immuneML.ml_methods package

Subpackages

Submodules

immuneML.ml_methods.AtchleyKmerMILClassifier module

class immuneML.ml_methods.AtchleyKmerMILClassifier.AtchleyKmerMILClassifier(iteration_count: int = None, threshold: float = None, evaluate_at: int = None, use_early_stopping: bool = None, random_seed: int = None, learning_rate: float = None, zero_abundance_weight_init: bool = None, number_of_threads: int = None, result_path: Path = None, initialization_count: int = None, pytorch_device_name: str = None)[source]

Bases: MLMethod

A binary Repertoire classifier which uses the data encoded by AtchleyKmer encoder to predict the repertoire label.

The original publication: Ostmeyer J, Christley S, Toby IT, Cowell LG. Biophysicochemical motifs in T cell receptor sequences distinguish repertoires from tumor-infiltrating lymphocytes and adjacent healthy tissue. Cancer Res. Published online January 1, 2019:canres.2292.2018. doi:10.1158/0008-5472.CAN-18-2292 .

Parameters:
  • iteration_count (int) – max number of training iterations

  • threshold (float) – loss threshold at which to stop training if reached

  • evaluate_at (int) – log model performance every ‘evaluate_at’ iterations and store the model every ‘evaluate_at’ iterations if early stopping

  • used (is) –

  • use_early_stopping (bool) – whether to use early stopping

  • learning_rate (float) – learning rate for stochastic gradient descent

  • random_seed (int) – random seed used

  • zero_abundance_weight_init (bool) – whether to use 0 as initial weight for abundance term (if not, a random value is sampled from normal

  • total_number_of_features (distribution with mean 0 and variance 1 /) –

  • number_of_threads – number of threads to be used for training

  • initialization_count (int) – how many times to repeat the fitting procedure from the beginning before choosing the optimal model (trains the model with multiple random initializations)

  • pytorch_device_name (str) – The name of the pytorch device to use. This name will be passed to torch.device(pytorch_device_name).

YAML specification:

my_kmer_mil_classifier:
    AtchleyKmerMILClassifier:
        iteration_count: 100
        evaluate_at: 15
        use_early_stopping: False
        learning_rate: 0.01
        random_seed: 100
        zero_abundance_weight_init: True
        number_of_threads: 8
        threshold: 0.00001
        initialization_count: 4
MAX_SEED_VALUE = 100000
MIN_SEED_VALUE = 1
can_predict_proba() bool[source]

Returns whether the ML model can be used to predict class probabilities or class assignment only.

check_if_exists(path) bool[source]

The check_if_exists function checks if there is a stored model on the given path. Might be useful in the future for implementing checkpoints. See SklearnMethod for example usage.

Parameters:

path (Path) – path to folder where it should be checked if the model was stored previously

Returns:

whether the model was stored previously on the given path or not

Return type:

True/False

fit(encoded_data: EncodedData, label: Label, cores_for_training: int = 2)[source]

The fit function fits the parameters of the machine learning model.

Parameters:
  • encoded_data (EncodedData) – an instance of EncodedData class which includes encoded examples (repertoires, receptors or sequences), their labels, names of the features and other additional information. Most often, only examples and labels will be used. Examples are either a dense numpy matrix or a sparse matrix, where columns correspond to features and rows correspond to examples. There are a few encodings which make multidimensional outputs that do not follow this pattern, but they are tailored to specific ML methods which require such input (for instance, one hot encoding and ReceptorCNN method).

  • label (Label) – the label for which the classifier will be created. immuneML also supports multi-label classification, but it is handled outside MLMethod class by creating an MLMethod instance for each label. This means that each MLMethod should handle only one label.

  • cores_for_training (int) – if parallelization is available in the MLMethod (and the availability depends on the specific classifier), this is the number of processes that will be creating when fitting the model to speed up the computation.

Returns:

it doesn’t return anything, but fits the model parameters instead

fit_by_cross_validation(encoded_data: EncodedData, number_of_splits: int = 5, label: Label = None, cores_for_training: int = -1, optimization_metric=None)[source]

The fit_by_cross_validation function should implement finding the best model hyperparameters through cross-validation. In immuneML, preprocessing, encoding and ML hyperparameters can be optimized by using nested cross-validation (see TrainMLModelInstruction for more details). This function is in that setting the third level of nested cross-validation as it can optimize only over the model hyperparameters. It represents an alternative to optimizing the model hyperparameters in the TrainMLModelInstruction. Which one should be used depends on the use-case and specific models: models based on scikit-learn implementations come with this option by default (see SklearnMethod class), while custom classifiers typically do not implement this and just call fit() function and throw a warning instead.

Parameters:
  • encoded_data (EncodedData) – an instance of EncodedData class which includes encoded examples (repertoires, receptors or sequences), their labels, names of the features and other additional information. Most often, only examples and labels will be used. Examples are either a dense numpy matrix or a sparse matrix, where columns correspond to features and rows correspond to examples. There are a few encodings which make multidimensional outputs that do not follow this pattern, but they are tailored to specific ML methods which require such input (for instance, one hot encoding and ReceptorCNN method).

  • number_of_splits (int) – number of splits for the cross-validation to be performed for selection the best hyperparameters of the ML model; note that if this is used in combination with nested cross-validation in TrainMLModel instruction, it can result in very few examples in each split depending on the orginal dataset size and the nested cross-validation setup.

  • label (Label) – the label for which the classifier will be created. immuneML also supports multi-label classification, but it is handled outside MLMethod class by creating an MLMethod instance for each label. This means that each MLMethod should handle only one label.

  • cores_for_training (int) – number of processes to be used during the cross-validation for model selection

  • optimization_metric (str) – the name of the optimization metric to be used to select the best model during cross-validation; when used with TrainMLModel instruction which is almost exclusively the case when the immuneML is run from the specification, this maps to the optimization metric in the instruction.

Returns:

it doesn’t return anything, but fits the model parameters instead

get_class_mapping() dict[source]

Returns a dictionary containing the mapping between label values and values internally used in the classifier

get_compatible_encoders()[source]
get_feature_names() list[source]

Returns the list of feature names (a list of strings) if available where the feature names were provided by the encoder in the EncodedData object.

get_label_name()[source]

Returns the name of the label for which the model was fitted.

get_package_info() str[source]

Returns the package and version used for implementing the ML method if an external package was used or immuneML version if it is custom implementation. See py:mod:immuneML.ml_methods.SklearnMethod.SklearnMethod and py:mod:immuneML.ml_methods.ProbabilisticBinaryClassifier.ProbabilisticBinaryClassifier for examples of both.

get_params()[source]

Returns the model parameters in a readable yaml-friendly way (consisting of lists, dictionaries and strings).

load(path)[source]

The load function can load the model given the folder where the same class of the model was previously stored using the store function. It reads in the parameters of the model and sets the values to the object attributes so that the model can be reused. For instance, this is used in MLApplication instruction when the previously trained model is applied on a new dataset.

Parameters:

path (Path) – path to the folder where the model was stored using store() function

Returns:

it does not have a return value, but sets the attribute values of the object instead

predict(encoded_data: EncodedData, label: Label)[source]

The predict function predicts the class for the given label across examples provided in encoded data.

Parameters:
  • encoded_data (EncodedData) – an instance of EncodedData class which includes encoded examples (repertoires, receptors or sequences), their labels, names of the features and other additional information. Most often, only examples and labels will be used. Examples are either a dense numpy matrix or a sparse matrix, where columns correspond to features and rows correspond to examples. There are a few encodings which make multidimensional outputs that do not follow this pattern, but they are tailored to specific ML methods which require such input (for instance, one hot encoding and ReceptorCNN method).

  • label – (Label): the label for which the classifier will be created. immuneML also supports multi-label classification, but it is handled outside MLMethod class by creating an MLMethod instance for each label. This means that each MLMethod should handle only one label.

Returns:

e.g., {label_name: [class1, class2, class2, class1]}

Return type:

a dictionary where the key is the label_name and the value is a list of class predictions (one prediction per example)

predict_proba(encoded_data: EncodedData, label: Label)[source]

The predict_proba function predicts class probabilities for the given label if the model supports probabilistic output. If not, it should raise a warning and return predicted classes without probabilities.

The function will return a nested dictionary. The key(s) of the outer dictionary represent the label name(s), and the keys of the inner dictionary the class names of the respective label. The utility function py:mod:immuneML.ml_methods.util.Util.Util.make_binary_class_mapping may be used to handle mapping of class names to an internal representation for binary classification.

Parameters:
  • encoded_data (EncodedData) – an object of EncodedData class where the examples attribute should be used to make predictions. examples

  • format (attribute includes encoded examples in matrix) –

  • labels (provided here can include) –

  • labels) (been) –

  • function (so the labels attribute of the EncodedData object should NOT be used in this) –

  • set. (even if it is) –

  • label (Label) – the label for which the prediction should be made. It can be used to check if it matches the label that the

  • not (model has been trained for and if) –

  • any (an exception should be thrown. It is often an AssertionError as this can be checked before) –

  • made (prediction is) –

  • cases (but could also be a RuntimeError. It both) –

  • message. (it should include a user-friendly) –

Returns:

a nested dictionary where the outer keys represent label names, inner keys represent class names for the respective label, and innermost values are 1D numpy arrays with class probabilities. For example for instance for label CMV where the class can be either True or False and there are 3 examples to predict the class probabilities for: {CMV: {True: [0.2, 0.55, 0.98], False: [0.8, 0.45, 0.02]}}

store(path: Path, feature_names=None, details_path: Path = None)[source]

The store function stores the object on which it is called so that it can be imported later using load function. It typically uses pickle, yaml or similar modules to store the information. It can store one or multiple files.

Parameters:
  • path (Path) – path to folder where to store the model

  • feature_names (list) – list of feature names in the encoded data; this can be stored as well to make it easier to map linear models to specific features as provided by the encoded (e.g., in case of logistic regression, this feature list defines what coefficients refer to)

  • details_path (Path) – path to folder where to store the details of the model. The details can be there to better understand the model but are not mandatory and are typically not loaded with the model afterwards. This is user-friendly file that can be examined manually by the user. It does not have to be created or can be created at the same folder as the path parameters points to. In practice, when used with TrainMLModel instruction, this parameter will either be None or have the same value as path parameter.

Returns:

it does not have a return value

immuneML.ml_methods.DeepRC module

class immuneML.ml_methods.DeepRC.DeepRC(validation_part, add_positional_information, kernel_size, n_kernels, n_additional_convs, n_attention_network_layers, n_attention_network_units, n_output_network_units, consider_seq_counts, sequence_reduction_fraction, reduction_mb_size, n_updates, n_torch_threads, learning_rate, l1_weight_decay, l2_weight_decay, evaluate_at, sample_n_sequences, training_batch_size, n_workers, sequence_counts_scaling_fn, keep_dataset_in_ram, pytorch_device_name)[source]

Bases: MLMethod

This classifier uses the DeepRC method for repertoire classification. The DeepRC ML method should be used in combination with the DeepRC encoder. Also consider using the DeepRCMotifDiscovery report for interpretability.

Notes:

  • DeepRC uses PyTorch functionalities that depend on GPU. Therefore, DeepRC does not work on a CPU.

  • This wrapper around DeepRC currently only supports binary classification.

Reference: Michael Widrich, Bernhard Schäfl, Milena Pavlović, Geir Kjetil Sandve, Sepp Hochreiter, Victor Greiff, Günter Klambauer ‘DeepRC: Immune repertoire classification with attention-based deep massive multiple instance learning’. bioRxiv preprint doi: https://doi.org/10.1101/2020.04.12.038158

Parameters:
  • validation_part (float) – the part of the data that will be used for validation, the rest will be used for training.

  • add_positional_information (bool) – whether positional information should be included in the input features.

  • kernel_size (int) – the size of the 1D-CNN kernels.

  • n_kernels (int) – the number of 1D-CNN kernels in each layer.

  • n_additional_convs (int) – Number of additional 1D-CNN layers after first layer

  • n_attention_network_layers (int) – Number of attention layers to compute keys

  • n_attention_network_units (int) – Number of units in each attention layer

  • n_output_network_units (int) – Number of units in the output layer

  • consider_seq_counts (bool) – whether the input data should be scaled by the receptor sequence counts.

  • sequence_reduction_fraction (float) – Fraction of number of sequences to which to reduce the number of sequences per bag based on attention weights. Has to be in range [0,1].

  • reduction_mb_size (int) – Reduction of sequences per bag is performed using minibatches of reduction_mb_size` sequences to compute the attention weights.

  • n_updates (int) – Number of updates to train for

  • n_torch_threads (int) – Number of parallel threads to allow PyTorch

  • learning_rate (float) – Learning rate for adam optimizer

  • l1_weight_decay (float) – l1 weight decay factor. l1 weight penalty will be added to loss, scaled by l1_weight_decay

  • l2_weight_decay (float) – l2 weight decay factor. l2 weight penalty will be added to loss, scaled by l2_weight_decay

  • sequence_counts_scaling_fn – it can either be log (logarithmic scaling of sequence counts) or None

  • evaluate_at (int) – Evaluate model on training and validation set every evaluate_at updates. This will also check for a new best model for early stopping.

  • sample_n_sequences (int) – Optional random sub-sampling of sample_n_sequences sequences per repertoire. Number of sequences per repertoire might be smaller than sample_n_sequences if repertoire is smaller or random indices have been drawn multiple times. If None, all sequences will be loaded for each repertoire.

  • training_batch_size (int) – Number of repertoires per minibatch during training.

  • n_workers (int) – Number of background processes to use for converting dataset to hdf5 container and training set data loader.

  • pytorch_device_name (str) – The name of the pytorch device to use. This name will be passed to torch.device(self.pytorch_device_name). The default value is cuda:0

YAML specification:

my_deeprc_method:
    DeepRC:
        validation_part: 0.2
        add_positional_information: True
        kernel_size: 9
can_predict_proba() bool[source]

Returns whether the ML model can be used to predict class probabilities or class assignment only.

check_if_exists(path)[source]

The check_if_exists function checks if there is a stored model on the given path. Might be useful in the future for implementing checkpoints. See SklearnMethod for example usage.

Parameters:

path (Path) – path to folder where it should be checked if the model was stored previously

Returns:

whether the model was stored previously on the given path or not

Return type:

True/False

check_is_fitted(label_name: str)[source]
fit(encoded_data: EncodedData, label: Label, cores_for_training: int = 2)[source]

The fit function fits the parameters of the machine learning model.

Parameters:
  • encoded_data (EncodedData) – an instance of EncodedData class which includes encoded examples (repertoires, receptors or sequences), their labels, names of the features and other additional information. Most often, only examples and labels will be used. Examples are either a dense numpy matrix or a sparse matrix, where columns correspond to features and rows correspond to examples. There are a few encodings which make multidimensional outputs that do not follow this pattern, but they are tailored to specific ML methods which require such input (for instance, one hot encoding and ReceptorCNN method).

  • label (Label) – the label for which the classifier will be created. immuneML also supports multi-label classification, but it is handled outside MLMethod class by creating an MLMethod instance for each label. This means that each MLMethod should handle only one label.

  • cores_for_training (int) – if parallelization is available in the MLMethod (and the availability depends on the specific classifier), this is the number of processes that will be creating when fitting the model to speed up the computation.

Returns:

it doesn’t return anything, but fits the model parameters instead

fit_by_cross_validation(encoded_data: EncodedData, number_of_splits: int = 5, label: Label = None, cores_for_training: int = -1, optimization_metric=None)[source]

The fit_by_cross_validation function should implement finding the best model hyperparameters through cross-validation. In immuneML, preprocessing, encoding and ML hyperparameters can be optimized by using nested cross-validation (see TrainMLModelInstruction for more details). This function is in that setting the third level of nested cross-validation as it can optimize only over the model hyperparameters. It represents an alternative to optimizing the model hyperparameters in the TrainMLModelInstruction. Which one should be used depends on the use-case and specific models: models based on scikit-learn implementations come with this option by default (see SklearnMethod class), while custom classifiers typically do not implement this and just call fit() function and throw a warning instead.

Parameters:
  • encoded_data (EncodedData) – an instance of EncodedData class which includes encoded examples (repertoires, receptors or sequences), their labels, names of the features and other additional information. Most often, only examples and labels will be used. Examples are either a dense numpy matrix or a sparse matrix, where columns correspond to features and rows correspond to examples. There are a few encodings which make multidimensional outputs that do not follow this pattern, but they are tailored to specific ML methods which require such input (for instance, one hot encoding and ReceptorCNN method).

  • number_of_splits (int) – number of splits for the cross-validation to be performed for selection the best hyperparameters of the ML model; note that if this is used in combination with nested cross-validation in TrainMLModel instruction, it can result in very few examples in each split depending on the orginal dataset size and the nested cross-validation setup.

  • label (Label) – the label for which the classifier will be created. immuneML also supports multi-label classification, but it is handled outside MLMethod class by creating an MLMethod instance for each label. This means that each MLMethod should handle only one label.

  • cores_for_training (int) – number of processes to be used during the cross-validation for model selection

  • optimization_metric (str) – the name of the optimization metric to be used to select the best model during cross-validation; when used with TrainMLModel instruction which is almost exclusively the case when the immuneML is run from the specification, this maps to the optimization metric in the instruction.

Returns:

it doesn’t return anything, but fits the model parameters instead

get_class_mapping() dict[source]

Returns a dictionary containing the mapping between label values and values internally used in the classifier

get_compatible_encoders()[source]
get_feature_names() list[source]

Returns the list of feature names (a list of strings) if available where the feature names were provided by the encoder in the EncodedData object.

get_label_name() str[source]

Returns the name of the label for which the model was fitted.

get_package_info() str[source]

Returns the package and version used for implementing the ML method if an external package was used or immuneML version if it is custom implementation. See py:mod:immuneML.ml_methods.SklearnMethod.SklearnMethod and py:mod:immuneML.ml_methods.ProbabilisticBinaryClassifier.ProbabilisticBinaryClassifier for examples of both.

get_params()[source]

Returns the model parameters in a readable yaml-friendly way (consisting of lists, dictionaries and strings).

load(path: Path, details_path: Path = None)[source]

The load function can load the model given the folder where the same class of the model was previously stored using the store function. It reads in the parameters of the model and sets the values to the object attributes so that the model can be reused. For instance, this is used in MLApplication instruction when the previously trained model is applied on a new dataset.

Parameters:

path (Path) – path to the folder where the model was stored using store() function

Returns:

it does not have a return value, but sets the attribute values of the object instead

make_data_loader(full_dataset, indices, label_name, eval_only: bool, is_train: bool, n_workers=1)[source]

Creates a pytorch dataloader using DeepRC’s RepertoireDataReaderBinary

Parameters:
  • hdf5_filepath – the path to the HDF5 file

  • pre_loaded_hdf5_file – Optional: It is faster to load the hdf5 file into the RAM as dictionary instead of keeping it on the disk. pre_loaded_hdf5_file is the loaded hdf5 file as dictionary. If None, the hdf5 file will be read from the disk and consume less RAM.

  • indices – indices of the subset of repertoires in the data that will be used for this dataset. If ‘None’, all repertoires will be used.

  • label_name – the name of the label to be predicted

  • eval_only – whether the dataloader will only be used for evaluation (no training). if false, sample_n_sequences can be set

  • is_train – whether this is a dataloader for training data. If true, self.training_batch_size is used.

  • n_workers – the number of workers used in torch.utils.data.DataLoader

Returns:

a Pytorch dataloader

predict(encoded_data: EncodedData, label: Label)[source]

The predict function predicts the class for the given label across examples provided in encoded data.

Parameters:
  • encoded_data (EncodedData) – an instance of EncodedData class which includes encoded examples (repertoires, receptors or sequences), their labels, names of the features and other additional information. Most often, only examples and labels will be used. Examples are either a dense numpy matrix or a sparse matrix, where columns correspond to features and rows correspond to examples. There are a few encodings which make multidimensional outputs that do not follow this pattern, but they are tailored to specific ML methods which require such input (for instance, one hot encoding and ReceptorCNN method).

  • label – (Label): the label for which the classifier will be created. immuneML also supports multi-label classification, but it is handled outside MLMethod class by creating an MLMethod instance for each label. This means that each MLMethod should handle only one label.

Returns:

e.g., {label_name: [class1, class2, class2, class1]}

Return type:

a dictionary where the key is the label_name and the value is a list of class predictions (one prediction per example)

predict_proba(encoded_data: EncodedData, label: Label)[source]

The predict_proba function predicts class probabilities for the given label if the model supports probabilistic output. If not, it should raise a warning and return predicted classes without probabilities.

The function will return a nested dictionary. The key(s) of the outer dictionary represent the label name(s), and the keys of the inner dictionary the class names of the respective label. The utility function py:mod:immuneML.ml_methods.util.Util.Util.make_binary_class_mapping may be used to handle mapping of class names to an internal representation for binary classification.

Parameters:
  • encoded_data (EncodedData) – an object of EncodedData class where the examples attribute should be used to make predictions. examples

  • format (attribute includes encoded examples in matrix) –

  • labels (provided here can include) –

  • labels) (been) –

  • function (so the labels attribute of the EncodedData object should NOT be used in this) –

  • set. (even if it is) –

  • label (Label) – the label for which the prediction should be made. It can be used to check if it matches the label that the

  • not (model has been trained for and if) –

  • any (an exception should be thrown. It is often an AssertionError as this can be checked before) –

  • made (prediction is) –

  • cases (but could also be a RuntimeError. It both) –

  • message. (it should include a user-friendly) –

Returns:

a nested dictionary where the outer keys represent label names, inner keys represent class names for the respective label, and innermost values are 1D numpy arrays with class probabilities. For example for instance for label CMV where the class can be either True or False and there are 3 examples to predict the class probabilities for: {CMV: {True: [0.2, 0.55, 0.98], False: [0.8, 0.45, 0.02]}}

store(path, feature_names=None, details_path: Path = None)[source]

The store function stores the object on which it is called so that it can be imported later using load function. It typically uses pickle, yaml or similar modules to store the information. It can store one or multiple files.

Parameters:
  • path (Path) – path to folder where to store the model

  • feature_names (list) – list of feature names in the encoded data; this can be stored as well to make it easier to map linear models to specific features as provided by the encoded (e.g., in case of logistic regression, this feature list defines what coefficients refer to)

  • details_path (Path) – path to folder where to store the details of the model. The details can be there to better understand the model but are not mandatory and are typically not loaded with the model afterwards. This is user-friendly file that can be examined manually by the user. It does not have to be created or can be created at the same folder as the path parameters points to. In practice, when used with TrainMLModel instruction, this parameter will either be None or have the same value as path parameter.

Returns:

it does not have a return value

immuneML.ml_methods.KNN module

class immuneML.ml_methods.KNN.KNN(parameter_grid: dict = None, parameters: dict = None)[source]

Bases: SklearnMethod

This is a wrapper of scikit-learn’s KNeighborsClassifier class. This ML method creates a distance matrix using the given encoded data. If the encoded data is already a distance matrix (for example, when using the Distance or CompAIRRDistance encoders), please use PrecomputedKNN instead.

Please see the scikit-learn documentation of KNeighborsClassifier for the parameters.

For usage instructions, check SklearnMethod.

YAML specification:

my_knn_method:
    KNN:
        # sklearn parameters (same names as in original sklearn class)
        weights: uniform # always use this setting for weights
        n_neighbors: [5, 10, 15] # find the optimal number of neighbors
        # Additional parameter that determines whether to print convergence warnings
        show_warnings: True
    # if any of the parameters under KNN is a list and model_selection_cv is True,
    # a grid search will be done over the given parameters, using the number of folds specified in model_selection_n_folds,
    # and the optimal model will be selected
    model_selection_cv: True
    model_selection_n_folds: 5
# alternative way to define ML method with default values:
my_default_knn: KNN
can_predict_proba() bool[source]

Returns whether the ML model can be used to predict class probabilities or class assignment only.

get_compatible_encoders()[source]
static get_documentation()[source]
get_params()[source]

Returns the model parameters in a readable yaml-friendly way (consisting of lists, dictionaries and strings).

immuneML.ml_methods.LogisticRegression module

class immuneML.ml_methods.LogisticRegression.LogisticRegression(parameter_grid: dict = None, parameters: dict = None)[source]

Bases: SklearnMethod

This is a wrapper of scikit-learn’s LogisticRegression class. Please see the scikit-learn documentation of LogisticRegression for the parameters.

Note: if you are interested in plotting the coefficients of the logistic regression model, consider running the Coefficients report.

For usage instructions, check SklearnMethod.

YAML specification:

my_logistic_regression: # user-defined method name
    LogisticRegression: # name of the ML method
        # sklearn parameters (same names as in original sklearn class)
        penalty: l1 # always use penalty l1
        C: [0.01, 0.1, 1, 10, 100] # find the optimal value for C
        # Additional parameter that determines whether to print convergence warnings
        show_warnings: True
    # if any of the parameters under LogisticRegression is a list and model_selection_cv is True,
    # a grid search will be done over the given parameters, using the number of folds specified in model_selection_n_folds,
    # and the optimal model will be selected
    model_selection_cv: True
    model_selection_n_folds: 5
# alternative way to define ML method with default values:
my_default_logistic_regression: LogisticRegression
can_predict_proba() bool[source]

Returns whether the ML model can be used to predict class probabilities or class assignment only.

default_parameters = {'max_iter': 1000, 'solver': 'saga'}
static get_documentation()[source]
get_params()[source]

Returns the model parameters in a readable yaml-friendly way (consisting of lists, dictionaries and strings).

immuneML.ml_methods.MLMethod module

class immuneML.ml_methods.MLMethod.MLMethod[source]

Bases: object

Base class for different machine learning methods, defining which functions should be implemented. These public functions are the only ones that will be used outside the method, during training, assessment or while making predictions. Most often the methods will be classifiers (binary or multi-class) that should learn some label on either immune repertoires (sets of receptor sequences), receptors (paired sequences) or receptor sequences (lists of amino acids).

Here we refer to machine learning methods (algorithms) as a method that, given a set of examples and corresponding labels, constructs a model (such as logistic regression), whereas we define the model to be already fit to data using the learning method (algorithm), such as logistic regression with specific coefficients.

The functions of this class provide a standard set of ML functions: fitting the model (with or without cross-validation) and making predictions (either class predictions or class probabilities if possible). Other functions provide for various utilities, such as storing and loading the model, checking if it was fit already, retrieving coefficients for user-friendly output etc.

Note that when providing class probabilities the classes should have a specific (constant) order, and in case of binary classification, they should be ordered so that the negative class comes first and the positive one comes second. For this handling classes, see py:immuneML.ml_methods.util.Util.Util.make_class_mapping method that will automatically create class mapping for classification.

abstract can_predict_proba() bool[source]

Returns whether the ML model can be used to predict class probabilities or class assignment only.

check_encoder_compatibility(encoder)[source]

Checks whether the given encoder is compatible with this ML method, and throws an error if it is not.

abstract check_if_exists(path: Path) bool[source]

The check_if_exists function checks if there is a stored model on the given path. Might be useful in the future for implementing checkpoints. See SklearnMethod for example usage.

Parameters:

path (Path) – path to folder where it should be checked if the model was stored previously

Returns:

whether the model was stored previously on the given path or not

Return type:

True/False

abstract fit(encoded_data: ~immuneML.data_model.encoded_data.EncodedData.EncodedData, label: <module 'immuneML.environment.Label' from '/Users/milenpa/PycharmProjects/BMIimmuneML/immuneML/environment/Label.py'>, cores_for_training: int = 2)[source]

The fit function fits the parameters of the machine learning model.

Parameters:
  • encoded_data (EncodedData) – an instance of EncodedData class which includes encoded examples (repertoires, receptors or sequences), their labels, names of the features and other additional information. Most often, only examples and labels will be used. Examples are either a dense numpy matrix or a sparse matrix, where columns correspond to features and rows correspond to examples. There are a few encodings which make multidimensional outputs that do not follow this pattern, but they are tailored to specific ML methods which require such input (for instance, one hot encoding and ReceptorCNN method).

  • label (Label) – the label for which the classifier will be created. immuneML also supports multi-label classification, but it is handled outside MLMethod class by creating an MLMethod instance for each label. This means that each MLMethod should handle only one label.

  • cores_for_training (int) – if parallelization is available in the MLMethod (and the availability depends on the specific classifier), this is the number of processes that will be creating when fitting the model to speed up the computation.

Returns:

it doesn’t return anything, but fits the model parameters instead

abstract fit_by_cross_validation(encoded_data: ~immuneML.data_model.encoded_data.EncodedData.EncodedData, number_of_splits: int = 5, label: <module 'immuneML.environment.Label' from '/Users/milenpa/PycharmProjects/BMIimmuneML/immuneML/environment/Label.py'> = None, cores_for_training: int = -1, optimization_metric=None)[source]

The fit_by_cross_validation function should implement finding the best model hyperparameters through cross-validation. In immuneML, preprocessing, encoding and ML hyperparameters can be optimized by using nested cross-validation (see TrainMLModelInstruction for more details). This function is in that setting the third level of nested cross-validation as it can optimize only over the model hyperparameters. It represents an alternative to optimizing the model hyperparameters in the TrainMLModelInstruction. Which one should be used depends on the use-case and specific models: models based on scikit-learn implementations come with this option by default (see SklearnMethod class), while custom classifiers typically do not implement this and just call fit() function and throw a warning instead.

Parameters:
  • encoded_data (EncodedData) – an instance of EncodedData class which includes encoded examples (repertoires, receptors or sequences), their labels, names of the features and other additional information. Most often, only examples and labels will be used. Examples are either a dense numpy matrix or a sparse matrix, where columns correspond to features and rows correspond to examples. There are a few encodings which make multidimensional outputs that do not follow this pattern, but they are tailored to specific ML methods which require such input (for instance, one hot encoding and ReceptorCNN method).

  • number_of_splits (int) – number of splits for the cross-validation to be performed for selection the best hyperparameters of the ML model; note that if this is used in combination with nested cross-validation in TrainMLModel instruction, it can result in very few examples in each split depending on the orginal dataset size and the nested cross-validation setup.

  • label (Label) – the label for which the classifier will be created. immuneML also supports multi-label classification, but it is handled outside MLMethod class by creating an MLMethod instance for each label. This means that each MLMethod should handle only one label.

  • cores_for_training (int) – number of processes to be used during the cross-validation for model selection

  • optimization_metric (str) – the name of the optimization metric to be used to select the best model during cross-validation; when used with TrainMLModel instruction which is almost exclusively the case when the immuneML is run from the specification, this maps to the optimization metric in the instruction.

Returns:

it doesn’t return anything, but fits the model parameters instead

abstract get_class_mapping() dict[source]

Returns a dictionary containing the mapping between label values and values internally used in the classifier

get_classes()[source]

The get_classes function returns a list of classes for which the method was trained.

abstract get_compatible_encoders()[source]
abstract get_feature_names() list[source]

Returns the list of feature names (a list of strings) if available where the feature names were provided by the encoder in the EncodedData object.

abstract get_label_name() str[source]

Returns the name of the label for which the model was fitted.

abstract get_package_info() str[source]

Returns the package and version used for implementing the ML method if an external package was used or immuneML version if it is custom implementation. See py:mod:immuneML.ml_methods.SklearnMethod.SklearnMethod and py:mod:immuneML.ml_methods.ProbabilisticBinaryClassifier.ProbabilisticBinaryClassifier for examples of both.

abstract get_params()[source]

Returns the model parameters in a readable yaml-friendly way (consisting of lists, dictionaries and strings).

get_positive_class()[source]

The get_positive_class function returns the positive class for which the method was trained.

abstract load(path: Path)[source]

The load function can load the model given the folder where the same class of the model was previously stored using the store function. It reads in the parameters of the model and sets the values to the object attributes so that the model can be reused. For instance, this is used in MLApplication instruction when the previously trained model is applied on a new dataset.

Parameters:

path (Path) – path to the folder where the model was stored using store() function

Returns:

it does not have a return value, but sets the attribute values of the object instead

abstract predict(encoded_data: ~immuneML.data_model.encoded_data.EncodedData.EncodedData, label: <module 'immuneML.environment.Label' from '/Users/milenpa/PycharmProjects/BMIimmuneML/immuneML/environment/Label.py'>)[source]

The predict function predicts the class for the given label across examples provided in encoded data.

Parameters:
  • encoded_data (EncodedData) – an instance of EncodedData class which includes encoded examples (repertoires, receptors or sequences), their labels, names of the features and other additional information. Most often, only examples and labels will be used. Examples are either a dense numpy matrix or a sparse matrix, where columns correspond to features and rows correspond to examples. There are a few encodings which make multidimensional outputs that do not follow this pattern, but they are tailored to specific ML methods which require such input (for instance, one hot encoding and ReceptorCNN method).

  • label – (Label): the label for which the classifier will be created. immuneML also supports multi-label classification, but it is handled outside MLMethod class by creating an MLMethod instance for each label. This means that each MLMethod should handle only one label.

Returns:

e.g., {label_name: [class1, class2, class2, class1]}

Return type:

a dictionary where the key is the label_name and the value is a list of class predictions (one prediction per example)

abstract predict_proba(encoded_data: ~immuneML.data_model.encoded_data.EncodedData.EncodedData, Label: <module 'immuneML.environment.Label' from '/Users/milenpa/PycharmProjects/BMIimmuneML/immuneML/environment/Label.py'>)[source]

The predict_proba function predicts class probabilities for the given label if the model supports probabilistic output. If not, it should raise a warning and return predicted classes without probabilities.

The function will return a nested dictionary. The key(s) of the outer dictionary represent the label name(s), and the keys of the inner dictionary the class names of the respective label. The utility function py:mod:immuneML.ml_methods.util.Util.Util.make_binary_class_mapping may be used to handle mapping of class names to an internal representation for binary classification.

Parameters:
  • encoded_data (EncodedData) – an object of EncodedData class where the examples attribute should be used to make predictions. examples

  • format (attribute includes encoded examples in matrix) –

  • labels (provided here can include) –

  • labels) (been) –

  • function (so the labels attribute of the EncodedData object should NOT be used in this) –

  • set. (even if it is) –

  • label (Label) – the label for which the prediction should be made. It can be used to check if it matches the label that the

  • not (model has been trained for and if) –

  • any (an exception should be thrown. It is often an AssertionError as this can be checked before) –

  • made (prediction is) –

  • cases (but could also be a RuntimeError. It both) –

  • message. (it should include a user-friendly) –

Returns:

a nested dictionary where the outer keys represent label names, inner keys represent class names for the respective label, and innermost values are 1D numpy arrays with class probabilities. For example for instance for label CMV where the class can be either True or False and there are 3 examples to predict the class probabilities for: {CMV: {True: [0.2, 0.55, 0.98], False: [0.8, 0.45, 0.02]}}

abstract store(path: Path, feature_names: list = None, details_path: Path = None)[source]

The store function stores the object on which it is called so that it can be imported later using load function. It typically uses pickle, yaml or similar modules to store the information. It can store one or multiple files.

Parameters:
  • path (Path) – path to folder where to store the model

  • feature_names (list) – list of feature names in the encoded data; this can be stored as well to make it easier to map linear models to specific features as provided by the encoded (e.g., in case of logistic regression, this feature list defines what coefficients refer to)

  • details_path (Path) – path to folder where to store the details of the model. The details can be there to better understand the model but are not mandatory and are typically not loaded with the model afterwards. This is user-friendly file that can be examined manually by the user. It does not have to be created or can be created at the same folder as the path parameters points to. In practice, when used with TrainMLModel instruction, this parameter will either be None or have the same value as path parameter.

Returns:

it does not have a return value

immuneML.ml_methods.PrecomputedKNN module

class immuneML.ml_methods.PrecomputedKNN.PrecomputedKNN(parameter_grid: dict = None, parameters: dict = None)[source]

Bases: SklearnMethod

This is a wrapper of scikit-learn’s KNeighborsClassifier class. This ML method takes a pre-computed distance matrix, as created by the Distance or CompAIRRDistance encoders. If you would like to use a different encoding in combination with KNN, please use KNN instead.

Please see the scikit-learn documentation of KNeighborsClassifier for the parameters.

For usage instructions, check SklearnMethod.

YAML specification:

my_knn_method:
    PrecomputedKNN:
        # sklearn parameters (same names as in original sklearn class)
        weights: uniform # always use this setting for weights
        n_neighbors: [5, 10, 15] # find the optimal number of neighbors
        # Additional parameter that determines whether to print convergence warnings
        show_warnings: True
    # if any of the parameters under KNN is a list and model_selection_cv is True,
    # a grid search will be done over the given parameters, using the number of folds specified in model_selection_n_folds,
    # and the optimal model will be selected
    model_selection_cv: True
    model_selection_n_folds: 5
# alternative way to define ML method with default values:
my_default_knn: PrecomputedKNN
can_predict_proba() bool[source]

Returns whether the ML model can be used to predict class probabilities or class assignment only.

get_compatible_encoders()[source]
static get_documentation()[source]
get_params()[source]

Returns the model parameters in a readable yaml-friendly way (consisting of lists, dictionaries and strings).

immuneML.ml_methods.ProbabilisticBinaryClassifier module

class immuneML.ml_methods.ProbabilisticBinaryClassifier.ProbabilisticBinaryClassifier(max_iterations: int = None, update_rate: float = None, likelihood_threshold: float = None)[source]

Bases: MLMethod

ProbabilisticBinaryClassifier predicts the class assignment in binary classification case based on encoding examples by number of successful trials and total number of trials. It models this ratio by one beta distribution per class and predicts the class of the new examples using log-posterior odds ratio with threshold at 0.

ProbabilisticBinaryClassifier is based on the paper (details on the classification can be found in the Online Methods section): Emerson, Ryan O., William S. DeWitt, Marissa Vignali, Jenna Gravley, Joyce K. Hu, Edward J. Osborne, Cindy Desmarais, et al. ‘Immunosequencing Identifies Signatures of Cytomegalovirus Exposure History and HLA-Mediated Effects on the T Cell Repertoire’. Nature Genetics 49, no. 5 (May 2017): 659–65. doi.org/10.1038/ng.3822.

Parameters:
  • max_iterations (int) – maximum number of iterations while optimizing the parameters of the beta distribution (same for both classes)

  • update_rate (float) – how much the computed gradient should influence the updated value of the parameters of the beta distribution

  • likelihood_threshold (float) – at which threshold to stop the optimization (default -1e-10)

YAML specification:

my_probabilistic_classifier: # user-defined name of the ML method
    ProbabilisticBinaryClassifier: # method name
        max_iterations: 1000
        update_rate: 0.01
SMALL_POSITIVE_NUMBER = 1e-15
can_predict_proba() bool[source]

Returns whether the ML model can be used to predict class probabilities or class assignment only.

check_if_exists(path)[source]

The check_if_exists function checks if there is a stored model on the given path. Might be useful in the future for implementing checkpoints. See SklearnMethod for example usage.

Parameters:

path (Path) – path to folder where it should be checked if the model was stored previously

Returns:

whether the model was stored previously on the given path or not

Return type:

True/False

fit(encoded_data: EncodedData, label: Label, cores_for_training: int = 2)[source]

The fit function fits the parameters of the machine learning model.

Parameters:
  • encoded_data (EncodedData) – an instance of EncodedData class which includes encoded examples (repertoires, receptors or sequences), their labels, names of the features and other additional information. Most often, only examples and labels will be used. Examples are either a dense numpy matrix or a sparse matrix, where columns correspond to features and rows correspond to examples. There are a few encodings which make multidimensional outputs that do not follow this pattern, but they are tailored to specific ML methods which require such input (for instance, one hot encoding and ReceptorCNN method).

  • label (Label) – the label for which the classifier will be created. immuneML also supports multi-label classification, but it is handled outside MLMethod class by creating an MLMethod instance for each label. This means that each MLMethod should handle only one label.

  • cores_for_training (int) – if parallelization is available in the MLMethod (and the availability depends on the specific classifier), this is the number of processes that will be creating when fitting the model to speed up the computation.

Returns:

it doesn’t return anything, but fits the model parameters instead

fit_by_cross_validation(encoded_data: EncodedData, number_of_splits: int = 5, label: Label = None, cores_for_training: int = -1, optimization_metric=None)[source]

The fit_by_cross_validation function should implement finding the best model hyperparameters through cross-validation. In immuneML, preprocessing, encoding and ML hyperparameters can be optimized by using nested cross-validation (see TrainMLModelInstruction for more details). This function is in that setting the third level of nested cross-validation as it can optimize only over the model hyperparameters. It represents an alternative to optimizing the model hyperparameters in the TrainMLModelInstruction. Which one should be used depends on the use-case and specific models: models based on scikit-learn implementations come with this option by default (see SklearnMethod class), while custom classifiers typically do not implement this and just call fit() function and throw a warning instead.

Parameters:
  • encoded_data (EncodedData) – an instance of EncodedData class which includes encoded examples (repertoires, receptors or sequences), their labels, names of the features and other additional information. Most often, only examples and labels will be used. Examples are either a dense numpy matrix or a sparse matrix, where columns correspond to features and rows correspond to examples. There are a few encodings which make multidimensional outputs that do not follow this pattern, but they are tailored to specific ML methods which require such input (for instance, one hot encoding and ReceptorCNN method).

  • number_of_splits (int) – number of splits for the cross-validation to be performed for selection the best hyperparameters of the ML model; note that if this is used in combination with nested cross-validation in TrainMLModel instruction, it can result in very few examples in each split depending on the orginal dataset size and the nested cross-validation setup.

  • label (Label) – the label for which the classifier will be created. immuneML also supports multi-label classification, but it is handled outside MLMethod class by creating an MLMethod instance for each label. This means that each MLMethod should handle only one label.

  • cores_for_training (int) – number of processes to be used during the cross-validation for model selection

  • optimization_metric (str) – the name of the optimization metric to be used to select the best model during cross-validation; when used with TrainMLModel instruction which is almost exclusively the case when the immuneML is run from the specification, this maps to the optimization metric in the instruction.

Returns:

it doesn’t return anything, but fits the model parameters instead

get_class_mapping() dict[source]

Returns a dictionary containing the mapping between label values and values internally used in the classifier

get_classes() list[source]

The get_classes function returns a list of classes for which the method was trained.

get_compatible_encoders()[source]
get_feature_names() list[source]

Returns the list of feature names (a list of strings) if available where the feature names were provided by the encoder in the EncodedData object.

get_label_name()[source]

Returns the name of the label for which the model was fitted.

get_package_info() str[source]

Returns the package and version used for implementing the ML method if an external package was used or immuneML version if it is custom implementation. See py:mod:immuneML.ml_methods.SklearnMethod.SklearnMethod and py:mod:immuneML.ml_methods.ProbabilisticBinaryClassifier.ProbabilisticBinaryClassifier for examples of both.

get_params()[source]

Returns the model parameters in a readable yaml-friendly way (consisting of lists, dictionaries and strings).

load(path: Path)[source]

The load function can load the model given the folder where the same class of the model was previously stored using the store function. It reads in the parameters of the model and sets the values to the object attributes so that the model can be reused. For instance, this is used in MLApplication instruction when the previously trained model is applied on a new dataset.

Parameters:

path (Path) – path to the folder where the model was stored using store() function

Returns:

it does not have a return value, but sets the attribute values of the object instead

predict(encoded_data: EncodedData, label: Label)[source]

Predict the class assignment for examples in X (where X is validation or test set - examples not seen during training).

\[\begin{split}\widehat{c} \, (k, n) = \left\{\begin{matrix} 0, & F(k, n) \leq 0\\ 1, & F(k, n) > 0 \end{matrix}\right\end{split}\]
Parameters:
  • encoded_data (EncodedData) – EncodedData object with examples attribute which is a design matrix of shape

  • features] ([number of examples x number of) –

  • sequences (where number of features is 2 (the first feature is the number of disease-associated) –

  • example) (and the second is the total number of sequences per) –

  • label (str) – the label used for classification (e.g. CMV)

Returns:

{label_name: predictions} where predictions is a list of predicted classes for each example

Return type:

a dictionary of the following format

predict_proba(encoded_data: EncodedData, label: Label)[source]

Predict the probability of the class for examples in X.

\[\widehat{c} \, (k, n) = '\left\{\begin{matrix} 0, & F(k, n) \leq 0\ 1, & F(k, n) > 0 \end{matrix}\right\]
Parameters:
  • encoded_data (EncodedData) – EncodedData object with examples attribute which is a design matrix of shape, where number of features is 2

  • example) ((the first feature is the number of disease-associated sequences and the second is the total number of sequences per) –

  • label (str) – the label used for classification (e.g. CMV)

Returns:

class probabilities for all examples in X

store(path: Path, feature_names=None, details_path=None)[source]

The store function stores the object on which it is called so that it can be imported later using load function. It typically uses pickle, yaml or similar modules to store the information. It can store one or multiple files.

Parameters:
  • path (Path) – path to folder where to store the model

  • feature_names (list) – list of feature names in the encoded data; this can be stored as well to make it easier to map linear models to specific features as provided by the encoded (e.g., in case of logistic regression, this feature list defines what coefficients refer to)

  • details_path (Path) – path to folder where to store the details of the model. The details can be there to better understand the model but are not mandatory and are typically not loaded with the model afterwards. This is user-friendly file that can be examined manually by the user. It does not have to be created or can be created at the same folder as the path parameters points to. In practice, when used with TrainMLModel instruction, this parameter will either be None or have the same value as path parameter.

Returns:

it does not have a return value

immuneML.ml_methods.RandomForestClassifier module

class immuneML.ml_methods.RandomForestClassifier.RandomForestClassifier(parameter_grid: dict = None, parameters: dict = None)[source]

Bases: SklearnMethod

This is a wrapper of scikit-learn’s RandomForestClassifier class. Please see the scikit-learn documentation of RandomForestClassifier for the parameters.

Note: if you are interested in plotting the coefficients of the random forest classifier model, consider running the Coefficients report.

For usage instructions, check SklearnMethod.

YAML specification:

my_random_forest_classifier: # user-defined method name
    RandomForestClassifier: # name of the ML method
        # sklearn parameters (same names as in original sklearn class)
        random_state: 100 # always use this value for random state
        n_estimators: [10, 50, 100] # find the optimal number of trees in the forest
        # Additional parameter that determines whether to print convergence warnings
        show_warnings: True
    # if any of the parameters under RandomForestClassifier is a list and model_selection_cv is True,
    # a grid search will be done over the given parameters, using the number of folds specified in model_selection_n_folds,
    # and the optimal model will be selected
    model_selection_cv: True
    model_selection_n_folds: 5
# alternative way to define ML method with default values:
my_default_random_forest: RandomForestClassifier
can_predict_proba() bool[source]

Returns whether the ML model can be used to predict class probabilities or class assignment only.

static get_documentation()[source]
get_params()[source]

Returns the model parameters in a readable yaml-friendly way (consisting of lists, dictionaries and strings).

immuneML.ml_methods.ReceptorCNN module

class immuneML.ml_methods.ReceptorCNN.ReceptorCNN(kernel_count: int = None, kernel_size=None, positional_channels: int = None, sequence_type: str = None, device=None, number_of_threads: int = None, random_seed: int = None, learning_rate: float = None, iteration_count: int = None, l1_weight_decay: float = None, l2_weight_decay: float = None, batch_size: int = None, training_percentage: float = None, evaluate_at: int = None, background_probabilities=None, result_path: Path = None)[source]

Bases: MLMethod

A CNN which separately detects motifs using CNN kernels in each chain of paired receptor data, combines the kernel activations into a unique representation of the receptor and uses this representation to predict the antigen binding.

_images/receptor_cnn_immuneML.png

The architecture of the CNN for paired-chain receptor data

Requires one-hot encoded data as input (as produced by OneHot encoder), where use_positional_info must be set to True.

Notes:

  • ReceptorCNN can only be used with ReceptorDatasets, it does not work with SequenceDatasets

  • ReceptorCNN can only be used for binary classification, not multi-class classification.

Parameters:
  • kernel_count (count) – number of kernels that will look for motifs for one chain

  • kernel_size (list) – sizes of the kernels = how many amino acids to consider at the same time in the chain sequence, can be a tuple of values; e.g. for value [3, 4] of kernel_size, kernel_count*len(kernel_size) kernels will be created, with kernel_count kernels of size 3 and kernel_count kernels of size 4 per chain

  • positional_channels (int) – how many positional channels where included in one-hot encoding of the receptor sequences (OneHot encoder adds 3 positional channels positional information is enabled)

  • sequence_type (SequenceType) – type of the sequence

  • device – which device to use for the model (cpu or gpu) - for more details see PyTorch documentation on device parameter

  • number_of_threads (int) – how many threads to use

  • random_seed (int) – number used as a seed for random initialization

  • learning_rate (float) – learning rate scaling the step size for optimization algorithm

  • iteration_count (int) – for how many iterations to train the model

  • l1_weight_decay (float) – weight decay l1 value for the CNN; encourages sparser representations

  • l2_weight_decay (float) – weight decay l2 value for the CNN; shrinks weight coefficients towards zero

  • batch_size (int) – how many receptors to process at once

  • training_percentage (float) – what percentage of data to use for training (the rest will be used for validation); values between 0 and 1

  • evaluate_at (int) – when to evaluate the model, e.g. every 100 iterations

  • background_probabilities – used for rescaling the kernel values to produce information gain matrix; represents the background probability of each amino acid (without positional information); if not specified, uniform background is assumed

YAML specification:

my_receptor_cnn:
    ReceptorCNN:
        kernel_count: 5
        kernel_size: [3]
        positional_channels: 3
        sequence_type: amino_acid
        device: cpu
        number_of_threads: 16
        random_seed: 100
        learning_rate: 0.01
        iteration_count: 10000
        l1_weight_decay: 0
        l2_weight_decay: 0
        batch_size: 5000
can_predict_proba() bool[source]

Returns whether the ML model can be used to predict class probabilities or class assignment only.

check_encoder_compatibility(encoder)[source]

Checks whether the given encoder is compatible with this ML method, and throws an error if it is not.

check_if_exists(path)[source]

The check_if_exists function checks if there is a stored model on the given path. Might be useful in the future for implementing checkpoints. See SklearnMethod for example usage.

Parameters:

path (Path) – path to folder where it should be checked if the model was stored previously

Returns:

whether the model was stored previously on the given path or not

Return type:

True/False

fit(encoded_data: EncodedData, label: Label, cores_for_training: int = 2)[source]

The fit function fits the parameters of the machine learning model.

Parameters:
  • encoded_data (EncodedData) – an instance of EncodedData class which includes encoded examples (repertoires, receptors or sequences), their labels, names of the features and other additional information. Most often, only examples and labels will be used. Examples are either a dense numpy matrix or a sparse matrix, where columns correspond to features and rows correspond to examples. There are a few encodings which make multidimensional outputs that do not follow this pattern, but they are tailored to specific ML methods which require such input (for instance, one hot encoding and ReceptorCNN method).

  • label (Label) – the label for which the classifier will be created. immuneML also supports multi-label classification, but it is handled outside MLMethod class by creating an MLMethod instance for each label. This means that each MLMethod should handle only one label.

  • cores_for_training (int) – if parallelization is available in the MLMethod (and the availability depends on the specific classifier), this is the number of processes that will be creating when fitting the model to speed up the computation.

Returns:

it doesn’t return anything, but fits the model parameters instead

fit_by_cross_validation(encoded_data: EncodedData, number_of_splits: int = 5, label: Label = None, cores_for_training: int = -1, optimization_metric=None)[source]

The fit_by_cross_validation function should implement finding the best model hyperparameters through cross-validation. In immuneML, preprocessing, encoding and ML hyperparameters can be optimized by using nested cross-validation (see TrainMLModelInstruction for more details). This function is in that setting the third level of nested cross-validation as it can optimize only over the model hyperparameters. It represents an alternative to optimizing the model hyperparameters in the TrainMLModelInstruction. Which one should be used depends on the use-case and specific models: models based on scikit-learn implementations come with this option by default (see SklearnMethod class), while custom classifiers typically do not implement this and just call fit() function and throw a warning instead.

Parameters:
  • encoded_data (EncodedData) – an instance of EncodedData class which includes encoded examples (repertoires, receptors or sequences), their labels, names of the features and other additional information. Most often, only examples and labels will be used. Examples are either a dense numpy matrix or a sparse matrix, where columns correspond to features and rows correspond to examples. There are a few encodings which make multidimensional outputs that do not follow this pattern, but they are tailored to specific ML methods which require such input (for instance, one hot encoding and ReceptorCNN method).

  • number_of_splits (int) – number of splits for the cross-validation to be performed for selection the best hyperparameters of the ML model; note that if this is used in combination with nested cross-validation in TrainMLModel instruction, it can result in very few examples in each split depending on the orginal dataset size and the nested cross-validation setup.

  • label (Label) – the label for which the classifier will be created. immuneML also supports multi-label classification, but it is handled outside MLMethod class by creating an MLMethod instance for each label. This means that each MLMethod should handle only one label.

  • cores_for_training (int) – number of processes to be used during the cross-validation for model selection

  • optimization_metric (str) – the name of the optimization metric to be used to select the best model during cross-validation; when used with TrainMLModel instruction which is almost exclusively the case when the immuneML is run from the specification, this maps to the optimization metric in the instruction.

Returns:

it doesn’t return anything, but fits the model parameters instead

get_class_mapping() dict[source]

Returns a dictionary containing the mapping between label values and values internally used in the classifier

get_compatible_encoders()[source]
get_feature_names() list[source]

Returns the list of feature names (a list of strings) if available where the feature names were provided by the encoder in the EncodedData object.

get_label_name()[source]

Returns the name of the label for which the model was fitted.

get_package_info() str[source]

Returns the package and version used for implementing the ML method if an external package was used or immuneML version if it is custom implementation. See py:mod:immuneML.ml_methods.SklearnMethod.SklearnMethod and py:mod:immuneML.ml_methods.ProbabilisticBinaryClassifier.ProbabilisticBinaryClassifier for examples of both.

get_params()[source]

Returns the model parameters in a readable yaml-friendly way (consisting of lists, dictionaries and strings).

load(path)[source]

The load function can load the model given the folder where the same class of the model was previously stored using the store function. It reads in the parameters of the model and sets the values to the object attributes so that the model can be reused. For instance, this is used in MLApplication instruction when the previously trained model is applied on a new dataset.

Parameters:

path (Path) – path to the folder where the model was stored using store() function

Returns:

it does not have a return value, but sets the attribute values of the object instead

predict(encoded_data: EncodedData, label: Label)[source]

The predict function predicts the class for the given label across examples provided in encoded data.

Parameters:
  • encoded_data (EncodedData) – an instance of EncodedData class which includes encoded examples (repertoires, receptors or sequences), their labels, names of the features and other additional information. Most often, only examples and labels will be used. Examples are either a dense numpy matrix or a sparse matrix, where columns correspond to features and rows correspond to examples. There are a few encodings which make multidimensional outputs that do not follow this pattern, but they are tailored to specific ML methods which require such input (for instance, one hot encoding and ReceptorCNN method).

  • label – (Label): the label for which the classifier will be created. immuneML also supports multi-label classification, but it is handled outside MLMethod class by creating an MLMethod instance for each label. This means that each MLMethod should handle only one label.

Returns:

e.g., {label_name: [class1, class2, class2, class1]}

Return type:

a dictionary where the key is the label_name and the value is a list of class predictions (one prediction per example)

predict_proba(encoded_data: EncodedData, label: Label)[source]

The predict_proba function predicts class probabilities for the given label if the model supports probabilistic output. If not, it should raise a warning and return predicted classes without probabilities.

The function will return a nested dictionary. The key(s) of the outer dictionary represent the label name(s), and the keys of the inner dictionary the class names of the respective label. The utility function py:mod:immuneML.ml_methods.util.Util.Util.make_binary_class_mapping may be used to handle mapping of class names to an internal representation for binary classification.

Parameters:
  • encoded_data (EncodedData) – an object of EncodedData class where the examples attribute should be used to make predictions. examples

  • format (attribute includes encoded examples in matrix) –

  • labels (provided here can include) –

  • labels) (been) –

  • function (so the labels attribute of the EncodedData object should NOT be used in this) –

  • set. (even if it is) –

  • label (Label) – the label for which the prediction should be made. It can be used to check if it matches the label that the

  • not (model has been trained for and if) –

  • any (an exception should be thrown. It is often an AssertionError as this can be checked before) –

  • made (prediction is) –

  • cases (but could also be a RuntimeError. It both) –

  • message. (it should include a user-friendly) –

Returns:

a nested dictionary where the outer keys represent label names, inner keys represent class names for the respective label, and innermost values are 1D numpy arrays with class probabilities. For example for instance for label CMV where the class can be either True or False and there are 3 examples to predict the class probabilities for: {CMV: {True: [0.2, 0.55, 0.98], False: [0.8, 0.45, 0.02]}}

set_background_probabilities()[source]
store(path: Path, feature_names=None, details_path: Path = None)[source]

The store function stores the object on which it is called so that it can be imported later using load function. It typically uses pickle, yaml or similar modules to store the information. It can store one or multiple files.

Parameters:
  • path (Path) – path to folder where to store the model

  • feature_names (list) – list of feature names in the encoded data; this can be stored as well to make it easier to map linear models to specific features as provided by the encoded (e.g., in case of logistic regression, this feature list defines what coefficients refer to)

  • details_path (Path) – path to folder where to store the details of the model. The details can be there to better understand the model but are not mandatory and are typically not loaded with the model afterwards. This is user-friendly file that can be examined manually by the user. It does not have to be created or can be created at the same folder as the path parameters points to. In practice, when used with TrainMLModel instruction, this parameter will either be None or have the same value as path parameter.

Returns:

it does not have a return value

immuneML.ml_methods.SVC module

class immuneML.ml_methods.SVC.SVC(parameter_grid: dict = None, parameters: dict = None)[source]

Bases: SklearnMethod

This is a wrapper of scikit-learn’s LinearSVC class. Please see the scikit-learn documentation of SVC for the parameters.

Note: if you are interested in plotting the coefficients of the SVC model, consider running the Coefficients report.

For usage instructions, check SklearnMethod.

YAML specification:

my_svc: # user-defined method name
    SVC: # name of the ML method
        # sklearn parameters (same names as in original sklearn class)
        C: [0.01, 0.1, 1, 10, 100] # find the optimal value for C
        # Additional parameter that determines whether to print convergence warnings
        show_warnings: True
    # if any of the parameters under SVC is a list and model_selection_cv is True,
    # a grid search will be done over the given parameters, using the number of folds specified in model_selection_n_folds,
    # and the optimal model will be selected
    model_selection_cv: True
    model_selection_n_folds: 5
# alternative way to define ML method with default values:
my_default_svc: SVC
can_predict_proba() bool[source]

Returns whether the ML model can be used to predict class probabilities or class assignment only.

static get_documentation()[source]
get_params()[source]

Returns the model parameters in a readable yaml-friendly way (consisting of lists, dictionaries and strings).

immuneML.ml_methods.SVM module

class immuneML.ml_methods.SVM.SVM(parameter_grid: dict = None, parameters: dict = None)[source]

Bases: SklearnMethod

This is a wrapper of scikit-learn’s SVC class. Please see the scikit-learn documentation of SVC for the parameters.

Note: if you are interested in plotting the coefficients of the SVM model, consider running the Coefficients report.

For usage instructions, check SklearnMethod.

YAML specification:

my_svm: # user-defined method name
    SVM: # name of the ML method
        # sklearn parameters (same names as in original sklearn class)
        C: [0.01, 0.1, 1, 10, 100] # find the optimal value for C
        kernel: linear
        # Additional parameter that determines whether to print convergence warnings
        show_warnings: True
    # if any of the parameters under SVM is a list and model_selection_cv is True,
    # a grid search will be done over the given parameters, using the number of folds specified in model_selection_n_folds,
    # and the optimal model will be selected
    model_selection_cv: True
    model_selection_n_folds: 5
# alternative way to define ML method with default values:
my_default_svm: SVM
can_predict_proba() bool[source]

Returns whether the ML model can be used to predict class probabilities or class assignment only.

static get_documentation()[source]
get_params()[source]

Returns the model parameters in a readable yaml-friendly way (consisting of lists, dictionaries and strings).

immuneML.ml_methods.SklearnMethod module

class immuneML.ml_methods.SklearnMethod.SklearnMethod(parameter_grid: dict = None, parameters: dict = None)[source]

Bases: MLMethod

Base class for ML methods imported from scikit-learn. The classes inheriting SklearnMethod acting as wrappers around imported ML methods from scikit-learn have to implement:

  • the __init__() method,

  • get_params(label) and

  • _get_ml_model()

Other methods can also be overwritten if needed. The arguments and specification described bellow applied for all classes inheriting SklearnMethod.

Parameters:
  • parameters – a dictionary of parameters that will be directly passed to scikit-learn’s class upon calling __init__() method; for detailed list see scikit-learn’s documentation of the specific class inheriting SklearnMethod

  • parameter_grid – a dictionary of parameters which all have to be valid arguments for scikit-learn’s corresponding class’ __init__() method (same as parameters), but unlike parameters argument can contain list of values instead of one value; if this is specified and “model_selection_cv” is True (in the specification) or just if fit_by_cross_validation() is called, a grid search will be performed over these parameters and the optimal model will be kept

YAML specification:

ml_methods:
log_reg:
LogisticRegression: # name of the class inheriting SklearnMethod

# sklearn parameters (same names as in original sklearn class) max_iter: 1000 # specific parameter value penalty: l1 # Additional parameter that determines whether to print convergence warnings show_warnings: True

# if any of the parameters under LogisticRegression is a list and model_selection_cv is True, # a grid search will be done over the given parameters, using the number of folds specified in model_selection_n_folds, # and the optimal model will be selected model_selection_cv: True model_selection_n_folds: 5

svm_with_cv:
SVM: # name of another class inheriting SklearnMethod

# sklearn parameters (same names as in original sklearn class) alpha: 10 # Additional parameter that determines whether to print convergence warnings show_warnings: True

# no grid search will be done model_selection_cv: False

FIT = 'fit'
FIT_CV = 'fit_CV'
can_predict_proba() bool[source]

Returns whether the ML model can be used to predict class probabilities or class assignment only.

check_if_exists(path: Path)[source]

The check_if_exists function checks if there is a stored model on the given path. Might be useful in the future for implementing checkpoints. See SklearnMethod for example usage.

Parameters:

path (Path) – path to folder where it should be checked if the model was stored previously

Returns:

whether the model was stored previously on the given path or not

Return type:

True/False

check_is_fitted(label_name: str)[source]
fit(encoded_data: EncodedData, label: Label, cores_for_training: int = 2)[source]

The fit function fits the parameters of the machine learning model.

Parameters:
  • encoded_data (EncodedData) – an instance of EncodedData class which includes encoded examples (repertoires, receptors or sequences), their labels, names of the features and other additional information. Most often, only examples and labels will be used. Examples are either a dense numpy matrix or a sparse matrix, where columns correspond to features and rows correspond to examples. There are a few encodings which make multidimensional outputs that do not follow this pattern, but they are tailored to specific ML methods which require such input (for instance, one hot encoding and ReceptorCNN method).

  • label (Label) – the label for which the classifier will be created. immuneML also supports multi-label classification, but it is handled outside MLMethod class by creating an MLMethod instance for each label. This means that each MLMethod should handle only one label.

  • cores_for_training (int) – if parallelization is available in the MLMethod (and the availability depends on the specific classifier), this is the number of processes that will be creating when fitting the model to speed up the computation.

Returns:

it doesn’t return anything, but fits the model parameters instead

fit_by_cross_validation(encoded_data: EncodedData, number_of_splits: int = 5, label: Label = None, cores_for_training: int = -1, optimization_metric='balanced_accuracy')[source]

The fit_by_cross_validation function should implement finding the best model hyperparameters through cross-validation. In immuneML, preprocessing, encoding and ML hyperparameters can be optimized by using nested cross-validation (see TrainMLModelInstruction for more details). This function is in that setting the third level of nested cross-validation as it can optimize only over the model hyperparameters. It represents an alternative to optimizing the model hyperparameters in the TrainMLModelInstruction. Which one should be used depends on the use-case and specific models: models based on scikit-learn implementations come with this option by default (see SklearnMethod class), while custom classifiers typically do not implement this and just call fit() function and throw a warning instead.

Parameters:
  • encoded_data (EncodedData) – an instance of EncodedData class which includes encoded examples (repertoires, receptors or sequences), their labels, names of the features and other additional information. Most often, only examples and labels will be used. Examples are either a dense numpy matrix or a sparse matrix, where columns correspond to features and rows correspond to examples. There are a few encodings which make multidimensional outputs that do not follow this pattern, but they are tailored to specific ML methods which require such input (for instance, one hot encoding and ReceptorCNN method).

  • number_of_splits (int) – number of splits for the cross-validation to be performed for selection the best hyperparameters of the ML model; note that if this is used in combination with nested cross-validation in TrainMLModel instruction, it can result in very few examples in each split depending on the orginal dataset size and the nested cross-validation setup.

  • label (Label) – the label for which the classifier will be created. immuneML also supports multi-label classification, but it is handled outside MLMethod class by creating an MLMethod instance for each label. This means that each MLMethod should handle only one label.

  • cores_for_training (int) – number of processes to be used during the cross-validation for model selection

  • optimization_metric (str) – the name of the optimization metric to be used to select the best model during cross-validation; when used with TrainMLModel instruction which is almost exclusively the case when the immuneML is run from the specification, this maps to the optimization metric in the instruction.

Returns:

it doesn’t return anything, but fits the model parameters instead

get_class_mapping() dict[source]

Returns a dictionary containing the mapping between label values and values internally used in the classifier

get_compatible_encoders()[source]
get_feature_names() list[source]

Returns the list of feature names (a list of strings) if available where the feature names were provided by the encoder in the EncodedData object.

get_label_name()[source]

Returns the name of the label for which the model was fitted.

get_package_info() str[source]

Returns the package and version used for implementing the ML method if an external package was used or immuneML version if it is custom implementation. See py:mod:immuneML.ml_methods.SklearnMethod.SklearnMethod and py:mod:immuneML.ml_methods.ProbabilisticBinaryClassifier.ProbabilisticBinaryClassifier for examples of both.

abstract get_params()[source]

Returns the model parameters in a readable yaml-friendly way (consisting of lists, dictionaries and strings).

static get_usage_documentation(model_name)[source]
load(path: Path, details_path: Path = None)[source]

The load function can load the model given the folder where the same class of the model was previously stored using the store function. It reads in the parameters of the model and sets the values to the object attributes so that the model can be reused. For instance, this is used in MLApplication instruction when the previously trained model is applied on a new dataset.

Parameters:

path (Path) – path to the folder where the model was stored using store() function

Returns:

it does not have a return value, but sets the attribute values of the object instead

predict(encoded_data: EncodedData, label: Label)[source]

The predict function predicts the class for the given label across examples provided in encoded data.

Parameters:
  • encoded_data (EncodedData) – an instance of EncodedData class which includes encoded examples (repertoires, receptors or sequences), their labels, names of the features and other additional information. Most often, only examples and labels will be used. Examples are either a dense numpy matrix or a sparse matrix, where columns correspond to features and rows correspond to examples. There are a few encodings which make multidimensional outputs that do not follow this pattern, but they are tailored to specific ML methods which require such input (for instance, one hot encoding and ReceptorCNN method).

  • label – (Label): the label for which the classifier will be created. immuneML also supports multi-label classification, but it is handled outside MLMethod class by creating an MLMethod instance for each label. This means that each MLMethod should handle only one label.

Returns:

e.g., {label_name: [class1, class2, class2, class1]}

Return type:

a dictionary where the key is the label_name and the value is a list of class predictions (one prediction per example)

predict_proba(encoded_data: EncodedData, label: Label)[source]

The predict_proba function predicts class probabilities for the given label if the model supports probabilistic output. If not, it should raise a warning and return predicted classes without probabilities.

The function will return a nested dictionary. The key(s) of the outer dictionary represent the label name(s), and the keys of the inner dictionary the class names of the respective label. The utility function py:mod:immuneML.ml_methods.util.Util.Util.make_binary_class_mapping may be used to handle mapping of class names to an internal representation for binary classification.

Parameters:
  • encoded_data (EncodedData) – an object of EncodedData class where the examples attribute should be used to make predictions. examples

  • format (attribute includes encoded examples in matrix) –

  • labels (provided here can include) –

  • labels) (been) –

  • function (so the labels attribute of the EncodedData object should NOT be used in this) –

  • set. (even if it is) –

  • label (Label) – the label for which the prediction should be made. It can be used to check if it matches the label that the

  • not (model has been trained for and if) –

  • any (an exception should be thrown. It is often an AssertionError as this can be checked before) –

  • made (prediction is) –

  • cases (but could also be a RuntimeError. It both) –

  • message. (it should include a user-friendly) –

Returns:

a nested dictionary where the outer keys represent label names, inner keys represent class names for the respective label, and innermost values are 1D numpy arrays with class probabilities. For example for instance for label CMV where the class can be either True or False and there are 3 examples to predict the class probabilities for: {CMV: {True: [0.2, 0.55, 0.98], False: [0.8, 0.45, 0.02]}}

store(path: Path, feature_names=None, details_path: Path = None)[source]

The store function stores the object on which it is called so that it can be imported later using load function. It typically uses pickle, yaml or similar modules to store the information. It can store one or multiple files.

Parameters:
  • path (Path) – path to folder where to store the model

  • feature_names (list) – list of feature names in the encoded data; this can be stored as well to make it easier to map linear models to specific features as provided by the encoded (e.g., in case of logistic regression, this feature list defines what coefficients refer to)

  • details_path (Path) – path to folder where to store the details of the model. The details can be there to better understand the model but are not mandatory and are typically not loaded with the model afterwards. This is user-friendly file that can be examined manually by the user. It does not have to be created or can be created at the same folder as the path parameters points to. In practice, when used with TrainMLModel instruction, this parameter will either be None or have the same value as path parameter.

Returns:

it does not have a return value

immuneML.ml_methods.TCRdistClassifier module

class immuneML.ml_methods.TCRdistClassifier.TCRdistClassifier(percentage: float, show_warnings: bool = True)[source]

Bases: SklearnMethod

Implementation of a nearest neighbors classifier based on TCR distances as presented in Dash P, Fiore-Gartland AJ, Hertz T, et al. Quantifiable predictive features define epitope-specific T cell receptor repertoires. Nature. 2017; 547(7661):89-93. doi:10.1038/nature22383.

This method is implemented using scikit-learn’s KNeighborsClassifier with k determined at runtime from the training dataset size and weights linearly scaled to decrease with the distance of examples.

Parameters:
  • percentage (float) – percentage of nearest neighbors to consider when determining receptor specificity based on known receptors (between 0 and 1)

  • show_warnings (bool) – whether to show warnings generated by scikit-learn, by default this is True.

YAML specification:

my_tcr_method:
    TCRdistClassifier:
        percentage: 0.1
        show_warnings: True
can_predict_proba() bool[source]

Returns whether the ML model can be used to predict class probabilities or class assignment only.

get_compatible_encoders()[source]
get_params()[source]

Returns the model parameters in a readable yaml-friendly way (consisting of lists, dictionaries and strings).

Module contents