{
  "dataloader": {
    "kwargs": {
      "batch_size": 1024
    },
    "type": "pytorch_dataloader"
  },
  "dataset": {
    "kwargs": {
      "feature_columns": [],
      "label_columns": [
        "label1"
      ],
      "mask": {
        "kwargs": {},
        "type": "pad_mask"
      }
    },
    "type": "TransformerDatasetWithDict"
  },
  "loss": {
    "kwargs": {},
    "type": "CorrLoss"
  },
  "max_epochs": 30,
  "min_epochs": 10,
  "model": {
    "kwargs": {
      "embedding_config": {
        "kwargs": {
          "dropout": 0.2,
          "fixed": false,
          "inlinear": true
        },
        "type": "Exp2dEmbedding"
      },
      "embedding_size": 256,
      "encoder_config": {
        "kwargs": {
          "activation": "gelu",
          "d_ff": 256,
          "d_k": 32,
          "d_v": 32,
          "layer_num": 4,
          "n_heads": 32,
          "res_dropout": 0.1
        },
        "type": "TSTPlusEncoder"
      },
      "feature_size": 256,
      "head_config": {
        "kwargs": {
          "dropout": 0.5,
          "hidden_size": 256,
          "num_layers": 1
        },
        "type": "FCHead"
      },
      "input_size": 0,
      "pooling_config": {
        "kwargs": {},
        "type": "LastPooling"
      },
      "y_len": 1
    },
    "type": "TransformerModel"
  },
  "model_name": "TransformerTest",
  "monitor": "val_loss",
  "optimizer": {
    "kwargs": {
      "lr": 5e-05,
      "weight_decay": 1e-06
    },
    "type": "Adam"
  },
  "patience": 10,
  "pp": "test_0",
  "scheduler": {
    "kwargs": {
      "gamma": 0.5,
      "milestones": [
        10,
        15
      ]
    },
    "type": "MultiStepLR"
  },
  "seq_length": 10,
  "test_dataset": "toy_dataset/test.csv",
  "train_dataset": "toy_dataset/train.csv",
  "transformers_version": "4.10.0.dev0",
  "use_cache": true,
  "valid_dataset": "toy_dataset/valid.csv"
}