Sagar8229/data-preprocessing-ml

1import numpy as np
2import pandas as pd
3import gradio as gr
4
5from sklearn.impute import SimpleImputer
6from sklearn.compose import ColumnTransformer
7from sklearn.preprocessing import OneHotEncoder, StandardScaler, LabelEncoder
8
9def preprocess():
10
11    # Load dataset
12    dataset = pd.read_csv("dataset.csv")
13
14    output = "Original Dataset:\n"
15    output += dataset.to_string()
16    output += "\n\n"
17
18    # Features and Target
19    X = dataset.iloc[:,1:-1].values
20    y = dataset.iloc[:,-1].values
21
22    # Handle missing data
23    imputer = SimpleImputer(missing_values=np.nan, strategy="mean")
24    X[:,0:2] = imputer.fit_transform(X[:,0:2])
25
26    # Encode categorical data
27    ct = ColumnTransformer(
28        transformers=[("encoder", OneHotEncoder(), [2])],
29        remainder="passthrough"
30    )
31
32    X = ct.fit_transform(X)
33
34    # Encode target variable
35    le = LabelEncoder()
36    y = le.fit_transform(y)
37
38    # Feature scaling
39    scaler = StandardScaler()
40    X = scaler.fit_transform(X)
41
42    output += "Features after scaling:\n"
43    output += str(X)
44    output += "\n\nTarget:\n"
45    output += str(y)
46    output += "\n\nPreprocessing Complete!"
47
48    return output
49
50
51demo = gr.Interface(
52    fn=preprocess,
53    inputs=None,
54    outputs="text",
55    title="Data Preprocessing ML App",
56    description="Basic Machine Learning Data Preprocessing (Missing Values, Encoding, Scaling)"
57)
58
59demo.launch()