last-checkpoint/trainer_state.json

{
  "best_metric": 0.3915156871409633,
  "best_model_checkpoint": "/mimer/NOBACKUP/groups/naiss2023-6-290/stefano/models//PROTAC-Splitter-EncoderDecoder-lr_reduce-opt25/trial-number=8-learning_rate=2.0e-05-warmup_ratio=0.020-min_lr=0.000-factor=0.820/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": 1.0092626810073853,
      "learning_rate": 1.9500595004156286e-05,
      "loss": 1.3581,
      "step": 500
    },
    {
      "epoch": 0.09863878477017163,
      "grad_norm": 0.6695255637168884,
      "learning_rate": 1.9500595004156286e-05,
      "loss": 0.3853,
      "step": 1000
    },
    {
      "epoch": 0.14795817715525744,
      "grad_norm": 0.5797414183616638,
      "learning_rate": 1.9500595004156286e-05,
      "loss": 0.2014,
      "step": 1500
    },
    {
      "epoch": 0.19727756954034326,
      "grad_norm": 0.49628201127052307,
      "learning_rate": 1.9500595004156286e-05,
      "loss": 0.1258,
      "step": 2000
    },
    {
      "epoch": 0.24659696192542907,
      "grad_norm": 0.3971586525440216,
      "learning_rate": 1.9500595004156286e-05,
      "loss": 0.0839,
      "step": 2500
    },
    {
      "epoch": 0.2959163543105149,
      "grad_norm": 0.30195024609565735,
      "learning_rate": 1.9500595004156286e-05,
      "loss": 0.0596,
      "step": 3000
    },
    {
      "epoch": 0.3452357466956007,
      "grad_norm": 0.2556217610836029,
      "learning_rate": 1.9500595004156286e-05,
      "loss": 0.0446,
      "step": 3500
    },
    {
      "epoch": 0.3945551390806865,
      "grad_norm": 0.2777057886123657,
      "learning_rate": 1.9500595004156286e-05,
      "loss": 0.0349,
      "step": 4000
    },
    {
      "epoch": 0.4438745314657723,
      "grad_norm": 0.246878519654274,
      "learning_rate": 1.9500595004156286e-05,
      "loss": 0.028,
      "step": 4500
    },
    {
      "epoch": 0.49319392385085814,
      "grad_norm": 0.27177679538726807,
      "learning_rate": 1.9500595004156286e-05,
      "loss": 0.0231,
      "step": 5000
    },
    {
      "epoch": 0.49319392385085814,
      "eval_all_ligands_equal": 0.3915156871409633,
      "eval_e3_equal": 0.7765797613787009,
      "eval_e3_graph_edit_distance": 9.999999999999999e+63,
      "eval_e3_graph_edit_distance_norm": 1.0,
      "eval_e3_has_attachment_point(s)": 0.9932832523199293,
      "eval_e3_heavy_atoms_difference": 0.2896155545735749,
      "eval_e3_heavy_atoms_difference_norm": 0.0020662380654388832,
      "eval_e3_tanimoto_similarity": 0.0,
      "eval_e3_valid": 0.9932832523199293,
      "eval_has_all_attachment_points": 0.9840919133893062,
      "eval_has_three_substructures": 0.9990278391515687,
      "eval_heavy_atoms_difference": 8.075386654882898,
      "eval_heavy_atoms_difference_norm": 0.10478609381959636,
      "eval_linker_equal": 0.563323022536456,
      "eval_linker_graph_edit_distance": 9.999999999999999e+63,
      "eval_linker_graph_edit_distance_norm": 1.0,
      "eval_linker_has_attachment_point(s)": 0.9960229783473266,
      "eval_linker_heavy_atoms_difference": 0.9376049491825011,
      "eval_linker_heavy_atoms_difference_norm": 0.02394921771244689,
      "eval_linker_tanimoto_similarity": 0.0,
      "eval_linker_valid": 0.9960229783473266,
      "eval_loss": 0.286180704832077,
      "eval_num_fragments": 2.9999116217410515,
      "eval_poi_equal": 0.7011047282368538,
      "eval_poi_graph_edit_distance": 9.999999999999999e+63,
      "eval_poi_graph_edit_distance_norm": 1.0,
      "eval_poi_has_attachment_point(s)": 0.9188687582854618,
      "eval_poi_heavy_atoms_difference": 2.6504639858594787,
      "eval_poi_heavy_atoms_difference_norm": 0.0799410695670892,
      "eval_poi_tanimoto_similarity": 0.0,
      "eval_poi_valid": 0.9188687582854618,
      "eval_reassembly": 0.39752540874944764,
      "eval_reassembly_nostereo": 0.41688024745912505,
      "eval_runtime": 1468.607,
      "eval_samples_per_second": 7.705,
      "eval_steps_per_second": 0.121,
      "eval_tanimoto_similarity": 0.0,
      "eval_valid": 0.9114449845338047,
      "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=8-learning_rate=2.0e-05-warmup_ratio=0.020-min_lr=0.000-factor=0.820",
  "trial_params": {
    "factor": 0.8200000000000001,
    "learning_rate": 1.9500595004156286e-05,
    "min_lr": 1.163109026324453e-09,
    "warmup_ratio": 0.02
  }
}