last-checkpoint/trainer_state.json

{
  "best_metric": 0.517454706142289,
  "best_model_checkpoint": "/mimer/NOBACKUP/groups/naiss2023-6-290/stefano/models//PROTAC-Splitter-EncoderDecoder-lr_reduce-opt25/trial-number=23-learning_rate=5.8e-05-warmup_ratio=0.050-min_lr=0.000-factor=0.800/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": 0.6019228100776672,
      "learning_rate": 5.8123802212920094e-05,
      "loss": 0.9116,
      "step": 500
    },
    {
      "epoch": 0.09863878477017163,
      "grad_norm": 0.4120740592479706,
      "learning_rate": 5.8123802212920094e-05,
      "loss": 0.1513,
      "step": 1000
    },
    {
      "epoch": 0.14795817715525744,
      "grad_norm": 0.29680919647216797,
      "learning_rate": 5.8123802212920094e-05,
      "loss": 0.0565,
      "step": 1500
    },
    {
      "epoch": 0.19727756954034326,
      "grad_norm": 0.22217430174350739,
      "learning_rate": 5.8123802212920094e-05,
      "loss": 0.0309,
      "step": 2000
    },
    {
      "epoch": 0.24659696192542907,
      "grad_norm": 0.1675247699022293,
      "learning_rate": 5.8123802212920094e-05,
      "loss": 0.0195,
      "step": 2500
    },
    {
      "epoch": 0.2959163543105149,
      "grad_norm": 0.19688111543655396,
      "learning_rate": 5.8123802212920094e-05,
      "loss": 0.014,
      "step": 3000
    },
    {
      "epoch": 0.3452357466956007,
      "grad_norm": 0.17143021523952484,
      "learning_rate": 5.8123802212920094e-05,
      "loss": 0.011,
      "step": 3500
    },
    {
      "epoch": 0.3945551390806865,
      "grad_norm": 0.12186994403600693,
      "learning_rate": 5.8123802212920094e-05,
      "loss": 0.0089,
      "step": 4000
    },
    {
      "epoch": 0.4438745314657723,
      "grad_norm": 0.14182540774345398,
      "learning_rate": 5.8123802212920094e-05,
      "loss": 0.0072,
      "step": 4500
    },
    {
      "epoch": 0.49319392385085814,
      "grad_norm": 0.16419431567192078,
      "learning_rate": 5.8123802212920094e-05,
      "loss": 0.0068,
      "step": 5000
    },
    {
      "epoch": 0.49319392385085814,
      "eval_all_ligands_equal": 0.517454706142289,
      "eval_e3_equal": 0.798232434821034,
      "eval_e3_graph_edit_distance": 9.999999999999999e+63,
      "eval_e3_graph_edit_distance_norm": 1.0,
      "eval_e3_has_attachment_point(s)": 0.9901016349977906,
      "eval_e3_heavy_atoms_difference": 0.5490057445868316,
      "eval_e3_heavy_atoms_difference_norm": 0.011477999514524074,
      "eval_e3_tanimoto_similarity": 0.0,
      "eval_e3_valid": 0.9901016349977906,
      "eval_has_all_attachment_points": 0.9792311091471498,
      "eval_has_three_substructures": 0.9985859478568272,
      "eval_heavy_atoms_difference": 8.138930623066726,
      "eval_heavy_atoms_difference_norm": 0.10671091458695267,
      "eval_linker_equal": 0.747326557666814,
      "eval_linker_graph_edit_distance": 9.999999999999999e+63,
      "eval_linker_graph_edit_distance_norm": 1.0,
      "eval_linker_has_attachment_point(s)": 0.994962439239947,
      "eval_linker_heavy_atoms_difference": 0.4178524083075563,
      "eval_linker_heavy_atoms_difference_norm": 0.004257830368338666,
      "eval_linker_tanimoto_similarity": 0.0,
      "eval_linker_valid": 0.994962439239947,
      "eval_loss": 0.30216675996780396,
      "eval_num_fragments": 3.000353513035793,
      "eval_poi_equal": 0.7433495360141406,
      "eval_poi_graph_edit_distance": 9.999999999999999e+63,
      "eval_poi_graph_edit_distance_norm": 1.0,
      "eval_poi_has_attachment_point(s)": 0.9159522757401679,
      "eval_poi_heavy_atoms_difference": 2.62076889085285,
      "eval_poi_heavy_atoms_difference_norm": 0.08558461375931094,
      "eval_poi_tanimoto_similarity": 0.0,
      "eval_poi_valid": 0.9159522757401679,
      "eval_reassembly": 0.5247901016349978,
      "eval_reassembly_nostereo": 0.5511268228015908,
      "eval_runtime": 1580.4016,
      "eval_samples_per_second": 7.16,
      "eval_steps_per_second": 0.112,
      "eval_tanimoto_similarity": 0.0,
      "eval_valid": 0.9047282368537339,
      "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=23-learning_rate=5.8e-05-warmup_ratio=0.050-min_lr=0.000-factor=0.800",
  "trial_params": {
    "factor": 0.8,
    "learning_rate": 5.8123802212920094e-05,
    "min_lr": 2.921430983672087e-10,
    "warmup_ratio": 0.05
  }
}