{
  "configuration": {
    "train_jsonl": "data\\enhanced\\train.jsonl",
    "output_dir": "models\\codet5-java-to-python",
    "val_jsonl": "data\\enhanced\\validation.jsonl",
    "model_name": "Salesforce/codet5-small",
    "direction": "java_to_python",
    "eval_ratio": 0.1,
    "max_input_length": 512,
    "max_target_length": 512,
    "batch_size": 4,
    "epochs": 8,
    "lr": 2e-05,
    "gradient_accumulation_steps": 2,
    "seed": 42,
    "early_stopping_patience": 2
  },
  "base_model": "Salesforce/codet5-small",
  "device": "cpu",
  "cuda_available": false,
  "mixed_precision_fp16": false,
  "pytorch_version": "2.13.0+cpu",
  "training_examples": 1000,
  "validation_examples": 140,
  "token_length_statistics": {
    "train": {
      "inputs": {
        "minimum": 73,
        "median": 107,
        "p95": 149,
        "maximum": 161,
        "limit": 512,
        "truncated_examples": 0
      },
      "targets": {
        "minimum": 23,
        "median": 49,
        "p95": 87,
        "maximum": 97,
        "limit": 512,
        "truncated_examples": 0
      }
    },
    "validation": {
      "inputs": {
        "minimum": 73,
        "median": 107,
        "p95": 149,
        "maximum": 149,
        "limit": 512,
        "truncated_examples": 0
      },
      "targets": {
        "minimum": 23,
        "median": 49,
        "p95": 87,
        "maximum": 87,
        "limit": 512,
        "truncated_examples": 0
      }
    }
  },
  "duration_seconds": 4517.891,
  "best_checkpoint": "models\\codet5-java-to-python\\checkpoint-1000",
  "best_metric": 0.0002744992380030453,
  "train_metrics": {
    "train_runtime": 4509.2089,
    "train_samples_per_second": 1.774,
    "train_steps_per_second": 0.222,
    "total_flos": 268101535137792.0,
    "train_loss": 0.11644796460866928,
    "epoch": 8.0
  },
  "validation_metrics": {
    "eval_loss": 0.0002744992380030453,
    "eval_runtime": 8.4931,
    "eval_samples_per_second": 16.484,
    "eval_steps_per_second": 4.121,
    "epoch": 8.0
  },
  "log_history": [
    {
      "loss": 0.8178,
      "grad_norm": 1.9174737930297852,
      "learning_rate": 1.752e-05,
      "epoch": 1.0,
      "step": 125
    },
    {
      "eval_loss": 0.027854034677147865,
      "eval_runtime": 9.3606,
      "eval_samples_per_second": 14.956,
      "eval_steps_per_second": 3.739,
      "epoch": 1.0,
      "step": 125
    },
    {
      "loss": 0.0566,
      "grad_norm": 1.2266449928283691,
      "learning_rate": 1.5020000000000002e-05,
      "epoch": 2.0,
      "step": 250
    },
    {
      "eval_loss": 0.0061431629583239555,
      "eval_runtime": 12.3133,
      "eval_samples_per_second": 11.37,
      "eval_steps_per_second": 2.842,
      "epoch": 2.0,
      "step": 250
    },
    {
      "loss": 0.0202,
      "grad_norm": 0.3443121016025543,
      "learning_rate": 1.252e-05,
      "epoch": 3.0,
      "step": 375
    },
    {
      "eval_loss": 0.0017448368016630411,
      "eval_runtime": 12.4117,
      "eval_samples_per_second": 11.28,
      "eval_steps_per_second": 2.82,
      "epoch": 3.0,
      "step": 375
    },
    {
      "loss": 0.0117,
      "grad_norm": 0.22224248945713043,
      "learning_rate": 1.002e-05,
      "epoch": 4.0,
      "step": 500
    },
    {
      "eval_loss": 0.0007071287254802883,
      "eval_runtime": 16.8838,
      "eval_samples_per_second": 8.292,
      "eval_steps_per_second": 2.073,
      "epoch": 4.0,
      "step": 500
    },
    {
      "loss": 0.0082,
      "grad_norm": 0.07997278869152069,
      "learning_rate": 7.520000000000001e-06,
      "epoch": 5.0,
      "step": 625
    },
    {
      "eval_loss": 0.00043417332926765084,
      "eval_runtime": 11.6114,
      "eval_samples_per_second": 12.057,
      "eval_steps_per_second": 3.014,
      "epoch": 5.0,
      "step": 625
    },
    {
      "loss": 0.0064,
      "grad_norm": 0.06534432619810104,
      "learning_rate": 5.02e-06,
      "epoch": 6.0,
      "step": 750
    },
    {
      "eval_loss": 0.0003266753919888288,
      "eval_runtime": 7.8595,
      "eval_samples_per_second": 17.813,
      "eval_steps_per_second": 4.453,
      "epoch": 6.0,
      "step": 750
    },
    {
      "loss": 0.0053,
      "grad_norm": 0.22041985392570496,
      "learning_rate": 2.52e-06,
      "epoch": 7.0,
      "step": 875
    },
    {
      "eval_loss": 0.0002930415503215045,
      "eval_runtime": 8.6639,
      "eval_samples_per_second": 16.159,
      "eval_steps_per_second": 4.04,
      "epoch": 7.0,
      "step": 875
    },
    {
      "loss": 0.0053,
      "grad_norm": 0.8123140931129456,
      "learning_rate": 2e-08,
      "epoch": 8.0,
      "step": 1000
    },
    {
      "eval_loss": 0.0002744992380030453,
      "eval_runtime": 7.9757,
      "eval_samples_per_second": 17.553,
      "eval_steps_per_second": 4.388,
      "epoch": 8.0,
      "step": 1000
    },
    {
      "train_runtime": 4509.2089,
      "train_samples_per_second": 1.774,
      "train_steps_per_second": 0.222,
      "total_flos": 268101535137792.0,
      "train_loss": 0.11644796460866928,
      "epoch": 8.0,
      "step": 1000
    },
    {
      "eval_loss": 0.0002744992380030453,
      "eval_runtime": 8.4931,
      "eval_samples_per_second": 16.484,
      "eval_steps_per_second": 4.121,
      "epoch": 8.0,
      "step": 1000
    }
  ]
}