trial-number=0-learning_rate=1.3e-05-warmup_ratio=0.100-num_cycles=8/checkpoint-5000/trainer_state.json

{
  "best_metric": 0.01104728236853734,
  "best_model_checkpoint": "/mimer/NOBACKUP/groups/naiss2023-6-290/stefano/models//PROTAC-Splitter-EncoderDecoder-lr_cosine_restarts-opt25/trial-number=0-learning_rate=1.3e-05-warmup_ratio=0.100-num_cycles=8/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": 1.1029289960861206,
      "learning_rate": 1.0150642174439675e-05,
      "loss": 2.1079,
      "step": 500
    },
    {
      "epoch": 0.09863878477017163,
      "grad_norm": 1.069559931755066,
      "learning_rate": 2.2939541438839045e-06,
      "loss": 0.612,
      "step": 1000
    },
    {
      "epoch": 0.14795817715525744,
      "grad_norm": 1.0086921453475952,
      "learning_rate": 1.2735337205476001e-05,
      "loss": 0.5012,
      "step": 1500
    },
    {
      "epoch": 0.19727756954034326,
      "grad_norm": 0.7310407161712646,
      "learning_rate": 5.909777868036016e-06,
      "loss": 0.3491,
      "step": 2000
    },
    {
      "epoch": 0.24659696192542907,
      "grad_norm": 0.7290641069412231,
      "learning_rate": 1.2011018124209218e-07,
      "loss": 0.275,
      "step": 2500
    },
    {
      "epoch": 0.2959163543105149,
      "grad_norm": 0.6321003437042236,
      "learning_rate": 9.785390029657761e-06,
      "loss": 0.2385,
      "step": 3000
    },
    {
      "epoch": 0.3452357466956007,
      "grad_norm": 0.42309775948524475,
      "learning_rate": 1.9841350601091295e-06,
      "loss": 0.1832,
      "step": 3500
    },
    {
      "epoch": 0.3945551390806865,
      "grad_norm": 0.5520086288452148,
      "learning_rate": 1.2554068257970625e-05,
      "loss": 0.1685,
      "step": 4000
    },
    {
      "epoch": 0.4438745314657723,
      "grad_norm": 0.4456807076931,
      "learning_rate": 5.492313896403925e-06,
      "loss": 0.1392,
      "step": 4500
    },
    {
      "epoch": 0.49319392385085814,
      "grad_norm": 0.4616454541683197,
      "learning_rate": 5.347201046140292e-08,
      "loss": 0.1218,
      "step": 5000
    },
    {
      "epoch": 0.49319392385085814,
      "eval_all_ligands_equal": 0.01104728236853734,
      "eval_e3_equal": 0.4543526292532037,
      "eval_e3_graph_edit_distance": 9.999999999999999e+63,
      "eval_e3_graph_edit_distance_norm": 1.0,
      "eval_e3_has_attachment_point(s)": 0.9961997348652232,
      "eval_e3_heavy_atoms_difference": -0.7910737958462218,
      "eval_e3_heavy_atoms_difference_norm": -0.04760925516085696,
      "eval_e3_tanimoto_similarity": 0.0,
      "eval_e3_valid": 0.9961997348652232,
      "eval_has_all_attachment_points": 0.971895713654441,
      "eval_has_three_substructures": 0.9972602739726028,
      "eval_heavy_atoms_difference": 13.540079540433053,
      "eval_heavy_atoms_difference_norm": 0.17612445546772115,
      "eval_linker_equal": 0.03296509058771542,
      "eval_linker_graph_edit_distance": 9.999999999999999e+63,
      "eval_linker_graph_edit_distance_norm": 1.0,
      "eval_linker_has_attachment_point(s)": 0.9920459566946531,
      "eval_linker_heavy_atoms_difference": 2.8389748121961995,
      "eval_linker_heavy_atoms_difference_norm": 0.09419788075518785,
      "eval_linker_tanimoto_similarity": 0.0,
      "eval_linker_valid": 0.9920459566946531,
      "eval_loss": 0.3278193175792694,
      "eval_num_fragments": 2.998851082633672,
      "eval_poi_equal": 0.44604507291206363,
      "eval_poi_graph_edit_distance": 9.999999999999999e+63,
      "eval_poi_graph_edit_distance_norm": 1.0,
      "eval_poi_has_attachment_point(s)": 0.8559434379142731,
      "eval_poi_heavy_atoms_difference": 4.626159964648696,
      "eval_poi_heavy_atoms_difference_norm": 0.13707884773495643,
      "eval_poi_tanimoto_similarity": 0.0,
      "eval_poi_valid": 0.8559434379142731,
      "eval_reassembly": 0.011224038886433937,
      "eval_reassembly_nostereo": 0.014582412726469289,
      "eval_runtime": 1604.1755,
      "eval_samples_per_second": 7.053,
      "eval_steps_per_second": 0.11,
      "eval_tanimoto_similarity": 0.0,
      "eval_valid": 0.8517012814847548,
      "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=0-learning_rate=1.3e-05-warmup_ratio=0.100-num_cycles=8",
  "trial_params": {
    "learning_rate": 1.329291894316217e-05,
    "num_cycles": 8,
    "warmup_ratio": 0.09999999999999999
  }
}