last-checkpoint/trainer_state.json

{
  "best_metric": 0.004507291206363235,
  "best_model_checkpoint": "/mimer/NOBACKUP/groups/naiss2023-6-290/stefano/models//PROTAC-Splitter-EncoderDecoder-lr_reduce-opt25/trial-number=21-learning_rate=6.4e-06-warmup_ratio=0.050-min_lr=0.000-factor=0.850/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": 1.1671305894851685,
      "learning_rate": 6.43023792571843e-06,
      "loss": 2.1019,
      "step": 500
    },
    {
      "epoch": 0.09863878477017163,
      "grad_norm": 1.1441779136657715,
      "learning_rate": 6.43023792571843e-06,
      "loss": 0.8011,
      "step": 1000
    },
    {
      "epoch": 0.14795817715525744,
      "grad_norm": 1.0528101921081543,
      "learning_rate": 6.43023792571843e-06,
      "loss": 0.5442,
      "step": 1500
    },
    {
      "epoch": 0.19727756954034326,
      "grad_norm": 0.9411584734916687,
      "learning_rate": 6.43023792571843e-06,
      "loss": 0.4046,
      "step": 2000
    },
    {
      "epoch": 0.24659696192542907,
      "grad_norm": 0.9701049327850342,
      "learning_rate": 6.43023792571843e-06,
      "loss": 0.3171,
      "step": 2500
    },
    {
      "epoch": 0.2959163543105149,
      "grad_norm": 0.7730458974838257,
      "learning_rate": 6.43023792571843e-06,
      "loss": 0.2574,
      "step": 3000
    },
    {
      "epoch": 0.3452357466956007,
      "grad_norm": 0.45782116055488586,
      "learning_rate": 6.43023792571843e-06,
      "loss": 0.2138,
      "step": 3500
    },
    {
      "epoch": 0.3945551390806865,
      "grad_norm": 0.5900447964668274,
      "learning_rate": 6.43023792571843e-06,
      "loss": 0.1834,
      "step": 4000
    },
    {
      "epoch": 0.4438745314657723,
      "grad_norm": 0.5426287651062012,
      "learning_rate": 6.43023792571843e-06,
      "loss": 0.1595,
      "step": 4500
    },
    {
      "epoch": 0.49319392385085814,
      "grad_norm": 0.5176225900650024,
      "learning_rate": 6.43023792571843e-06,
      "loss": 0.1413,
      "step": 5000
    },
    {
      "epoch": 0.49319392385085814,
      "eval_all_ligands_equal": 0.004507291206363235,
      "eval_e3_equal": 0.40459566946531156,
      "eval_e3_graph_edit_distance": 9.999999999999999e+63,
      "eval_e3_graph_edit_distance_norm": 1.0,
      "eval_e3_has_attachment_point(s)": 0.9962881131241714,
      "eval_e3_heavy_atoms_difference": -0.7212549712770658,
      "eval_e3_heavy_atoms_difference_norm": -0.04417093492101355,
      "eval_e3_tanimoto_similarity": 0.0,
      "eval_e3_valid": 0.9962881131241714,
      "eval_has_all_attachment_points": 0.9608484312859037,
      "eval_has_three_substructures": 0.997348652231551,
      "eval_heavy_atoms_difference": 15.638356164383561,
      "eval_heavy_atoms_difference_norm": 0.20120441984493373,
      "eval_linker_equal": 0.018382677861246133,
      "eval_linker_graph_edit_distance": 9.999999999999999e+63,
      "eval_linker_graph_edit_distance_norm": 1.0,
      "eval_linker_has_attachment_point(s)": 0.9916040653999116,
      "eval_linker_heavy_atoms_difference": 3.2921785240830754,
      "eval_linker_heavy_atoms_difference_norm": 0.10986259482255506,
      "eval_linker_tanimoto_similarity": 0.0,
      "eval_linker_valid": 0.9916040653999116,
      "eval_loss": 0.3394020199775696,
      "eval_num_fragments": 2.9975254087494476,
      "eval_poi_equal": 0.4064516129032258,
      "eval_poi_graph_edit_distance": 9.999999999999999e+63,
      "eval_poi_graph_edit_distance_norm": 1.0,
      "eval_poi_has_attachment_point(s)": 0.8352629253203712,
      "eval_poi_heavy_atoms_difference": 5.534688466637207,
      "eval_poi_heavy_atoms_difference_norm": 0.164553420169889,
      "eval_poi_tanimoto_similarity": 0.0,
      "eval_poi_valid": 0.8352629253203712,
      "eval_reassembly": 0.004507291206363235,
      "eval_reassembly_nostereo": 0.006628369421122404,
      "eval_runtime": 1469.5715,
      "eval_samples_per_second": 7.7,
      "eval_steps_per_second": 0.12,
      "eval_tanimoto_similarity": 0.0,
      "eval_valid": 0.8312859036676977,
      "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=21-learning_rate=6.4e-06-warmup_ratio=0.050-min_lr=0.000-factor=0.850",
  "trial_params": {
    "factor": 0.8500000000000001,
    "learning_rate": 6.43023792571843e-06,
    "min_lr": 4.575542841847297e-10,
    "warmup_ratio": 0.05
  }
}