development

immuneML/IO/dataset_export/AIRRExporter.py

@@ -59,7 +59,7 @@ def export_repertoire_dataset(dataset: RepertoireDataset, path: Path, omit_colum
         else:
 
             new_metadata_df = pd.read_csv(path / f"{dataset.name}.csv")
-            labels = {label: list(new_metadata_df[label].unique())
+            labels = {label: new_metadata_df[label].unique().tolist()
                       for label in dataset.get_label_names(refresh=True)}
 
             dataset_yaml = RepertoireDataset.create_metadata_dict(metadata_file=path / f"{dataset.name}.csv",

immuneML/config/default_params/ml_methods/progen_params.yaml

@@ -0,0 +1,14 @@
+locus: beta
+device: cpu
+num_frozen_layers: 22
+num_epochs: 2
+learning_rate: 0.00004
+fp16: False
+prefix_text: '<|bos|>1'
+suffix_text: '2<|eos|>'
+max_new_tokens: 1024
+temperature: 1
+top_p: 0.9
+prompt: '1'
+num_gen_batches: 1
+per_device_train_batch_size: 2

immuneML/config/default_params/reports/vj_gene_distribution_params.yaml

@@ -1,3 +1,4 @@
-split_by_label: False
+split_by_label: false
 label: null
-is_sequence_label: False
+is_sequence_label: false
+show_joint_dist: true

immuneML/data_model/EncodedData.py

@@ -27,7 +27,7 @@ class EncodedData:
 
     def __init__(self, examples, labels: dict = None, example_ids: list = None, feature_names: list = None,
                  feature_annotations: pd.DataFrame = None, encoding: str = None, example_weights: list = None, info: dict = None,
-                 dimensionality_reduced_data: np.ndarray = None):
+                 dimensionality_reduced_data: np.ndarray = None, dim_names: list = None):
 
         assert feature_names is None or examples.shape[1] == len(feature_names), \
             (f"EncodedData: the length of feature_names ({len(feature_names)}) must match the feature dimension of the "
@@ -59,6 +59,7 @@ def __init__(self, examples, labels: dict = None, example_ids: list = None, feat
         self.encoding = encoding
         self.example_weights = example_weights
         self.info = info
+        self.dim_names = dim_names
         self.dimensionality_reduced_data = dimensionality_reduced_data
 
     def __getstate__(self):

immuneML/dsl/instruction_parsers/ClusteringParser.py

@@ -1,5 +1,6 @@
 import copy
 import inspect
+import logging
 from pathlib import Path
 from typing import List
 
@@ -152,7 +153,13 @@ def parse_clustering_settings(key: str, instruction: dict, symbol_table: SymbolT
                                        instruction)
         settings_objs.append(setting_obj)
 
-    return settings_objs
+    unique_objs = list(set(settings_objs))
+
+    if len(unique_objs) < len(settings_objs):
+        logging.warning(f"Clustering parser: clustering_settings contains {len(settings_objs) - len(unique_objs)} "
+                        f"duplicate settings, keep the following: {[obj.get_key() for obj in unique_objs]}")
+
+    return unique_objs
 
 
 def make_setting_obj(setting, valid_encodings, valid_clusterings, valid_dim_red, symbol_table, instruction):

immuneML/environment/Constants.py

@@ -1,6 +1,6 @@
 class Constants:
 
-    VERSION = "3.0.16"
+    VERSION = "3.0.17"
 
     # encoding constants
     FEATURE_DELIMITER = "_"

immuneML/ml_methods/dim_reduction/DimRedMethod.py

@@ -1,13 +1,14 @@
 import abc
 import logging
+from abc import ABC
 from typing import List
 
 import numpy as np
 
 from immuneML.data_model.datasets.Dataset import Dataset
 
 
-class DimRedMethod:
+class DimRedMethod(ABC):
     """
     Dimensionality reduction methods are algorithms which can be used to reduce the dimensionality
     of encoded datasets, in order to uncover and analyze patterns present in the data.
@@ -21,14 +22,12 @@ def __init__(self, name: str = None):
         self.method = None
         self.name = name
 
-    @abc.abstractmethod
     def fit(self, dataset: Dataset = None, design_matrix: np.ndarray = None):
         if dataset is None:
             self.method.fit(design_matrix)
         else:
             self.method.fit(dataset.encoded_data.get_examples_as_np_matrix())
 
-    @abc.abstractmethod
     def transform(self, dataset: Dataset = None, design_matrix: np.ndarray = None):
         if dataset is None:
             return self.method.transform(design_matrix)