
    repo_type: dataset
    url: https://modelscope.cn/datasets/OneScience/pdenneval
    revision: main
    readme_path: README.md
    manifest_path: onescience_run_manifest.yaml
  mirrors: []
  access:
    visibility: public
    license: unknown

website_integration:
  enabled: true
  click_target:
    platform: modelscope
    resource_url: https://modelscope.cn/datasets/OneScience/pdenneval
  llm_handoff:
    readme_required: true
    manifest_required: true
    download_readme_first: true
    resolve_related_models: true
    default_run_goal: dataset_validation
    cwd_note: 如果使用 modelscope download --cache_dir 下载数据集，请先 cd 到实际下载后的数据集仓库根目录；模型侧将 ONESCIENCE_PDENNEVAL_DATA_DIR 指向本仓库 data 目录。

runtime:
  enabled: true
  onescience_domain: cfd
  min_onescience_version: null
  supported_execution: [local_dataset_validation, model_training_input, model_evaluation_input]
  environment:
    exported_env:
      ONESCIENCE_PDENNEVAL_DATA_DIR: <dataset_repo_root>/data
  dependencies:
    python: ">=3.10"
    python_packages: [h5py, numpy, pyyaml]

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 cfd
  source_paths:
    - onescience/src/onescience/datapipes/cfd/PDENNEval.py
  compatibility:
    examples_path: onescience/examples/cfd/PDENNEval
    status: examples_compatible
  datapipe: onescience.datapipes.cfd.PDENNEval

runtime_package:
  kind: standard_runtime_package
  package_root: .
  standard_layout:
    workdir: .
    data_dir: data
    metadata_dir: metadata
    output_dir: validation_outputs
  apply_policy:
    mode: direct_use
    target: session_workdir
    overwrite: false
    protect_installed_onescience: true
  entry_files:
    - data/2D_DarcyFlow_beta0.1_Train.hdf5
    - data/1D_Burgers_Sols_Nu0.001.hdf5
    - metadata/pdenneval_schema.json
    - scripts/validate_pdenneval_dataset.py
    - files_sha256.jsonl
  entrypoints:
    preflight: scripts/validate_pdenneval_dataset.py
    validate: scripts/validate_pdenneval_dataset.py
    inference: null
    train: null
    finetune: null
    evaluate: null
    visualize: null
    deploy: null

files:
  model_files: []
  dataset_files:
    - path: data/2D_DarcyFlow_beta0.1_Train.hdf5
      role: train_eval_data
      description_zh: 2D Darcy Flow HDF5 数据，包含 nu 系数场、tensor 解场和 x/y 坐标。
      required: true
      required_for: [dataset_validation, train, evaluate]
      source: {platform: modelscope, repo_id: OneScience/pdenneval, repo_type: dataset, download_method: command_ref, command_ref: commands.download.download_dataset}
      local_path: data/2D_DarcyFlow_beta0.1_Train.hdf5
      sha256: 74bbdf8b5db9cfd36168b003114673261337fd0422fa492ab1cb0c79f5d157e8
      size: 1310724488
    - path: data/1D_Burgers_Sols_Nu0.001.hdf5
      role: train_eval_data
      description_zh: 1D Burgers HDF5 数据，包含 tensor 解场、x 坐标和 t 坐标。
      required: true
      required_for: [dataset_validation, train, evaluate]
      source: {platform: modelscope, repo_id: OneScience/pdenneval, repo_type: dataset, download_method: command_ref, command_ref: commands.download.download_dataset}
      local_path: data/1D_Burgers_Sols_Nu0.001.hdf5
      sha256: afeeed1c40ce01d2ba5e1702f4f66b6bf95c65943eb0e4847691e9a10d1cb50d
      size: 8232968312
    - path: data/1D_Advection_Sols_beta1.0.hdf5
      role: supplemental_data
      description_zh: 1D Advection 可读 HDF5 数据，当前不作为默认模型运行场景。
      required: false
      required_for: [research_reference]
      source: {platform: modelscope, repo_id: OneScience/pdenneval, repo_type: dataset, download_method: command_ref, command_ref: commands.download.download_dataset}
      local_path: data/1D_Advection_Sols_beta1.0.hdf5
      sha256: 0ccd649b1d5ecca8a5ae417a506dec5ef57ea365f7bce46afa190d34fe02dcc7
      size: 4136968376
    - path: files_sha256.jsonl
      role: file_size_sha256_inventory
      description_zh: 标准化数据文件的大小和 SHA256 清单，用于完整性校验。
      required: true
      required_for: [dataset_validation]
      source: {platform: modelscope, repo_id: OneScience/pdenneval, repo_type: dataset, download_method: command_ref, command_ref: commands.download.download_dataset}
      local_path: files_sha256.jsonl
    - path: data_integrity_summary.json
      role: source_to_standardized_integrity_summary
      description_zh: 原始数据与标准化数据的文件名、大小、SHA256 对照摘要。
      required: true
      required_for: [dataset_validation]
      source: {platform: modelscope, repo_id: OneScience/pdenneval, repo_type: dataset, download_method: command_ref, command_ref: commands.download.download_dataset}
      local_path: data_integrity_summary.json
  config_files:
    - path: metadata/pdenneval_schema.json
      role: dataset_schema
      description_zh: 描述 PDENNEval HDF5 文件、字段、shape、dtype 和默认模型配置关系。
      required: true
      required_for: [dataset_validation, train, evaluate]
      source: {platform: modelscope, repo_id: OneScience/pdenneval, repo_type: dataset, download_method: command_ref, command_ref: commands.download.download_dataset}
      local_path: metadata/pdenneval_schema.json
  sample_files:
    - path: scripts/validate_pdenneval_dataset.py
      role: dataset_validation_entry
      description_zh: 数据集读取验证脚本，检查 HDF5 schema、可读性、文件大小和可选 SHA256。
      required: true
      required_for: [dataset_validation]
      source: {platform: modelscope, repo_id: OneScience/pdenneval, repo_type: dataset, download_method: command_ref, command_ref: commands.download.download_dataset}
      local_path: scripts/validate_pdenneval_dataset.py

dataset:
  format: [hdf5]
  sample_unit: 一个 PDE case 的时空网格解场或系数场。
  schema:
    path: metadata/pdenneval_schema.json
  splits:
    train: 由模型 datapipe 按配置内部切分。
    validation: 由模型 datapipe 按配置内部切分。
    test: null

relations:
  required_datasets: []
  optional_datasets: []
  compatible_models:
    - id: OneScience/PDENNEval
      role: train_data
      required_for: [preflight, train, evaluate]
      resource_ref:
        platform: modelscope
        repo_id: OneScience/PDENNEval
        repo_type: model
        url: https://modelscope.cn/models/OneScience/PDENNEval
        revision: main
        readme_path: README.md
        manifest_path: onescience_run_manifest.yaml
      expected_local_path: session_workdir/data
      expected_env: ONESCIENCE_PDENNEVAL_DATA_DIR

run_matrix:
  scenarios:
    - name: validate_hdf5_schema
      capability: dataset_validation
      default: true
      required_datasets:
        - id: OneScience/pdenneval
          role: train_data
          local_path: data
      required_model_files: []
      required_dataset_files:
        - data/2D_DarcyFlow_beta0.1_Train.hdf5
        - data/1D_Burgers_Sols_Nu0.001.hdf5
        - metadata/pdenneval_schema.json
      preconditions: [当前工作目录为数据集仓库根目录。]
      command_refs: [commands.preflight.validate_dataset]
      outputs:
        - "控制台输出 [OK] dataset validation completed"
    - name: validate_integrity_full_hash
      capability: dataset_validation
      default: false
      required_datasets:
        - id: OneScience/pdenneval
          role: train_data
          local_path: data
      required_model_files: []
      required_dataset_files: [files_sha256.jsonl, data/*.hdf5]
      preconditions: [当前工作目录为数据集仓库根目录。]
      command_refs: [commands.preflight.validate_dataset_full_hash]
      outputs:
        - "控制台输出 checksum manifest verified in size+sha256 mode"
    - name: provide_to_pdenneval_model
      capability: train
      default: true
      required_datasets:
        - id: OneScience/pdenneval
          role: train_data
          local_path: data
      required_model_files: [OneScience/PDENNEval:conf/fno_2d_darcy.yaml]
      required_dataset_files:
        - data/2D_DarcyFlow_beta0.1_Train.hdf5
        - data/1D_Burgers_Sols_Nu0.001.hdf5
      preconditions:
        - 模型仓库 OneScience/PDENNEval 已下载。
        - 设置 ONESCIENCE_PDENNEVAL_DATA_DIR 为本仓库 data 目录。
      command_refs: [commands.prepare.export_dataset_env]
      outputs:
        - 模型侧 scripts/preflight_check.py 可以读取数据集。

capabilities:
  dataset_validation: true
  train_input: true
  evaluation_input: true
  inference_input: false
  preflight: true
  inference: false
  train: false
  finetune: false
  evaluate: false
  visualize: false
  deploy: false

commands:
  download:
    - name: download_dataset
      command: modelscope download --dataset OneScience/pdenneval
      cwd: session_workdir
    - name: download_model
      command: modelscope download --model OneScience/PDENNEval
      cwd: session_workdir
  prepare:
    - name: export_dataset_env
      command: export ONESCIENCE_PDENNEVAL_DATA_DIR=/path/to/OneScience/pdenneval/data
      cwd: .
  preflight:
    - name: validate_dataset
      command: python scripts/validate_pdenneval_dataset.py
      cwd: .
    - name: validate_dataset_full_hash
      command: python scripts/validate_pdenneval_dataset.py --full-hash
      cwd: .
  inference: []
  train: []
  finetune: []
  evaluate: []
  visualize: []
  deploy: []

expected_outputs:
  - path: validation_outputs/
    type: directory
    description_zh: 预留的数据读取验证输出目录；默认验证命令主要输出控制台 OK/FAIL。

diagnostics:
  - name: missing_hdf5
    match: "missing expected HDF5 file"
    severity: error
    fix_zh: 确认已完整下载 OneScience/pdenneval，并在数据集仓库根目录执行验证脚本。
  - name: hdf5_schema_mismatch
    match: "missing dataset"
    severity: error
    fix_zh: 数据文件结构与 Manifest 不一致，重新下载或检查文件是否被替换。
  - name: checksum_mismatch
    match: "sha256 mismatch"
    severity: error
    fix_zh: 文件内容与上传清单不一致，重新下载数据集。
  - name: dependency_missing
    match: "ModuleNotFoundError"
    severity: error
    fix_zh: 安装 h5py、numpy 和 PyYAML。

domain_extension:
  cfd:
    equation_families: [darcy_flow, burgers, advection]
    benchmark_family: PDEBench
    data_format: HDF5
    default_model: OneScience/PDENNEval
    notes:
      - 1D_Advection_Sols_beta1.0.hdf5 随包上传但不作为默认运行场景。