metrics.json

{
  "history": [
    {
      "epoch": 1,
      "train_sum_loss": 6.272962152862549,
      "train_class_loss": 1.0185239918708802,
      "train_class_acc": 0.4368,
      "val_sum_loss": 5.8008860549926755,
      "val_class_loss": 1.1023406219482421,
      "val_class_acc": 0.304,
      "val_rouge1": 0.17660690248954364,
      "val_rouge2": 0.018371370658335634,
      "val_rougeL": 0.15689190075772386
    },
    {
      "epoch": 2,
      "train_sum_loss": 5.219159950256348,
      "train_class_loss": 1.099994719696045,
      "train_class_acc": 0.3462,
      "val_sum_loss": 5.416218318939209,
      "val_class_loss": 1.0967657680511476,
      "val_class_acc": 0.386,
      "val_rouge1": 0.21779871449217764,
      "val_rouge2": 0.028536134623968267,
      "val_rougeL": 0.19329056631523484
    },
    {
      "epoch": 3,
      "train_sum_loss": 4.404316672515869,
      "train_class_loss": 1.0899620735168456,
      "train_class_acc": 0.361,
      "val_sum_loss": 5.30168932723999,
      "val_class_loss": 1.7727555074691772,
      "val_class_acc": 0.342,
      "val_rouge1": 0.22501668712961082,
      "val_rouge2": 0.03326989668183761,
      "val_rougeL": 0.20778061653444058
    }
  ],
  "config": {
    "tokenizer": "t5-small",
    "max_input_len": 512,
    "max_target_len": 64,
    "hidden_dim": 512,
    "num_experts": 6,
    "top_k": 2,
    "batch_size": 8,
    "learning_rate": 0.0005,
    "num_epochs": 3,
    "seed": 42,
    "device": "cuda",
    "hf_repo": "Deepu1965/bonus2-multitask-moe"
  }
}