README.md

Data Distribution

EC

• train 15472
• test 1906
• valid 1719

GO

• train 28484
• valid 3177
• test 3153

EC Attributes

{
  "all_categories": {
    "ec_number": [...], # 包含所有可能的label
  },
  # labels里的每个字典表示一条样本，对应的feature和它的labels
  "labels": [ 
    {
      "id": "1RK6-A",
      "ec_number": [
        "3.5.1.-"
      ],     # 这条样本的标签
      "split": "train", # 这条样本属于哪个划分
      "feat_name": "1RK6-A_15801.safetensors"   # 特征名字
    },
    {
      "id": "1YRV-A",
      "ec_number": [
        "2.3.2.23",
        "2.3.2.-"
      ],    # 一个样本可能有多个标签
      "split": "test",
      "feat_name": "1YRV-A_15802.safetensors" # 对应feature下面的safetensors的名字
    },
  ]
}

GO Attributes

{
  "all_categories": {
    "molecular_function": [...], # 包含所有可能的molecular_function的label
    "biological_process": [...], # 包含所有可能的biological_process的label
    "cellular_component": [...], # 包含所有可能的cellular_component的label
  },
  "labels": [
    {
      "id": "11AS-A",
      "molecular_function": ["GO:0097367","GO:0042802",...],
      "biological_process": ["GO:0016053",],
      "cellular_component": ["GO:0005829"],
      "split": "test",
      "feat_name": "11AS-A_0.safetensors"
    },
    {
      "id": "154L-A",
      "molecular_function": ["GO:0004553","GO:0003796","GO:0061783","GO:0016798"],
      "biological_process": ["GO:0006950","GO:0042742","GO:0044248","GO:0006022",...],
      "cellular_component": [],  # 有的labels是空的
      "split": "train",
      "feat_name": "154L-A_1.safetensors" # 对应feature下面的safetensors的名字
    },
  ]
}