Skip to content

Latest commit

 

History

History
128 lines (80 loc) · 3.37 KB

README.md

File metadata and controls

128 lines (80 loc) · 3.37 KB

No-Code Machine Learning Model Building Platform

Welcome to the No-Code ML Model Building Platform! This platform enables users to perform data cleaning, transformation, visualization, and machine learning model building without writing any code. Link - NoCodeML

Features

1. Data Upload and Display

  • Upload your dataset in CSV format.
  • View and explore the data directly in the app.

"data_upload"

2. Data Cleaning

  • Handle missing values:

    • Drop rows with missing values.

      "drop_na"

    • Impute missing values using mean, median, or mode.

      "impute"

  • Replace values in columns.

    "replace"

  • Change column data types.

    "change_dtype"

  • Detect and remove outliers using:

    • Interquartile Range (IQR)
    • Z-Score

    "outliers"

  • Download Cleaned Data as CSV:

    "download_cleaned_data"

3. Data Analysis and Visualization

"EDA"

  • Unique Values: Explore unique values for selected columns.

    "unique_vals"

  • EDA (Exploratory Data Analysis):

    • Univariate Analysis: Visualize single-variable distributions.

      "univariate"

    • Bivariate Analysis: Analyze relationships between two variables.

      "bivariate"

    • Multivariate Analysis: Correlation matrix and heatmaps.

      "multivariate"

  • Feature Importance:

    • Calculate feature importance using methods like Random Forest, Chi-Square, ANOVA F-value, Point Biserial Correlation, and Mutual Information.

      "feat_impt_1"

      "feat_impt_2"

4. Data Preparation for Modeling

  • Drop irrelevant columns.

    "drop_feat"

  • Split data into train and test sets.

    "train_test_split"

  • Normalize numerical data and encode categorical data using:

    • Label Encoding

    • One-Hot Encoding

    • Ordinal Encoding

    • Standard Scaling

    • Min-Max Scaling

      "feat_norm_encode"

  • Handle class imbalance (coming soon).

  • Perform dimensionality reduction (coming soon).

    "class_imb_pca"

5. Machine Learning Model Building

  • Model Training:

    • Choose between classification or regression tasks.

    • Supported algorithms include Logistic Regression, Random Forest, Gradient Boosting, Support Vector Machines, and more.

      "model_train"

  • Model Testing:

    • Evaluate trained models on test data.

    • Download trained models as pickle files.

      "model_test"

6. Notes and Downloads

  • Take notes directly in the app.
  • Download cleaned data, transformed data, models, and notes.

Getting Started

  1. Clone the repository:

    git clone https://github.com/Dhruv-Limbani/NoCodeML.git

  2. Install the required Python dependencies:

    pip install -r requirements.txt

  3. Run the app:

    streamlit run app.py

  4. Open the app in your browser and upload your data to get started.