trial-number=18-learning_rate=2.0e-04-warmup_ratio=0.010-num_cycles=3/checkpoint-5000/trainer_state.json

{
  "best_metric": 0.5108263367211666,
  "best_model_checkpoint": "/mimer/NOBACKUP/groups/naiss2023-6-290/stefano/models//PROTAC-Splitter-EncoderDecoder-lr_cosine_restarts-opt25/trial-number=18-learning_rate=2.0e-04-warmup_ratio=0.010-num_cycles=3/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": 0.5011103749275208,
      "learning_rate": 0.00019333014963385616,
      "loss": 0.9833,
      "step": 500
    },
    {
      "epoch": 0.09863878477017163,
      "grad_norm": 0.30493640899658203,
      "learning_rate": 0.00016595114383716005,
      "loss": 0.1736,
      "step": 1000
    },
    {
      "epoch": 0.14795817715525744,
      "grad_norm": 0.20391501486301422,
      "learning_rate": 0.00012396207622088705,
      "loss": 0.065,
      "step": 1500
    },
    {
      "epoch": 0.19727756954034326,
      "grad_norm": 0.12541060149669647,
      "learning_rate": 7.66983385109593e-05,
      "loss": 0.0227,
      "step": 2000
    },
    {
      "epoch": 0.24659696192542907,
      "grad_norm": 0.10875938832759857,
      "learning_rate": 3.466803510220776e-05,
      "loss": 0.0121,
      "step": 2500
    },
    {
      "epoch": 0.2959163543105149,
      "grad_norm": 0.10840030014514923,
      "learning_rate": 7.215725637295107e-06,
      "loss": 0.0084,
      "step": 3000
    },
    {
      "epoch": 0.3452357466956007,
      "grad_norm": 0.28355872631073,
      "learning_rate": 0.0002000300854243022,
      "loss": 0.0154,
      "step": 3500
    },
    {
      "epoch": 0.3945551390806865,
      "grad_norm": 0.1145474836230278,
      "learning_rate": 0.00018461415129199327,
      "loss": 0.0275,
      "step": 4000
    },
    {
      "epoch": 0.4438745314657723,
      "grad_norm": 0.1480787992477417,
      "learning_rate": 0.00015043882563917473,
      "loss": 0.0093,
      "step": 4500
    },
    {
      "epoch": 0.49319392385085814,
      "grad_norm": 0.08729447424411774,
      "learning_rate": 0.00010510227795238147,
      "loss": 0.0059,
      "step": 5000
    },
    {
      "epoch": 0.49319392385085814,
      "eval_all_ligands_equal": 0.5108263367211666,
      "eval_e3_equal": 0.7976137870083959,
      "eval_e3_graph_edit_distance": 9.999999999999999e+63,
      "eval_e3_graph_edit_distance_norm": 1.0,
      "eval_e3_has_attachment_point(s)": 0.9916924436588599,
      "eval_e3_heavy_atoms_difference": 0.4874944763588157,
      "eval_e3_heavy_atoms_difference_norm": 0.006469055588544779,
      "eval_e3_tanimoto_similarity": 0.0,
      "eval_e3_valid": 0.9916924436588599,
      "eval_has_all_attachment_points": 0.9919575784357049,
      "eval_has_three_substructures": 0.9999116217410517,
      "eval_heavy_atoms_difference": 2.0157313300927973,
      "eval_heavy_atoms_difference_norm": 0.02104730460463736,
      "eval_linker_equal": 0.7404330534688467,
      "eval_linker_graph_edit_distance": 9.999999999999999e+63,
      "eval_linker_graph_edit_distance_norm": 1.0,
      "eval_linker_has_attachment_point(s)": 0.9988510826336722,
      "eval_linker_heavy_atoms_difference": 0.546442775077331,
      "eval_linker_heavy_atoms_difference_norm": 0.006276838082081687,
      "eval_linker_tanimoto_similarity": 0.0,
      "eval_linker_valid": 0.9988510826336722,
      "eval_loss": 0.32633018493652344,
      "eval_num_fragments": 2.9999116217410515,
      "eval_poi_equal": 0.744675209898365,
      "eval_poi_graph_edit_distance": 9.999999999999999e+63,
      "eval_poi_graph_edit_distance_norm": 1.0,
      "eval_poi_has_attachment_point(s)": 0.9913389306230668,
      "eval_poi_heavy_atoms_difference": -0.07856827220503756,
      "eval_poi_heavy_atoms_difference_norm": -0.022628294753132334,
      "eval_poi_tanimoto_similarity": 0.0,
      "eval_poi_valid": 0.9913389306230668,
      "eval_reassembly": 0.5181617322138754,
      "eval_reassembly_nostereo": 0.5433495360141405,
      "eval_runtime": 1536.1175,
      "eval_samples_per_second": 7.366,
      "eval_steps_per_second": 0.115,
      "eval_tanimoto_similarity": 0.0,
      "eval_valid": 0.982235969951392,
      "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=18-learning_rate=2.0e-04-warmup_ratio=0.010-num_cycles=3",
  "trial_params": {
    "learning_rate": 0.0002004749428706299,
    "num_cycles": 3,
    "warmup_ratio": 0.01
  }
}