Python for AI — NumPy आणि Pandas

pip install numpy pandas matplotlib scikit-learn tensorflow
# किंवा Anaconda distribution — सगळं built-in असतं
# Google Colab — free GPU, no setup needed!

import numpy as np

# Array बनवा
arr = np.array([1, 2, 3, 4, 5])
matrix = np.array([[1, 2, 3], [4, 5, 6], [7, 8, 9]])

# Shape बघा
print(arr.shape)     # (5,)
print(matrix.shape)  # (3, 3)

# Operations — element-wise (Python list पेक्षा 100x fast)
print(arr * 2)       # [2, 4, 6, 8, 10]
print(arr + arr)     # [2, 4, 6, 8, 10]
print(np.sum(arr))   # 15
print(np.mean(arr))  # 3.0
print(np.std(arr))   # standard deviation

# Random data (ML साठी)
X = np.random.randn(100, 5)  # 100 samples, 5 features
y = np.random.randint(0, 2, 100)  # 100 binary labels

import pandas as pd

# CSV load करा
df = pd.read_csv('students.csv')

# Data बघा
print(df.head())        # पहिले 5 rows
print(df.shape)         # (rows, columns)
print(df.info())        # dtypes, null counts
print(df.describe())    # statistics

# Column select
ages = df['age']
subset = df[['name', 'grade']]

# Filter
high_scorers = df[df['score'] > 80]
mumbai = df[df['city'] == 'Mumbai']

# Missing values handle करा
df['age'].fillna(df['age'].mean(), inplace=True)  # mean ने fill
df.dropna(inplace=True)  # rows with NaN drop

# New column
df['score_category'] = df['score'].apply(lambda x: 'High' if x > 80 else 'Low')

# Save
df.to_csv('cleaned_data.csv', index=False)