Inside Uncodemy's Data Science Course: What You Learn

import pandas as pd

df = pd.DataFrame({
    "customer": ["A", "B", "B", "C"],
    "city": [" delhi", "Noida ", "Noida ", None],
    "spend": ["1200", "850", "850", "n/a"],
})
df = df.drop_duplicates()
df["city"] = df["city"].str.strip().str.title().fillna("Unknown")
df["spend"] = pd.to_numeric(df["spend"], errors="coerce")
print(df)
print(df.groupby("city")["spend"].agg(["count", "mean"]))

-- Monthly spend per customer, with a running total
SELECT
  customer_id,
  DATE_TRUNC('month', order_date) AS month,
  SUM(amount) AS monthly_spend,
  SUM(SUM(amount)) OVER (
    PARTITION BY customer_id
    ORDER BY DATE_TRUNC('month', order_date)
  ) AS running_total
FROM orders
GROUP BY customer_id, DATE_TRUNC('month', order_date)
ORDER BY customer_id, month;

from sklearn.datasets import load_breast_cancer
from sklearn.ensemble import RandomForestClassifier
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import classification_report
from sklearn.model_selection import train_test_split
from sklearn.pipeline import make_pipeline
from sklearn.preprocessing import StandardScaler

X, y = load_breast_cancer(return_X_y=True)
X_tr, X_te, y_tr, y_te = train_test_split(X, y, test_size=0.25, stratify=y, random_state=42)

models = {
    "logistic": make_pipeline(StandardScaler(), LogisticRegression(max_iter=1000)),
    "forest": RandomForestClassifier(random_state=42),
}
for name, m in models.items():
    m.fit(X_tr, y_tr)
    print(name)
    print(classification_report(y_te, m.predict(X_te), digits=3))

import numpy as np
from scipy import stats

rng = np.random.default_rng(1)
version_a = rng.normal(5.0, 1.0, 200)   # simulated minutes on page
version_b = rng.normal(5.3, 1.0, 200)

t, p = stats.ttest_ind(version_a, version_b, equal_var=False)
print(f"mean A={version_a.mean():.2f}, mean B={version_b.mean():.2f}")
print(f"t={t:.2f}, p={p:.4f}")
print("difference unlikely to be chance" if p < 0.05 else "no clear difference")

# pip install fastapi uvicorn scikit-learn joblib
# Run: uvicorn api:app --port 8000
import joblib
from fastapi import FastAPI
from pydantic import BaseModel

app = FastAPI()
model = joblib.load("model.joblib")   # saved after training

class Features(BaseModel):
    values: list[float]

@app.post("/predict")
def predict(f: Features):
    proba = float(model.predict_proba([f.values])[0][1])
    return {"probability": round(proba, 3)}

Capstone project outline (example):

1. Problem: predict which customers may cancel a subscription
2. Data: load, clean and document quality issues
3. EDA: charts and key patterns
4. Baseline model and metric choice explained
5. Improved model with cross-validation
6. Error analysis: where the model fails
7. Simple API or app for predictions
8. README: problem, approach, results, limitations

1. Can I see the current official data science syllabus?
2. How long is the course, and what are the batch timings?
3. How many hours are hands-on practice?
4. Which projects are included, and are they reviewed?
5. Does the course cover SQL, statistics and deployment basics?
6. What career preparation is included?
7. What are the current fees and payment terms?

Module               Started   Practised   Project done
Python basics        [ ]       [ ]         [ ]
NumPy and pandas     [ ]       [ ]         [ ]
SQL                  [ ]       [ ]         [ ]
Statistics           [ ]       [ ]         [ ]
Visualisation        [ ]       [ ]         [ ]
Machine learning     [ ]       [ ]         [ ]
Deep learning intro  [ ]       [ ]         [ ]
Deployment basics    [ ]       [ ]         [ ]
Capstone             [ ]       [ ]         [ ]

Live classes:        ____ hours
Self-practice:       ____ hours (aim for at least equal to class time)
Project work:        ____ hours
Revision of notes:   ____ hours

This week's practice goal: ______________________
Question to ask the trainer: ______________________