import pandas as pd
import matplotlib.pyplot as plt

df = pd.read_csv("students.csv")      # columns: hours, attendance, score
print(df.describe().round(1))
print(df.isna().sum())
print(df[["hours", "attendance", "score"]].corr().round(2))

df.plot.scatter(x="hours", y="score", title="Study hours vs score")
plt.tight_layout()
plt.savefig("hours_vs_score.png")

from scipy import stats

# Did students in the weekend batch score differently from the weekday batch?
weekday = [72, 68, 75, 80, 66, 71, 78]
weekend = [70, 74, 79, 83, 77, 72, 81]
t, p = stats.ttest_ind(weekday, weekend, equal_var=False)
print(f"t = {t:.2f}, p = {p:.3f}")
# p < 0.05 would suggest a real difference; small samples need caution

from sklearn.datasets import load_wine
from sklearn.model_selection import cross_val_score
from sklearn.linear_model import LogisticRegression
from sklearn.ensemble import RandomForestClassifier
from sklearn.pipeline import make_pipeline
from sklearn.preprocessing import StandardScaler

X, y = load_wine(return_X_y=True)
models = {
    "logistic": make_pipeline(StandardScaler(), LogisticRegression(max_iter=1000)),
    "forest": RandomForestClassifier(n_estimators=200, random_state=42),
}
for name, m in models.items():
    s = cross_val_score(m, X, y, cv=5)
    print(f"{name:8} mean accuracy {s.mean():.3f} (± {s.std():.3f})")

from sklearn.cluster import KMeans
from sklearn.preprocessing import StandardScaler
import pandas as pd

customers = pd.read_csv("customers.csv")   # annual_spend, visits
X = StandardScaler().fit_transform(customers[["annual_spend", "visits"]])
customers["segment"] = KMeans(n_clusters=3, n_init=10, random_state=42).fit_predict(X)
print(customers.groupby("segment")[["annual_spend", "visits"]].mean().round(0))

Capstone structure
1. Business question and success metric
2. Data sourcing and cleaning log
3. Exploratory analysis with key charts
4. Baseline model
5. Improved models with cross-validation
6. Error analysis and limitations
7. Recommendation for the business
8. 5-7 minute presentation

Topics practised in mock interviews
☐ Python and pandas tasks
☐ SQL joins and window functions
☐ Statistics: p-values, confidence intervals, bias
☐ Bias-variance trade-off and overfitting
☐ Choosing evaluation metrics
☐ Explaining your capstone
☐ A business case question

Module Tracker
Module                     Classes   Exercises   Project   Confidence (1-5)
Python for data            __/__     ☐           ☐         ___
pandas + visualisation     __/__     ☐           ☐         ___
SQL                        __/__     ☐           ☐         ___
Statistics                 __/__     ☐           ☐         ___
Supervised learning        __/__     ☐           ☐         ___
Evaluation                 __/__     ☐           ☐         ___
Unsupervised learning      __/__     ☐           ☐         ___
Deep learning intro        __/__     ☐           ☐         ___

Weekly routine during the course
Class days: attend + note questions
Next day: redo the class notebook without looking
3 times a week: 30 min SQL or pandas practice
Weekend: project work, 3-4 hours
Sunday: push to GitHub, update module tracker

Questions for the counsellor
1. Can I see the current syllabus with module hours?
2. Which batches suit my schedule (weekday/weekend/online)?
3. How many projects and how are they reviewed?
4. What does placement support include?
5. What is the current fee and instalment option?