umami.models package#

Submodules#

umami.models.model_cads module#

Keras model of the CADS tagger.

umami.models.model_cads.create_cads_model(train_config: object, input_shape: tuple)#

Keras model definition of CADS.

Parameters:

  • train_config (object) – training config

  • input_shape (tuple) – dataset input shape

Returns:

  • keras model – CADS keras model

  • int – Number of epochs

  • int – Starting epoch number

umami.models.model_cads.train_cads(args, train_config)#

Training handling of CADS.

Parameters:

  • args (parser args) – Arguments from command line parser

  • train_config (object) – training configuration

Raises:

ValueError – If input is neither a h5 nor a directory.

umami.models.model_dips module#

Keras model of the DIPS tagger.

umami.models.model_dips.create_dips_model(train_config: object, input_shape: tuple)#

Keras model definition of DIPS.

Parameters:

  • train_config (object) – training config

  • input_shape (tuple) – dataset input shape

Returns:

  • keras model – Dips keras model

  • int – Number of epochs

  • int – Starting epoch number

umami.models.model_dips.train_dips(args, train_config)#

Training handling of DIPS tagger.

Parameters:

  • args (parser args) – Arguments from command line parser

  • train_config (object) – training configuration

Raises:

ValueError – If input is neither a h5 nor a directory.

umami.models.model_dl1 module#

Keras model of the DL1 tagger.

umami.models.model_dl1.create_dl1_model(train_config: object, input_shape: tuple, feature_connect_indices: list | None = None)#

Constructs or loads the DL1 model

Parameters:

  • train_config (object) – Training configuration with nn_structure sub-dict giving the structure of the NN.

  • input_shape (tuple) – Size of the input: (nFeatures,).

  • feature_connect_indices (list) – List with features that are feeded in another time.

Returns:

  • model (keras model) – Keras model.

  • nn_structure.epochs – number of epochs to be trained

  • init_epoch (int) – Starting epoch number

umami.models.model_dl1.train_dl1(args, train_config)#

Training handling of DL1 tagger.

Parameters:

  • args (parser args) – Arguments from command line parser

  • train_config (object) – training configuration

Raises:

ValueError – If input is neither a h5 nor a directory.

umami.models.model_umami module#

Keras model of the UMAMI tagger.

umami.models.model_umami.create_umami_model(train_config: object, input_shape: tuple, njet_features: int)#

Keras model definition of UMAMI tagger.

Parameters:

  • train_config (object) – training config

  • input_shape (tuple) – dataset input shape

  • njet_features (int) – number of jet features

Returns:

  • keras model – UMAMI keras model

  • int – Number of epochs

  • int – Starting epoch number

umami.models.model_umami.train_umami(args, train_config)#

Training handling of UMAMI tagger.

Parameters:

  • args (parser args) – Arguments from command line parser

  • train_config (object) – training configuration

Raises:

ValueError – If input is neither a h5 nor a directory.

umami.models.model_umami_cond_att module#

Keras model of the UMAMI with conditional attention tagger.

umami.models.model_umami_cond_att.create_umami_cond_att_model(train_config: object, input_shape: tuple, njet_features: int)#

Keras model definition of UMAMI tagger.

Parameters:

  • train_config (object) – training config

  • input_shape (tuple) – dataset input shape

  • njet_features (int) – number of jet features

Returns:

  • keras model – UMAMI with conditional attention keras model

  • int – number of epochs

  • int – Starting epoch number

umami.models.model_umami_cond_att.train_umami_cond_att(args, train_config)#

Training handling of UMAMI with conditional attention tagger.

Parameters:

  • args (parser args) – Arguments from command line parser

  • train_config (object) – training configuration

Raises:

ValueError – If input is neither a h5 nor a directory.

Module contents#