last-checkpoint/trainer_state.json

{
  "best_metric": 0.06389748121961997,
  "best_model_checkpoint": "/mimer/NOBACKUP/groups/naiss2023-6-290/stefano/models//PROTAC-Splitter-EncoderDecoder-lr_reduce-opt25/trial-number=13-learning_rate=9.3e-06-warmup_ratio=0.060-min_lr=0.000-factor=0.910/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": 1.1640962362289429,
      "learning_rate": 9.294928212675126e-06,
      "loss": 1.8212,
      "step": 500
    },
    {
      "epoch": 0.09863878477017163,
      "grad_norm": 1.0102286338806152,
      "learning_rate": 9.294928212675126e-06,
      "loss": 0.6458,
      "step": 1000
    },
    {
      "epoch": 0.14795817715525744,
      "grad_norm": 0.8785966038703918,
      "learning_rate": 9.294928212675126e-06,
      "loss": 0.4104,
      "step": 1500
    },
    {
      "epoch": 0.19727756954034326,
      "grad_norm": 0.7990570664405823,
      "learning_rate": 9.294928212675126e-06,
      "loss": 0.2907,
      "step": 2000
    },
    {
      "epoch": 0.24659696192542907,
      "grad_norm": 0.7908627986907959,
      "learning_rate": 9.294928212675126e-06,
      "loss": 0.2186,
      "step": 2500
    },
    {
      "epoch": 0.2959163543105149,
      "grad_norm": 0.571479082107544,
      "learning_rate": 9.294928212675126e-06,
      "loss": 0.1745,
      "step": 3000
    },
    {
      "epoch": 0.3452357466956007,
      "grad_norm": 0.37610748410224915,
      "learning_rate": 9.294928212675126e-06,
      "loss": 0.1433,
      "step": 3500
    },
    {
      "epoch": 0.3945551390806865,
      "grad_norm": 0.39717283844947815,
      "learning_rate": 9.294928212675126e-06,
      "loss": 0.1201,
      "step": 4000
    },
    {
      "epoch": 0.4438745314657723,
      "grad_norm": 0.4328235387802124,
      "learning_rate": 9.294928212675126e-06,
      "loss": 0.1006,
      "step": 4500
    },
    {
      "epoch": 0.49319392385085814,
      "grad_norm": 0.39815056324005127,
      "learning_rate": 9.294928212675126e-06,
      "loss": 0.0848,
      "step": 5000
    },
    {
      "epoch": 0.49319392385085814,
      "eval_all_ligands_equal": 0.06389748121961997,
      "eval_e3_equal": 0.5864781263809103,
      "eval_e3_graph_edit_distance": 9.999999999999999e+63,
      "eval_e3_graph_edit_distance_norm": 1.0,
      "eval_e3_has_attachment_point(s)": 0.9963764913831198,
      "eval_e3_heavy_atoms_difference": -0.3860362350861688,
      "eval_e3_heavy_atoms_difference_norm": -0.02804502844059732,
      "eval_e3_tanimoto_similarity": 0.0,
      "eval_e3_valid": 0.9963764913831198,
      "eval_has_all_attachment_points": 0.9783473265576668,
      "eval_has_three_substructures": 0.9984975695978789,
      "eval_heavy_atoms_difference": 11.214140521431728,
      "eval_heavy_atoms_difference_norm": 0.14692428805581007,
      "eval_linker_equal": 0.12037118868758286,
      "eval_linker_graph_edit_distance": 9.999999999999999e+63,
      "eval_linker_graph_edit_distance_norm": 1.0,
      "eval_linker_has_attachment_point(s)": 0.9952275740167918,
      "eval_linker_heavy_atoms_difference": 2.771984091913389,
      "eval_linker_heavy_atoms_difference_norm": 0.1170382519264519,
      "eval_linker_tanimoto_similarity": 0.0,
      "eval_linker_valid": 0.9952275740167918,
      "eval_loss": 0.3106701374053955,
      "eval_num_fragments": 2.9990278391515686,
      "eval_poi_equal": 0.5546619531595227,
      "eval_poi_graph_edit_distance": 9.999999999999999e+63,
      "eval_poi_graph_edit_distance_norm": 1.0,
      "eval_poi_has_attachment_point(s)": 0.8872293415819709,
      "eval_poi_heavy_atoms_difference": 3.580998674326116,
      "eval_poi_heavy_atoms_difference_norm": 0.11244772864400035,
      "eval_poi_tanimoto_similarity": 0.0,
      "eval_poi_valid": 0.8872293415819709,
      "eval_reassembly": 0.06575342465753424,
      "eval_reassembly_nostereo": 0.0758285461776403,
      "eval_runtime": 1503.1486,
      "eval_samples_per_second": 7.528,
      "eval_steps_per_second": 0.118,
      "eval_tanimoto_similarity": 0.0,
      "eval_valid": 0.8827220503756076,
      "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=13-learning_rate=9.3e-06-warmup_ratio=0.060-min_lr=0.000-factor=0.910",
  "trial_params": {
    "factor": 0.91,
    "learning_rate": 9.294928212675126e-06,
    "min_lr": 2.837616835512592e-09,
    "warmup_ratio": 0.06
  }
}