last-checkpoint/trainer_state.json

{
  "best_metric": 0.014670790985417587,
  "best_model_checkpoint": "/mimer/NOBACKUP/groups/naiss2023-6-290/stefano/models//PROTAC-Splitter-EncoderDecoder-lr_reduce-opt25/trial-number=12-learning_rate=7.4e-06-warmup_ratio=0.030-min_lr=0.000-factor=0.930/checkpoint-5000",
  "epoch": 0.49319392385085814,
  "eval_steps": 2500,
  "global_step": 5000,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.049319392385085814,
      "grad_norm": 1.1744908094406128,
      "learning_rate": 7.392845746133611e-06,
      "loss": 1.9933,
      "step": 500
    },
    {
      "epoch": 0.09863878477017163,
      "grad_norm": 1.0896804332733154,
      "learning_rate": 7.392845746133611e-06,
      "loss": 0.7405,
      "step": 1000
    },
    {
      "epoch": 0.14795817715525744,
      "grad_norm": 0.9823786616325378,
      "learning_rate": 7.392845746133611e-06,
      "loss": 0.492,
      "step": 1500
    },
    {
      "epoch": 0.19727756954034326,
      "grad_norm": 0.9321771264076233,
      "learning_rate": 7.392845746133611e-06,
      "loss": 0.3598,
      "step": 2000
    },
    {
      "epoch": 0.24659696192542907,
      "grad_norm": 0.9124600887298584,
      "learning_rate": 7.392845746133611e-06,
      "loss": 0.2774,
      "step": 2500
    },
    {
      "epoch": 0.2959163543105149,
      "grad_norm": 0.7268500328063965,
      "learning_rate": 7.392845746133611e-06,
      "loss": 0.2236,
      "step": 3000
    },
    {
      "epoch": 0.3452357466956007,
      "grad_norm": 0.4115571081638336,
      "learning_rate": 7.392845746133611e-06,
      "loss": 0.185,
      "step": 3500
    },
    {
      "epoch": 0.3945551390806865,
      "grad_norm": 0.5712406039237976,
      "learning_rate": 7.392845746133611e-06,
      "loss": 0.1587,
      "step": 4000
    },
    {
      "epoch": 0.4438745314657723,
      "grad_norm": 0.49448713660240173,
      "learning_rate": 7.392845746133611e-06,
      "loss": 0.1374,
      "step": 4500
    },
    {
      "epoch": 0.49319392385085814,
      "grad_norm": 0.4770694375038147,
      "learning_rate": 7.392845746133611e-06,
      "loss": 0.1204,
      "step": 5000
    },
    {
      "epoch": 0.49319392385085814,
      "eval_all_ligands_equal": 0.014670790985417587,
      "eval_e3_equal": 0.4694653115333628,
      "eval_e3_graph_edit_distance": 9.999999999999999e+63,
      "eval_e3_graph_edit_distance_norm": 1.0,
      "eval_e3_has_attachment_point(s)": 0.9965532479010164,
      "eval_e3_heavy_atoms_difference": -0.54909412284578,
      "eval_e3_heavy_atoms_difference_norm": -0.03696717892690007,
      "eval_e3_tanimoto_similarity": 0.0,
      "eval_e3_valid": 0.9965532479010164,
      "eval_has_all_attachment_points": 0.9708351745470615,
      "eval_has_three_substructures": 0.997348652231551,
      "eval_heavy_atoms_difference": 13.469553689792312,
      "eval_heavy_atoms_difference_norm": 0.17544293470078143,
      "eval_linker_equal": 0.03588157313300928,
      "eval_linker_graph_edit_distance": 9.999999999999999e+63,
      "eval_linker_graph_edit_distance_norm": 1.0,
      "eval_linker_has_attachment_point(s)": 0.9908086610693769,
      "eval_linker_heavy_atoms_difference": 3.279628811312417,
      "eval_linker_heavy_atoms_difference_norm": 0.12970442523287076,
      "eval_linker_tanimoto_similarity": 0.0,
      "eval_linker_valid": 0.9908086610693769,
      "eval_loss": 0.3284246027469635,
      "eval_num_fragments": 2.9978789217852406,
      "eval_poi_equal": 0.46981882456915597,
      "eval_poi_graph_edit_distance": 9.999999999999999e+63,
      "eval_poi_graph_edit_distance_norm": 1.0,
      "eval_poi_has_attachment_point(s)": 0.8617764030048608,
      "eval_poi_heavy_atoms_difference": 4.413610251878038,
      "eval_poi_heavy_atoms_difference_norm": 0.1345314085241158,
      "eval_poi_tanimoto_similarity": 0.0,
      "eval_poi_valid": 0.8617764030048608,
      "eval_reassembly": 0.014935925762262483,
      "eval_reassembly_nostereo": 0.01864781263809103,
      "eval_runtime": 1510.5239,
      "eval_samples_per_second": 7.491,
      "eval_steps_per_second": 0.117,
      "eval_tanimoto_similarity": 0.0,
      "eval_valid": 0.8566504639858595,
      "step": 5000
    }
  ],
  "logging_steps": 500,
  "max_steps": 10000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.33139477262336e+16,
  "train_batch_size": 128,
  "trial_name": "trial-number=12-learning_rate=7.4e-06-warmup_ratio=0.030-min_lr=0.000-factor=0.930",
  "trial_params": {
    "factor": 0.93,
    "learning_rate": 7.392845746133611e-06,
    "min_lr": 9.986795785570395e-09,
    "warmup_ratio": 0.03
  }
}