last-checkpoint/trainer_state.json

{
  "best_metric": 0.20972160848431287,
  "best_model_checkpoint": "/mimer/NOBACKUP/groups/naiss2023-6-290/stefano/models//PROTAC-Splitter-EncoderDecoder-lr_reduce-opt25/trial-number=9-learning_rate=1.3e-05-warmup_ratio=0.050-min_lr=0.000-factor=0.890/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": 1.0882785320281982,
      "learning_rate": 1.287188117510188e-05,
      "loss": 1.6004,
      "step": 500
    },
    {
      "epoch": 0.09863878477017163,
      "grad_norm": 0.7874086499214172,
      "learning_rate": 1.287188117510188e-05,
      "loss": 0.5219,
      "step": 1000
    },
    {
      "epoch": 0.14795817715525744,
      "grad_norm": 0.7524672150611877,
      "learning_rate": 1.287188117510188e-05,
      "loss": 0.3065,
      "step": 1500
    },
    {
      "epoch": 0.19727756954034326,
      "grad_norm": 0.5783708691596985,
      "learning_rate": 1.287188117510188e-05,
      "loss": 0.2066,
      "step": 2000
    },
    {
      "epoch": 0.24659696192542907,
      "grad_norm": 0.669140100479126,
      "learning_rate": 1.287188117510188e-05,
      "loss": 0.151,
      "step": 2500
    },
    {
      "epoch": 0.2959163543105149,
      "grad_norm": 0.4841144382953644,
      "learning_rate": 1.287188117510188e-05,
      "loss": 0.1148,
      "step": 3000
    },
    {
      "epoch": 0.3452357466956007,
      "grad_norm": 0.34032905101776123,
      "learning_rate": 1.287188117510188e-05,
      "loss": 0.0882,
      "step": 3500
    },
    {
      "epoch": 0.3945551390806865,
      "grad_norm": 0.32155144214630127,
      "learning_rate": 1.287188117510188e-05,
      "loss": 0.0698,
      "step": 4000
    },
    {
      "epoch": 0.4438745314657723,
      "grad_norm": 0.38137614727020264,
      "learning_rate": 1.287188117510188e-05,
      "loss": 0.056,
      "step": 4500
    },
    {
      "epoch": 0.49319392385085814,
      "grad_norm": 0.39012080430984497,
      "learning_rate": 1.287188117510188e-05,
      "loss": 0.0469,
      "step": 5000
    },
    {
      "epoch": 0.49319392385085814,
      "eval_all_ligands_equal": 0.20972160848431287,
      "eval_e3_equal": 0.7072028281042864,
      "eval_e3_graph_edit_distance": 9.999999999999999e+63,
      "eval_e3_graph_edit_distance_norm": 1.0,
      "eval_e3_has_attachment_point(s)": 0.9947856827220504,
      "eval_e3_heavy_atoms_difference": 0.09491825011047282,
      "eval_e3_heavy_atoms_difference_norm": -0.006944252067541642,
      "eval_e3_tanimoto_similarity": 0.0,
      "eval_e3_valid": 0.9947856827220504,
      "eval_has_all_attachment_points": 0.9820592134334953,
      "eval_has_three_substructures": 0.9984975695978789,
      "eval_heavy_atoms_difference": 9.720901458241272,
      "eval_heavy_atoms_difference_norm": 0.12639386628820087,
      "eval_linker_equal": 0.33548387096774196,
      "eval_linker_graph_edit_distance": 9.999999999999999e+63,
      "eval_linker_graph_edit_distance_norm": 1.0,
      "eval_linker_has_attachment_point(s)": 0.9950508174988952,
      "eval_linker_heavy_atoms_difference": 1.5590808661069377,
      "eval_linker_heavy_atoms_difference_norm": 0.05308081409891352,
      "eval_linker_tanimoto_similarity": 0.0,
      "eval_linker_valid": 0.9950508174988952,
      "eval_loss": 0.2932600677013397,
      "eval_num_fragments": 2.999381352187362,
      "eval_poi_equal": 0.636411842686699,
      "eval_poi_graph_edit_distance": 9.999999999999999e+63,
      "eval_poi_graph_edit_distance_norm": 1.0,
      "eval_poi_has_attachment_point(s)": 0.9011047282368537,
      "eval_poi_heavy_atoms_difference": 3.2722934158197083,
      "eval_poi_heavy_atoms_difference_norm": 0.1006642870975325,
      "eval_poi_tanimoto_similarity": 0.0,
      "eval_poi_valid": 0.9011047282368537,
      "eval_reassembly": 0.2139637649138312,
      "eval_reassembly_nostereo": 0.23932832523199293,
      "eval_runtime": 1512.9908,
      "eval_samples_per_second": 7.479,
      "eval_steps_per_second": 0.117,
      "eval_tanimoto_similarity": 0.0,
      "eval_valid": 0.8955368979231109,
      "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=9-learning_rate=1.3e-05-warmup_ratio=0.050-min_lr=0.000-factor=0.890",
  "trial_params": {
    "factor": 0.89,
    "learning_rate": 1.287188117510188e-05,
    "min_lr": 7.953202892241938e-10,
    "warmup_ratio": 0.05
  }
}