| input_dataframe |
Path to the input DataFrame. |
*mandatory* |
| output_directory |
Path to the folder where the results will be collected. |
results/ |
| preprocessed_filename |
Filename of the preprocessed DataFrame. |
None |
| graph_filename |
Filename of the graph file. |
None |
| numeric_columns |
List of columns to be treated as numeric. |
None |
| categorical_columns |
List of columns to be treated as categorical. |
[ ] |
| target_columns |
Column to be used as the target variable for coloring the graph and neighborhood statistics. |
[ ] |
| ignore_columns |
List of columns to ignore during preprocessing. |
[ ] |
| unknown_column_action |
Action to take on columns not specified as numeric, categorical, or ignored. |
“infer” |
| numeric_threshold |
Threshold for determining if a column is numeric. |
0.05 |
| numeric_scaling |
Method for scaling numeric columns. |
“standard” |
| categorical_encoding |
Method for encoding categorical columns. |
“one-hot” |
| nan_action |
Action to take on NaN values. |
“infer” |
| nan_threshold |
Threshold for dropping columns based on NaN ratio. |
0.5 |
| verbose |
Flag for detailed output. |
True |
| overwrite |
Overwrite or not the previous output. |
False |
| manifold_method |
Method for manifold learning on numeric columns. |
None |
| manifold_dimension |
Number of dimensions for manifold learning output. |
None |
| method |
Method for creating the graph. |
“knn” |
| k |
Number of neighbors for KNN graph creation. |
5 |
| distance_threshold |
Threshold for distance-based graph creation. |
None |
| similarity_threshold |
Threshold for similarity-based graph creation. |
None |
| neigh_prob_path |
Filename for neighborhood probability statistics. |
“neigh_prob.txt” |
| degree_distribution_filename |
Filename for the degree distribution plot. |
“degree.png” |
| community_filename |
Filename for the community composition histogram. |
“communities.png” |
| graph_visualization_filename |
Filename for the graph visualization. |
“graph.png” |
| prob_heatmap_filename |
Filename for the neighborhood probability heatmap. |
“neigh_prob_heatmap.png” |
| network_metrics_filename |
Filename for the other network metrics. |
None (will be displayed in terminal) |
