All notable changes to this project will be documented in this file.
The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.
-
function
transform_data_to_vectors
- the logic were change from os.replace(temp_file.name, model_file) to os.rename(temp_file_path, model_file) => speed
-
In features is better to have [0 or 1 or 2 or 3] compared to one hot encoded [0, 0, 0]. Pros: The file is more readable. The features are in Model embedded.
-
In case for
gene
prediction there cannot be all features from gff3 features such asmRNA
does not make sense cuz wheregene
there ismRNA
also featureexon
=> The user specifies inconfig.py
which features want from.gff3
file. e.gGFF3_FEATURES = ["miRNA", "rRNA", "gene"]