add mp20_gen ckpt

.gitattributes

@@ -34,3 +34,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 mp20_csp_s10/ema_avgs=0.044-epoch=3500-step=371105.ckpt.ema_state_dict filter=lfs diff=lfs merge=lfs -text
+mp_20_gen_s1000/ema-epoch=4920-step=521625.ckpt.ema_state_dict filter=lfs diff=lfs merge=lfs -text

mp_20_gen_s1000/.hydra/config.yaml

@@ -0,0 +1,318 @@
+data:
+  root_path: ${oc.env:PROJECT_ROOT}/data/mp_20
+  prop: formation_energy_per_atom
+  num_targets: 1
+  frac_coord_norm: 1.0
+  niggli: true
+  primitive: false
+  graph_method: crystalnn
+  lattice_scale_method: scale_length
+  preprocess_workers: 30
+  readout: mean
+  max_atoms: 20
+  otf_graph: false
+  eval_model_name: mp20
+  train_max_epochs: 5000
+  early_stopping_patience: 50000
+  teacher_forcing_max_epoch: 500
+  std: 7.0
+  datamodule:
+    _target_: crysbfn.pl_data.datamodule.CrystDataModule
+    datasets:
+      train:
+        _target_: crysbfn.pl_data.dataset.CrystDataset
+        name: Formation energy train
+        path: ${data.root_path}/train.csv
+        prop: ${data.prop}
+        niggli: ${data.niggli}
+        primitive: ${data.primitive}
+        graph_method: ${data.graph_method}
+        lattice_scale_method: ${data.lattice_scale_method}
+        preprocess_workers: ${data.preprocess_workers}
+      val:
+      - _target_: crysbfn.pl_data.dataset.CrystDataset
+        name: Formation energy val
+        path: ${data.root_path}/val.csv
+        prop: ${data.prop}
+        niggli: ${data.niggli}
+        primitive: ${data.primitive}
+        graph_method: ${data.graph_method}
+        lattice_scale_method: ${data.lattice_scale_method}
+        preprocess_workers: ${data.preprocess_workers}
+      test:
+      - _target_: crysbfn.pl_data.dataset.CrystDataset
+        name: Formation energy test
+        path: ${data.root_path}/test.csv
+        prop: ${data.prop}
+        niggli: ${data.niggli}
+        primitive: ${data.primitive}
+        graph_method: ${data.graph_method}
+        lattice_scale_method: ${data.lattice_scale_method}
+        preprocess_workers: ${data.preprocess_workers}
+    num_workers:
+      train: 0
+      val: 0
+      test: 0
+    batch_size:
+      train: 256
+      val: 256
+      test: 256
+  lattice_mean: 0.0
+  lattice_std: 3.866470446562982
+  length_normalizer: 1.0
+  num_atom_types: 89
+  atom_type_map:
+    1: 0
+    2: 1
+    3: 2
+    4: 3
+    5: 4
+    6: 5
+    7: 6
+    8: 7
+    9: 8
+    10: 9
+    11: 10
+    12: 11
+    13: 12
+    14: 13
+    15: 14
+    16: 15
+    17: 16
+    18: 17
+    19: 18
+    20: 19
+    21: 20
+    22: 21
+    23: 22
+    24: 23
+    25: 24
+    26: 25
+    27: 26
+    28: 27
+    29: 28
+    30: 29
+    31: 30
+    32: 31
+    33: 32
+    34: 33
+    35: 34
+    36: 35
+    37: 36
+    38: 37
+    39: 38
+    40: 39
+    41: 40
+    42: 41
+    43: 42
+    44: 43
+    45: 44
+    46: 45
+    47: 46
+    48: 47
+    49: 48
+    50: 49
+    51: 50
+    52: 51
+    53: 52
+    54: 53
+    55: 54
+    56: 55
+    57: 56
+    58: 57
+    59: 58
+    60: 59
+    61: 60
+    62: 61
+    63: 62
+    64: 63
+    65: 64
+    66: 65
+    67: 66
+    68: 67
+    69: 68
+    70: 69
+    71: 70
+    72: 71
+    73: 72
+    74: 73
+    75: 74
+    76: 75
+    77: 76
+    78: 77
+    79: 78
+    80: 79
+    81: 80
+    82: 81
+    83: 82
+    89: 83
+    90: 84
+    91: 85
+    92: 86
+    93: 87
+    94: 88
+  num_atoms_dist:
+    1: 110
+    2: 961
+    3: 924
+    4: 6964
+    5: 2130
+    6: 3795
+    7: 991
+    8: 3563
+    9: 1527
+    10: 4395
+    11: 590
+    12: 4396
+    13: 979
+    14: 2959
+    15: 666
+    16: 3026
+    17: 418
+    18: 2418
+    19: 468
+    20: 3949
+  norm_len: false
+  norm_len_log: false
+  norm_abc:
+  - 0.6377525345209046
+  - 0.7004837423370611
+  - 0.9260156168244864
+logging:
+  val_check_interval: 8
+  val_check_interval_step: 5000
+  gen_check_interval: 120
+  progress_bar_refresh_rate: 2
+  run_gen: true
+  debug_mode: false
+  gen_check:
+    num_samples: 128
+    start_epoch: 800
+    run_func_timeout: 14400
+  return_traj: false
+  wandb:
+    name: ${expname}
+    project: crystalbfn
+    entity: null
+    log_model: true
+    mode: online
+    group: ${expname}
+  wandb_watch:
+    log: all
+    log_freq: 500
+  lr_monitor:
+    logging_interval: step
+    log_momentum: false
+model:
+  decoder:
+    _target_: crysbfn.pl_modules.egnn.cspnet.CSPNet
+    hidden_dim: 512
+    time_dim: ${model.time_dim}
+    max_atoms: ${data.num_atom_types}
+    num_layers: 6
+    act_fn: silu
+    dis_emb: sin
+    num_freqs: 128
+    edge_style: fc
+    smooth: true
+    max_neighbors: ${model.max_neighbors}
+    cutoff: ${model.radius}
+    ln: true
+    ip: true
+    pred_type: true
+    period: ${model.T_max}-${model.T_min}
+    dtime_loss: ${model.BFN.dtime_loss}
+    add_n_atom_feat: false
+    cond_acc: ${model.BFN.cond_acc}
+  _target_: crysbfn.pl_modules.crysbfn_plmodel.CrysBFN_PL_Model
+  time_dim: 128
+  hidden_dim: 256
+  latent_dim: 256
+  fc_num_layers: 1
+  max_atoms: ${data.max_atoms}
+  cost_natom: 1.0
+  cost_coord: 1
+  cost_type: 1
+  cost_lattice: 1
+  cost_composition: 1.0
+  cost_edge: 10.0
+  cost_property: 1.0
+  beta: 0.01
+  teacher_forcing_lattice: true
+  teacher_forcing_max_epoch: ${data.teacher_forcing_max_epoch}
+  max_neighbors: 20
+  radius: 7.0
+  sigma_begin: 0.5
+  sigma_end: 0.005
+  type_sigma_begin: 5.0
+  type_sigma_end: 0.01
+  num_noise_level: 50
+  predict_property: true
+  T_min: -np.pi
+  T_max: np.pi
+  norm_weight: false
+  norm_beta: true
+  rej_samp: false
+  n_samples: 1
+  BFN:
+    _target_: crysbfn.pl_modules.crysbfn.CrysBFN
+    device: cuda
+    beta1_type: 0.5
+    beta1_coord: 1000.0
+    K: ${data.num_atom_types}
+    t_min: 1.0e-07
+    dtime_loss: true
+    dtime_loss_steps: 1000
+    sigma1_lattice: 0.03162277660168379
+    mult_constant: true
+    pred_mean: true
+    end_back: true
+    sim_cir_flow: true
+    cond_acc: ${model.BFN.sim_cir_flow}
+    sch_type: linear
+    disc_prob_loss: true
+optim:
+  optimizer:
+    _target_: torch.optim.AdamW
+    lr: 0.001
+    weight_decay: 0
+  use_lr_scheduler: true
+  lr_scheduler:
+    _target_: torch.optim.lr_scheduler.ReduceLROnPlateau
+    factor: 0.6
+    patience: 50
+    min_lr: 0.0001
+train:
+  deterministic: false
+  random_seed: 42
+  use_queue_clip: false
+  gradient_clip_val: 0
+  gradient_clip_algorithm: value
+  pl_trainer:
+    fast_dev_run: false
+    gpus:
+    - 0
+    precision: 32
+    max_steps: 2000000
+    max_epochs: ${data.train_max_epochs}
+    accumulate_grad_batches: 1
+    num_sanity_val_steps: 2
+    profiler: simple
+  monitor_metric: val_loss
+  monitor_metric_mode: min
+  early_stopping:
+    patience: ${data.early_stopping_patience}
+    verbose: false
+  resume: false
+  model_checkpoints:
+    save_top_k: 1
+    verbose: false
+  ema:
+    enable: true
+    decay: 0.995
+    device: cuda
+expname: mp_20_1e3_0.5_s1000_sim_dtime_cate_ema0.995cs_11-07-22-28-31
+core:
+  version: 0.0.1
+  tags:
+  - ${now:%Y-%m-%d}

mp_20_gen_s1000/.hydra/hydra.yaml

@@ -0,0 +1,170 @@
+hydra:
+  run:
+    dir: ${oc.env:HYDRA_JOBS}/singlerun/${now:%Y-%m-%d}/${expname}/
+  sweep:
+    dir: ${oc.env:HYDRA_JOBS}/multirun/${now:%Y-%m-%d}/${expname}/
+    subdir: ${hydra.job.num}_${hydra.job.id}
+  launcher:
+    _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
+  sweeper:
+    _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
+    max_batch_size: null
+  help:
+    app_name: ${hydra.job.name}
+    header: '${hydra.help.app_name} is powered by Hydra.
+
+      '
+    footer: 'Powered by Hydra (https://hydra.cc)
+
+      Use --hydra-help to view Hydra specific help
+
+      '
+    template: '${hydra.help.header}
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (group=option)
+
+
+      $APP_CONFIG_GROUPS
+
+
+      == Config ==
+
+      Override anything in the config (foo.bar=value)
+
+
+      $CONFIG
+
+
+      ${hydra.help.footer}
+
+      '
+  hydra_help:
+    template: 'Hydra (${hydra.runtime.version})
+
+      See https://hydra.cc for more info.
+
+
+      == Flags ==
+
+      $FLAGS_HELP
+
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (For example, append hydra/job_logging=disabled
+      to command line)
+
+
+      $HYDRA_CONFIG_GROUPS
+
+
+      Use ''--cfg hydra'' to Show the Hydra config.
+
+      '
+    hydra_help: ???
+  hydra_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][HYDRA] %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+    root:
+      level: INFO
+      handlers:
+      - console
+    loggers:
+      logging_example:
+        level: DEBUG
+    disable_existing_loggers: false
+  job_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+      file:
+        class: logging.FileHandler
+        formatter: simple
+        filename: ${hydra.job.name}.log
+    root:
+      level: INFO
+      handlers:
+      - console
+      - file
+    disable_existing_loggers: false
+  env: {}
+  searchpath: []
+  callbacks: {}
+  output_subdir: .hydra
+  overrides:
+    hydra: []
+    task:
+    - data=mp_20
+    - expname=mp_20_1e3_0.5_s1000_sim_dtime_cate_ema0.995cs_11-07-22-28-31
+    - train.resume=false
+    - model.BFN.dtime_loss_steps=1000
+    - model.BFN.beta1_coord=1e3
+    - model.BFN.beta1_type=0.5
+    - logging.gen_check.start_epoch=800
+    - logging.gen_check_interval=120
+    - logging.gen_check.num_samples=128
+    - train.ema.enable=true
+    - train.ema.decay=0.995
+    - model.BFN.disc_prob_loss=true
+    - model.BFN.sigma1_lattice=0.03162277660168379
+    - optim.lr_scheduler.patience=50
+  job:
+    name: run
+    override_dirname: data=mp_20,expname=mp_20_1e3_0.5_s1000_sim_dtime_cate_ema0.995cs_11-07-22-28-31,logging.gen_check.num_samples=128,logging.gen_check.start_epoch=800,logging.gen_check_interval=120,model.BFN.beta1_coord=1e3,model.BFN.beta1_type=0.5,model.BFN.disc_prob_loss=true,model.BFN.dtime_loss_steps=1000,model.BFN.sigma1_lattice=0.03162277660168379,optim.lr_scheduler.patience=50,train.ema.decay=0.995,train.ema.enable=true,train.resume=false
+    id: ???
+    num: ???
+    config_name: default
+    env_set:
+      WANDB_START_METHOD: thread
+      WANDB_DIR: ${oc.env:WABDB_DIR}
+    env_copy: []
+    config:
+      override_dirname:
+        kv_sep: '='
+        item_sep: ','
+        exclude_keys: []
+  runtime:
+    version: 1.1.0
+    cwd: /sharefs/material/crysbfn
+    config_sources:
+    - path: hydra.conf
+      schema: pkg
+      provider: hydra
+    - path: /sharefs/material/crysbfn/conf
+      schema: file
+      provider: main
+    - path: ''
+      schema: structured
+      provider: schema
+    choices:
+      train: default
+      optim: default
+      model: bfn
+      model/decoder: cspnet
+      logging: default
+      data: mp_20
+      hydra/env: default
+      hydra/callbacks: null
+      hydra/job_logging: default
+      hydra/hydra_logging: default
+      hydra/hydra_help: default
+      hydra/help: default
+      hydra/sweeper: basic
+      hydra/launcher: basic
+      hydra/output: default
+  verbose: false

mp_20_gen_s1000/.hydra/overrides.yaml

@@ -0,0 +1,14 @@
+- data=mp_20
+- expname=mp_20_1e3_0.5_s1000_sim_dtime_cate_ema0.995cs_11-07-22-28-31
+- train.resume=false
+- model.BFN.dtime_loss_steps=1000
+- model.BFN.beta1_coord=1e3
+- model.BFN.beta1_type=0.5
+- logging.gen_check.start_epoch=800
+- logging.gen_check_interval=120
+- logging.gen_check.num_samples=128
+- train.ema.enable=true
+- train.ema.decay=0.995
+- model.BFN.disc_prob_loss=true
+- model.BFN.sigma1_lattice=0.03162277660168379
+- optim.lr_scheduler.patience=50

mp_20_gen_s1000/ema-epoch=4920-step=521625.ckpt.ema_state_dict

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86f766653f2e815d1723408caeb06c2e02e0c5e6a4ef115c37995a47d12697a4
+size 49109650

mp_20_gen_s1000/hparams.yaml

@@ -0,0 +1,318 @@
+data:
+  root_path: ${oc.env:PROJECT_ROOT}/data/mp_20
+  prop: formation_energy_per_atom
+  num_targets: 1
+  frac_coord_norm: 1.0
+  niggli: true
+  primitive: false
+  graph_method: crystalnn
+  lattice_scale_method: scale_length
+  preprocess_workers: 30
+  readout: mean
+  max_atoms: 20
+  otf_graph: false
+  eval_model_name: mp20
+  train_max_epochs: 5000
+  early_stopping_patience: 50000
+  teacher_forcing_max_epoch: 500
+  std: 7.0
+  datamodule:
+    _target_: crysbfn.pl_data.datamodule.CrystDataModule
+    datasets:
+      train:
+        _target_: crysbfn.pl_data.dataset.CrystDataset
+        name: Formation energy train
+        path: ${data.root_path}/train.csv
+        prop: ${data.prop}
+        niggli: ${data.niggli}
+        primitive: ${data.primitive}
+        graph_method: ${data.graph_method}
+        lattice_scale_method: ${data.lattice_scale_method}
+        preprocess_workers: ${data.preprocess_workers}
+      val:
+      - _target_: crysbfn.pl_data.dataset.CrystDataset
+        name: Formation energy val
+        path: ${data.root_path}/val.csv
+        prop: ${data.prop}
+        niggli: ${data.niggli}
+        primitive: ${data.primitive}
+        graph_method: ${data.graph_method}
+        lattice_scale_method: ${data.lattice_scale_method}
+        preprocess_workers: ${data.preprocess_workers}
+      test:
+      - _target_: crysbfn.pl_data.dataset.CrystDataset
+        name: Formation energy test
+        path: ${data.root_path}/test.csv
+        prop: ${data.prop}
+        niggli: ${data.niggli}
+        primitive: ${data.primitive}
+        graph_method: ${data.graph_method}
+        lattice_scale_method: ${data.lattice_scale_method}
+        preprocess_workers: ${data.preprocess_workers}
+    num_workers:
+      train: 0
+      val: 0
+      test: 0
+    batch_size:
+      train: 256
+      val: 256
+      test: 256
+  lattice_mean: 0.0
+  lattice_std: 3.866470446562982
+  length_normalizer: 1.0
+  num_atom_types: 89
+  atom_type_map:
+    1: 0
+    2: 1
+    3: 2
+    4: 3
+    5: 4
+    6: 5
+    7: 6
+    8: 7
+    9: 8
+    10: 9
+    11: 10
+    12: 11
+    13: 12
+    14: 13
+    15: 14
+    16: 15
+    17: 16
+    18: 17
+    19: 18
+    20: 19
+    21: 20
+    22: 21
+    23: 22
+    24: 23
+    25: 24
+    26: 25
+    27: 26
+    28: 27
+    29: 28
+    30: 29
+    31: 30
+    32: 31
+    33: 32
+    34: 33
+    35: 34
+    36: 35
+    37: 36
+    38: 37
+    39: 38
+    40: 39
+    41: 40
+    42: 41
+    43: 42
+    44: 43
+    45: 44
+    46: 45
+    47: 46
+    48: 47
+    49: 48
+    50: 49
+    51: 50
+    52: 51
+    53: 52
+    54: 53
+    55: 54
+    56: 55
+    57: 56
+    58: 57
+    59: 58
+    60: 59
+    61: 60
+    62: 61
+    63: 62
+    64: 63
+    65: 64
+    66: 65
+    67: 66
+    68: 67
+    69: 68
+    70: 69
+    71: 70
+    72: 71
+    73: 72
+    74: 73
+    75: 74
+    76: 75
+    77: 76
+    78: 77
+    79: 78
+    80: 79
+    81: 80
+    82: 81
+    83: 82
+    89: 83
+    90: 84
+    91: 85
+    92: 86
+    93: 87
+    94: 88
+  num_atoms_dist:
+    1: 110
+    2: 961
+    3: 924
+    4: 6964
+    5: 2130
+    6: 3795
+    7: 991
+    8: 3563
+    9: 1527
+    10: 4395
+    11: 590
+    12: 4396
+    13: 979
+    14: 2959
+    15: 666
+    16: 3026
+    17: 418
+    18: 2418
+    19: 468
+    20: 3949
+  norm_len: false
+  norm_len_log: false
+  norm_abc:
+  - 0.6377525345209046
+  - 0.7004837423370611
+  - 0.9260156168244864
+logging:
+  val_check_interval: 8
+  val_check_interval_step: 5000
+  gen_check_interval: 120
+  progress_bar_refresh_rate: 2
+  run_gen: true
+  debug_mode: false
+  gen_check:
+    num_samples: 128
+    start_epoch: 800
+    run_func_timeout: 14400
+  return_traj: false
+  wandb:
+    name: ${expname}
+    project: crystalbfn
+    entity: null
+    log_model: true
+    mode: online
+    group: ${expname}
+  wandb_watch:
+    log: all
+    log_freq: 500
+  lr_monitor:
+    logging_interval: step
+    log_momentum: false
+model:
+  decoder:
+    _target_: crysbfn.pl_modules.egnn.cspnet.CSPNet
+    hidden_dim: 512
+    time_dim: ${model.time_dim}
+    max_atoms: ${data.num_atom_types}
+    num_layers: 6
+    act_fn: silu
+    dis_emb: sin
+    num_freqs: 128
+    edge_style: fc
+    smooth: true
+    max_neighbors: ${model.max_neighbors}
+    cutoff: ${model.radius}
+    ln: true
+    ip: true
+    pred_type: true
+    period: ${model.T_max}-${model.T_min}
+    dtime_loss: ${model.BFN.dtime_loss}
+    add_n_atom_feat: false
+    cond_acc: ${model.BFN.cond_acc}
+  _target_: crysbfn.pl_modules.crysbfn_plmodel.CrysBFN_PL_Model
+  time_dim: 128
+  hidden_dim: 256
+  latent_dim: 256
+  fc_num_layers: 1
+  max_atoms: ${data.max_atoms}
+  cost_natom: 1.0
+  cost_coord: 1
+  cost_type: 1
+  cost_lattice: 1
+  cost_composition: 1.0
+  cost_edge: 10.0
+  cost_property: 1.0
+  beta: 0.01
+  teacher_forcing_lattice: true
+  teacher_forcing_max_epoch: ${data.teacher_forcing_max_epoch}
+  max_neighbors: 20
+  radius: 7.0
+  sigma_begin: 0.5
+  sigma_end: 0.005
+  type_sigma_begin: 5.0
+  type_sigma_end: 0.01
+  num_noise_level: 50
+  predict_property: true
+  T_min: -np.pi
+  T_max: np.pi
+  norm_weight: false
+  norm_beta: true
+  rej_samp: false
+  n_samples: 1
+  BFN:
+    _target_: crysbfn.pl_modules.crysbfn.CrysBFN
+    device: cuda
+    beta1_type: 0.5
+    beta1_coord: 1000.0
+    K: ${data.num_atom_types}
+    t_min: 1.0e-07
+    dtime_loss: true
+    dtime_loss_steps: 1000
+    sigma1_lattice: 0.03162277660168379
+    mult_constant: true
+    pred_mean: true
+    end_back: true
+    sim_cir_flow: true
+    cond_acc: ${model.BFN.sim_cir_flow}
+    sch_type: linear
+    disc_prob_loss: true
+optim:
+  optimizer:
+    _target_: torch.optim.AdamW
+    lr: 0.001
+    weight_decay: 0
+  use_lr_scheduler: true
+  lr_scheduler:
+    _target_: torch.optim.lr_scheduler.ReduceLROnPlateau
+    factor: 0.6
+    patience: 50
+    min_lr: 0.0001
+train:
+  deterministic: false
+  random_seed: 42
+  use_queue_clip: false
+  gradient_clip_val: 0
+  gradient_clip_algorithm: value
+  pl_trainer:
+    fast_dev_run: false
+    gpus:
+    - 0
+    precision: 32
+    max_steps: 2000000
+    max_epochs: ${data.train_max_epochs}
+    accumulate_grad_batches: 1
+    num_sanity_val_steps: 2
+    profiler: simple
+  monitor_metric: val_loss
+  monitor_metric_mode: min
+  early_stopping:
+    patience: ${data.early_stopping_patience}
+    verbose: false
+  resume: false
+  model_checkpoints:
+    save_top_k: 1
+    verbose: false
+  ema:
+    enable: true
+    decay: 0.995
+    device: cuda
+expname: mp_20_1e3_0.5_s1000_sim_dtime_cate_ema0.995cs_11-07-22-28-31
+core:
+  version: 0.0.1
+  tags:
+  - ${now:%Y-%m-%d}