Homework 1: Introduction to Machine Learning
- Submissions
- 1625
- Completion
- 68.9%
- Median score
- 7
- Lecture time
- 3.5h
- Homework time
- 1h
- Total time
- 5h
Machine Learning Zoomcamp 2025
Course-level participation, homework, and project statistics.
Median time and score by homework. Hover values on desktop for interquartile ranges.
| Homework | Submissions | Completion | Lecture time | Homework time | Total time | Median score |
|---|---|---|---|---|---|---|
| Homework 1: Introduction to Machine Learning | 1625 | 68.9% | 3.5h | 1h | 5h | 7 |
| Homework 2: Machine Learning for Regression | 1025 | 43.5% | 5h | 3h | 8h | 5 |
| Homework 3: Machine Learning for Classification | 856 | 36.3% | 4h | 3h | 7.2h | 5 |
| Homework 4: Evaluation Metrics for Classification | 680 | 28.8% | 4h | 3h | 8h | 5 |
| Homework 5: Deploying Machine Learning Models | 537 | 22.8% | 5h | 3h | 8h | 5 |
| Homework 6: Decision Trees and Ensemble Learning | 393 | 16.7% | 5h | 3h | 8h | 6 |
| Homework 8: Neural Networks and Deep Learning | 336 | 14.2% | 6h | 3h | 10h | 6 |
| Homework 9: Serverless Deep Learning | 249 | 10.6% | 4h | 3h | 7.8h | 5 |
| Homework 10: Kubernetes | 187 | 7.9% | 4h | 2h | 6h | 7 |
Ranked by lowest median question score relative to the maximum achievable. Normalizing by question count makes homeworks of different lengths comparable; a lower percentage indicates a harder assignment.
| Rank | Homework | Median score | Score % | Completion |
|---|---|---|---|---|
| 1 | Homework 2: Machine Learning for Regression | 5 / 6 | 83.3% | 43.5% |
| 2 | Homework 3: Machine Learning for Classification | 5 / 6 | 83.3% | 36.3% |
| 3 | Homework 4: Evaluation Metrics for Classification | 5 / 6 | 83.3% | 28.8% |
| 4 | Homework 5: Deploying Machine Learning Models | 5 / 6 | 83.3% | 22.8% |
| 5 | Homework 8: Neural Networks and Deep Learning | 5 / 6 | 83.3% | 14.2% |
| 6 | Homework 9: Serverless Deep Learning | 5 / 6 | 83.3% | 10.6% |
| 7 | Homework 1: Introduction to Machine Learning | 7 / 7 | 100.0% | 68.9% |
| 8 | Homework 6: Decision Trees and Ensemble Learning | 6 / 6 | 100.0% | 16.7% |
| 9 | Homework 10: Kubernetes | 7 / 7 | 100.0% | 7.9% |
Share of submitted answers that were correct, per question. A lower percentage points to a harder or more confusing question. Participation-only questions are excluded.
| Homework | Question | Correct | Answers | % correct |
|---|---|---|---|---|
| Homework 1: Introduction to Machine Learning | Records count | 1588 | 1625 | 97.7% |
| Fuel types | 1560 | 1625 | 96.0% | |
| Missing values | 1503 | 1625 | 92.5% | |
| Max fuel efficiency | 1562 | 1625 | 96.1% | |
| Median value of horsepower | 1188 | 1625 | 73.1% | |
| Sum of weights | 1454 | 1625 | 89.5% | |
| Homework 2: Machine Learning for Regression | Missing values | 1008 | 1025 | 98.3% |
| Median for horse power | 979 | 1025 | 95.5% | |
| Filling NAs | 715 | 1025 | 69.8% | |
| Best regularization | 600 | 1025 | 58.5% | |
| RMSE Standard Deviation | 725 | 1025 | 70.7% | |
| Evaluation on test | 822 | 1025 | 80.2% | |
| Homework 3: Machine Learning for Classification | Mode for industry | 821 | 856 | 95.9% |
| Biggest correlation | 693 | 856 | 81.0% | |
| Biggest MI | 778 | 856 | 90.9% | |
| Accuracy | 701 | 856 | 81.9% | |
| Feature selection | 479 | 856 | 56.0% | |
| Parameter tuning | 602 | 856 | 70.3% | |
| Homework 4: Evaluation Metrics for Classification | ROC AUC feature importance | 608 | 680 | 89.4% |
| Model AUC | 649 | 680 | 95.4% | |
| Precision and recall | 379 | 680 | 55.7% | |
| F1 score | 516 | 680 | 75.9% | |
| 5-Fold CV standard deviation | 411 | 680 | 60.4% | |
| Best C | 506 | 680 | 74.4% | |
| Homework 5: Deploying Machine Learning Models | Lead score (v1) | 493 | 537 | 91.8% |
| Lead score (v2) | 487 | 537 | 90.7% | |
| Docker image size | 492 | 537 | 91.6% | |
| Lead score (v3) | 186 | 537 | 34.6% | |
| Homework 6: Decision Trees and Ensemble Learning | Most important featute | 374 | 393 | 95.2% |
| RMSE on validation | 366 | 393 | 93.1% | |
| Number of estimators | 351 | 393 | 89.3% | |
| Best max_depth | 326 | 393 | 83.0% | |
| Most important feature | 357 | 393 | 90.8% | |
| XGBoost eta | 319 | 393 | 81.2% | |
| Homework 8: Neural Networks and Deep Learning | Loss | 302 | 336 | 89.9% |
| Number of parameters | 307 | 336 | 91.4% | |
| Median of training accuracy for all the epochs | 272 | 336 | 81.0% | |
| Standard deviation of training loss for all the epochs | 214 | 336 | 63.7% | |
| Mean of test loss for all the epochs | 289 | 336 | 86.0% | |
| Average of test accuracy for the last 5 epochs | 296 | 336 | 88.1% | |
| Homework 9: Serverless Deep Learning | Output name | 233 | 249 | 93.6% |
| Image size | 237 | 249 | 95.2% | |
| First R value after pre-processing | 227 | 249 | 91.2% | |
| Model output | 170 | 249 | 68.3% | |
| Docker image size | 149 | 249 | 59.8% | |
| Model output (docker) | 157 | 249 | 63.1% | |
| Homework 10: Kubernetes | Probability of conversion | 166 | 187 | 88.8% |
| Smallest deployable compute unit | 176 | 187 | 94.1% | |
| Service type | 175 | 187 | 93.6% | |
| Command | 178 | 187 | 95.2% |
When homework submissions arrive relative to the deadline, across all assignments.
Homework submissions per week, across all assignments. Bars are scaled to the busiest week.
400 students completed the course and received certificates.