bactoscoop.curation¶
Classes and public operations¶
Curation¶
- class bactoscoop.curation.Curation(svm_df)¶
Prepare phase/morphology features and predict retained/rejected cells using a named SVM.
Supply the svm feature DataFrame produced by the collection workflow; compiled_curation loads a model, prepares features, and returns labels.
- Parameters:
svm_df – Input
svm_dfconsumed by this implementation; its exact use is visible in the linked source below. No maintained parameter description is present in the original docstring.- Returns:
None (no explicit return statement).
Assigned state:
self.svm_df,self.features,self.curated_df,self.svm_model.
Curation.compiled_curation¶
- Curation.compiled_curation(path_to_model, cols)¶
Perform compiled cell curation using a trained SVM model.
- Parameters:
path_to_model – Path to a serialized SVM carrying the
feature_names_in_training-column schema.cols – Legacy curation argument retained for compatibility; unused by the current curation preparation.
Return expressions in source:
self.curated_df
Assigned state:
self.curated_df.Original docstring (legacy wording may differ from the current signature)
Perform compiled cell curation using a trained SVM model. This method curates the dataset by making predictions using a trained Support Vector Machine (SVM) model. Cells that do not meet the criteria specified by the model are flagged for potential removal. Parameters ---------- path_to_model : str The path to the trained SVM model file. cols : int The number of columns to exclude in the dataset during curation. Returns ------- pd.DataFrame A DataFrame containing the curated dataset.
Curation.load_model¶
- Curation.load_model(path_to_model)¶
Load a trained SVM model from a file.
- Parameters:
path_to_model – Path to a serialized SVM carrying the
feature_names_in_training-column schema.- Returns:
None (no explicit return statement).
Assigned state:
self.svm_model.Original docstring (legacy wording may differ from the current signature)
Load a trained SVM model from a file. This method loads a trained Support Vector Machine (SVM) model from a file located at the given path. Parameters ---------- path_to_model : str The path to the trained SVM model file.
Curation.prepare_dataframe¶
- Curation.prepare_dataframe(cols)¶
Prepare the feature dataset for SVM curation.
- Parameters:
cols – Legacy curation argument retained for compatibility; unused by the current curation preparation.
- Returns:
None (no explicit return statement).
Assigned state:
self.cell_id_frame,self.features,self.features.select_dtypes(include=[np.number]).columns,self.features.select_dtypes,self.svm_df_nan,self.svm_df.Original docstring (legacy wording may differ from the current signature)
Prepare the feature dataset for SVM curation. This method prepares the feature dataset for cell curation using a Support Vector Machine (SVM) model. It extracts the relevant feature columns while ensuring that `cell_id` and `frame` are preserved. Parameters ---------- cols : int Retained for backward compatibility; feature columns are selected by name and this argument is currently unused.
Curation.get_label_proportions¶
- Curation.get_label_proportions()¶
Calculate the proportions of labels in the curated dataset.
Return expressions in source:
(proportion_label_1, proportion_label_0)
View current source, line 110.
Original docstring (legacy wording may differ from the current signature)
Calculate the proportions of labels in the curated dataset. This method calculates and returns the proportions of label 1 and label 0 in the curated dataset. It counts the occurrences of each label and divides them by the total number of samples in the dataset to compute the proportions. Returns ------- float The proportion of label 1 in the curated dataset. float The proportion of label 0 in the curated dataset.
Curation.make_predictions¶
- Curation.make_predictions()¶
Make predictions using the SVM model and reattach cell_id and frame.
Return expressions in source:
self.curated_df
Assigned state:
self.curated_df.Explicitly raised exceptions:
ValueError. Errors from called functions can also propagate or be recorded.View current source, line 133.
Original docstring (legacy wording may differ from the current signature)
Make predictions using the SVM model and reattach `cell_id` and `frame`. Returns ------- pd.DataFrame A DataFrame containing the predicted labels with corresponding `cell_id` and `frame`.
Curation.get_control¶
- Curation.get_control(num_pos=5, num_neg=5)¶
Randomly select retained and rejected curation examples.
Returns two lists of (cell_id, frame) pairs, or (None, None) when either requested class is too small.
- Parameters:
num_pos – The number of positive class cell IDs to retrieve. The default is 5.
num_neg – The number of negative class cell IDs to retrieve. The default is 5.
Return expressions in source:
(positive_cell_frames, negative_cell_frames) (None, None)
View current source, line 174.
Original docstring (legacy wording may differ from the current signature)
Get control cell IDs for positive and negative classes. This method retrieves cell IDs for positive and negative classes based on the curated dataset. It randomly selects a specified number of positive and negative cell IDs and returns them as lists, along with the corresponding 'frame' values. Parameters ---------- num_pos : int, optional The number of positive class cell IDs to retrieve. The default is 5. num_neg : int, optional The number of negative class cell IDs to retrieve. The default is 5. Returns ------- Tuple Two lists containing tuples of ('frame', 'cell_id') for positive and negative class control cell IDs. If there are not enough samples for one class, the corresponding list will be empty.