smartsolve

a mini-machine learning python package

smartsolve is a versatile machine learning package that empowers you to excel in data analysis and predictive modeling. With a rich array of machine learning models, it equips you to analyze diverse datasets, including text-based, numerical, and categorical data, and make precise predictions. The package is thoughtfully organized into three essential modules: preprocessing, models, and evaluation, each of which hosts a collection of classes with specialized functionalities.

Table of Contents

• Installation
• Usage
• License

Installation

Before installing and using smartsolve, please ensure that the following dependencies are installed on your system:

• numpy
• pandas
• matplotlib
• seaborn
• scikit-learn
• scipy

You can install these dependencies using pip:

pip install numpy pandas matplotlib seaborn scikit-learn scipy

Once the requirements are successfully installed, you can proceed to install smartsolve:

pip install smartsolve

pypi-page of smartsolve: pypi.org

Usage

smartsolve offers a comprehensive workflow for machine learning tasks. Here's how you can use it effectively:

1. Data Preparation: Utilize the preprocessing module to analyze and prepare your raw data. You can split your data into training, validation, and test sets, which can be used for various machine learning models available in the models module.

2. Model Training: Choose a machine learning model from the models module that suits your task. Train your model using the prepared data.

3. Model Evaluation: Once your model is trained, employ the evaluation module to evaluate its performance. You can assess various metrics like mean squared error (MSE), accuracy, precision, recall, F-score, and more.

Here's an example of how to use smartsolve to prepare data, train a model, and evaluate its performance:

from smartsolve.preprocessing import AnalyseData, SplitData
from smartsolve.models import LinearRegression
from smartsolve.evaluation import Validation
import numpy as np
import pandas as pd
import seaborn as sns

# Load and analyze data from a text-based file

path = "example_data.csv"
example = AnalyseData(path)
data = example.load_data()

info = example.infos()
example.stats()
heatmap = example.heat_map(columns='data columns')


# A linear regression model as an example

# Prepare data for splitting
labels = list(data['labels_column'])
feats = ['feature_columns']
features = [[feature[i] for feature in feats] for i in range(len(labels))]

# Combine labels and features
input_data = [(label, feature) for label, feature in zip(labels, features)]

# Split the data into training and testing sets
split = SplitData(data=input_data, method='Random', train=0.8, test=0.2)
train_data, _, test_data = split.random()

# Train the linear regression model
model = LinearRegression(train_data=train_data, max_iter=400, threshold=1e-6)
model.train()
weights = model.coefficients
bias = model.bias

# Prepare test data
test_set = 'test_sample'

# Predict using the trained model
predicted = model.predict(data=test_set)

# Evaluate the model
obj = Validation()
mse = obj.mean_squared_error(actual='actual labels', predicted=predicted)

For more examples, see this page :examples

For more explanation, see the Medium article: Introducing smartsolve: a mini machine learning package

License

This project is licensed under the MIT LICENSE.