a python package to benchmarks algorithms against various datasets
Project description
baselines
a python package used for benchmarking algorithms over various datasets
Installation
pip install base2lines
Introduction
Data class :
show_datasets:
show available datasets
load_dataset :
load specific dataset (pass name as argument)
Classification class :
classification method:
parameters:
dataset in csv format or pandas dataframe with classification column at last
result:
####### returns benchmark for dataset with various algorithms and its f1_score
Regression Class:
regressor method:
parameters:
dataset (pandas dataframe)
resultantColumn (name of column which has the value to be predicted)
Pipeline Class:
init method:
parameters:
dataset (pandas dataframe)
show_graph method:
parameters:
pipeline_show (sklearn Pipeline object which will be inturn turned to graph)
How to use
Data Class
from baseline.Data import Data
data = Data()
dataset_names = data.show_dataset()
#name should be only from dataset_names
dataset = data.load_dataset('name')
Classification Class
from baseline.classification import Classification
classifier = Classification()
# dataset's last column should be the column to be classified
results = classifier.classification(dataset)
print(results)
Regression Class
from baseline.regression import Regression
regressors = Regression()
# also pass the resultant column
results = regressors.regression(dataset , resultantColumn)
print(results)
Pipeline Class
from baseline.pipeline import Pipeline
graph = Pipeline(dataset)
#pipeline should only have sklearn components
graph.show_graph(pipelineToBeShown)
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
base2lines-0.0.14.tar.gz
(18.3 kB
view hashes)
Built Distribution
Close
Hashes for base2lines-0.0.14-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 15ae8a05cf7faf08acd4ab152c90f1f9df6da5db7a9eb0170658f59aeafb6ab4 |
|
MD5 | a56f55b2f706a20d1dd80422aad22e37 |
|
BLAKE2b-256 | 4f5c4a444750df2d3ea74d5c53e765d1c2b04700c2b58d1a89e4b98e91b0b61d |