trial-number=3-learning_rate=5.4e-05-warmup_ratio=0.060-num_cycles=8/checkpoint-5000/trainer_state.json

{
  "best_metric": 0.48351745470614227,
  "best_model_checkpoint": "/mimer/NOBACKUP/groups/naiss2023-6-290/stefano/models//PROTAC-Splitter-EncoderDecoder-lr_cosine_restarts-opt25/trial-number=3-learning_rate=5.4e-05-warmup_ratio=0.060-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": 0.7583757042884827,
      "learning_rate": 4.136371276196747e-05,
      "loss": 1.3436,
      "step": 500
    },
    {
      "epoch": 0.09863878477017163,
      "grad_norm": 0.5184589624404907,
      "learning_rate": 9.34782831136264e-06,
      "loss": 0.1858,
      "step": 1000
    },
    {
      "epoch": 0.14795817715525744,
      "grad_norm": 0.46950823068618774,
      "learning_rate": 5.1896305774682246e-05,
      "loss": 0.1246,
      "step": 1500
    },
    {
      "epoch": 0.19727756954034326,
      "grad_norm": 0.24211066961288452,
      "learning_rate": 2.4082255094758885e-05,
      "loss": 0.0679,
      "step": 2000
    },
    {
      "epoch": 0.24659696192542907,
      "grad_norm": 0.2779711186885834,
      "learning_rate": 4.894471651454904e-07,
      "loss": 0.0405,
      "step": 2500
    },
    {
      "epoch": 0.2959163543105149,
      "grad_norm": 0.2603333294391632,
      "learning_rate": 3.9875315816945065e-05,
      "loss": 0.0374,
      "step": 3000
    },
    {
      "epoch": 0.3452357466956007,
      "grad_norm": 0.17240776121616364,
      "learning_rate": 8.085320248404235e-06,
      "loss": 0.0221,
      "step": 3500
    },
    {
      "epoch": 0.3945551390806865,
      "grad_norm": 0.20610322058200836,
      "learning_rate": 5.115763756547701e-05,
      "loss": 0.0195,
      "step": 4000
    },
    {
      "epoch": 0.4438745314657723,
      "grad_norm": 0.1784711331129074,
      "learning_rate": 2.238109574795991e-05,
      "loss": 0.0173,
      "step": 4500
    },
    {
      "epoch": 0.49319392385085814,
      "grad_norm": 0.16437868773937225,
      "learning_rate": 2.1789763086122026e-07,
      "loss": 0.0108,
      "step": 5000
    },
    {
      "epoch": 0.49319392385085814,
      "eval_all_ligands_equal": 0.48351745470614227,
      "eval_e3_equal": 0.7912505523641185,
      "eval_e3_graph_edit_distance": 9.999999999999999e+63,
      "eval_e3_graph_edit_distance_norm": 1.0,
      "eval_e3_has_attachment_point(s)": 0.9908970393283252,
      "eval_e3_heavy_atoms_difference": 0.45373398144056565,
      "eval_e3_heavy_atoms_difference_norm": 0.005591781364094335,
      "eval_e3_tanimoto_similarity": 0.0,
      "eval_e3_valid": 0.9908970393283252,
      "eval_has_all_attachment_points": 0.9815289438798056,
      "eval_has_three_substructures": 0.999116217410517,
      "eval_heavy_atoms_difference": 7.107467962881131,
      "eval_heavy_atoms_difference_norm": 0.09374154126686561,
      "eval_linker_equal": 0.6897039328325232,
      "eval_linker_graph_edit_distance": 9.999999999999999e+63,
      "eval_linker_graph_edit_distance_norm": 1.0,
      "eval_linker_has_attachment_point(s)": 0.9963764913831198,
      "eval_linker_heavy_atoms_difference": 0.3809102960671675,
      "eval_linker_heavy_atoms_difference_norm": -0.0016906051655895635,
      "eval_linker_tanimoto_similarity": 0.0,
      "eval_linker_valid": 0.9963764913831198,
      "eval_loss": 0.2832363545894623,
      "eval_num_fragments": 3.0001767565178965,
      "eval_poi_equal": 0.7419354838709677,
      "eval_poi_graph_edit_distance": 9.999999999999999e+63,
      "eval_poi_graph_edit_distance_norm": 1.0,
      "eval_poi_has_attachment_point(s)": 0.9233760494918251,
      "eval_poi_heavy_atoms_difference": 2.1789659743703047,
      "eval_poi_heavy_atoms_difference_norm": 0.07135149949500147,
      "eval_poi_tanimoto_similarity": 0.0,
      "eval_poi_valid": 0.9233760494918251,
      "eval_reassembly": 0.4908528501988511,
      "eval_reassembly_nostereo": 0.5169244365885992,
      "eval_runtime": 1508.7434,
      "eval_samples_per_second": 7.5,
      "eval_steps_per_second": 0.117,
      "eval_tanimoto_similarity": 0.0,
      "eval_valid": 0.9135660627485639,
      "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=3-learning_rate=5.4e-05-warmup_ratio=0.060-num_cycles=8",
  "trial_params": {
    "learning_rate": 5.4168442890996515e-05,
    "num_cycles": 8,
    "warmup_ratio": 0.060000000000000005
  }
}