Questions score
- Min
- 1
- Median
- 6.0
- Max
- 6
- Q1
- 4.0
- Avg
- 5.3
- Q3
- 6.0
MLOps Zoomcamp 2025
Homework 5: Monitoring Statistics
Distribution of scores and reported study time for this homework.
Submissions
207
Median total score
6
Average total score
6
Score distribution
All values are points.
Learning in public score
- Min
- -
- Median
- 0.0
- Max
- 7
- Q1
- 0.0
- Avg
- 0.7
- Q3
- 0.0
Total score
- Min
- 1
- Median
- 6.0
- Max
- 14
- Q1
- 5.0
- Avg
- 6.1
- Q3
- 6.0
Time distribution
All values are hours reported by students.
Lectures
- Min
- 1.0
- Median
- 4.0
- Max
- 44.0
- Q1
- 3.0
- Avg
- 5.8
- Q3
- 6.0
Homework
- Min
- 0.5
- Median
- 3.0
- Max
- 72.0
- Q1
- 2.0
- Avg
- 5.1
- Q3
- 5.0
Question breakdown
Correctness and answer distribution per question.
1. Prepare the dataset
193 / 207 correct (93.2%)
1
72044
5 (2.4%)
2
78537
7 (3.4%)
3
57457
193 (93.2%)
4
54396
2 (1.0%)
2. Metric
207 / 207 correct (100.0%)
| Answer | Count |
|---|---|
| ColumnQuantileMetric | 25 |
| ColumnSummaryMetric | 13 |
| QuantileValue | 13 |
| ColumnMissingValuesMetric | 9 |
| trip_distance | 5 |
| ColumnValueRangeMetric | 5 |
| DatasetSummaryMetric | 4 |
| ColumnQuantileMetric(column_name="fare_amount", quantile=0.5) | 4 |
| ColumnDriftMetric | 4 |
| StdValue | 3 |
| ColumnCorrelationsMetric | 3 |
| MedianValue | 2 |
| ColumnMeanMetric | 2 |
| UniqueValueCount | 2 |
| DatasetDriftMetric | 2 |
| ColumnQuantileMetric(column_name='fare_amount', quantile=0.5) | 2 |
| QuantileValue() | 2 |
| QuantileValue, MissingValueCount | 2 |
| passenger_count | 2 |
| DatasetCorrelationsMetric | 2 |
| ColumnCorrelationsMetric(column_name='fare_amount') | 2 |
| RegressionQualityMetric() | 2 |
| MissingValueCount | 2 |
| ColumnQuantileMetric(column_name='fare_amount', quantile=0.5) and Metric DatasetCorrelationsMetric | 2 |
| DatasetMissingValuesMetric | 2 |
| 0.01 | 1 |
| prediction | 1 |
| RegressionQualityMetric(target="duration_min", prediction="prediction") | 1 |
| InRangeValueCount() | 1 |
| 3.805 minutes, 4.142 minutes | 1 |
| ColumnDriftMetric, DatasetDriftMetric, DatasetMissingValuesMetric, ColumnValueRangeMetric | 1 |
| DataQualityStabilityMetric | 1 |
| Missing Values Analysis | 1 |
| correlation between the prediction column and the trip_distance column | 1 |
| ColumnQuantileMetric(column_name='fare_amount', quantile=0.5), ColumnMissingValuesMetric(column_name='passenger_count') | 1 |
| MaxValue | 1 |
| ColumnQuantileMetric for fare_amount (quantile=0.5) | 1 |
| the mean of fare_amount | 1 |
| quantile_metric = QuantileValue(column="fare_amount", quantile=0.5) | 1 |
| evidently.metrics.ColumnQuantileMetric | 1 |
| ColumnSummaryMetric of the trip_distance | 1 |
| Mean of trip_distance | 1 |
| 2 | 1 |
| mean | 1 |
| Most common passenger count (taken from ColumnSummaryMetric for passenger count) | 1 |
| MaxValue(column="fare_amount") | 1 |
| ColumnQuantileMetric and DatasetCorrelationsMetric | 1 |
| Mean value fot `tip_amount` | 1 |
| DuplicatedRowCount | 1 |
| ColumnDriftMetric(column_name='trip_distance'), | 1 |
| StdValue, QuantileValue(column='fare_amount', quantile=0.5), | 1 |
| QuantileValue(column="total_amount", quantile=0.5) | 1 |
| fare_amount | 1 |
| UniqueValueCount for categorical variables | 1 |
| ColumnQuantileMetric, ColumnMissingValuesMetric | 1 |
| 0.5 | 1 |
| MeanValue and MissingValue | 1 |
| RegressionErrorNormality | 1 |
| • I chose to monitor the median (50th percentile) of the fare_amount column using the ColumnQuantileMetric. | 1 |
| We are monitoring the prediction latency (ms) of the model. | 1 |
| 96 | 1 |
| Fare Amount Quantile (0.5) - The median fare amount for each day, Column Drift - Detects statistical drift in the fare_amount column | 1 |
| I chose Standard deviation for fare_amount. It will be followed with "StdValue(column_name="fare_amount")". | 1 |
| Used ColumnQuantileMetric + ColumnMissingValuesMetric | 1 |
| I chose: ColumnQuantileMetric and ColumnMissingValuesMetric | 1 |
| ValueDrift(column="prediction") | 1 |
| ColumnValueListMetric | 1 |
| 0,5 | 1 |
| DatasetCorrelationsMetric() | 1 |
| ColumnDriftMetric(column_name='prediction'), DatasetDriftMetric(), DatasetMissingValuesMetric(), ColumnQuantileMetric(column_name="fare_amount", quantile=0.5), ColumnQuantileMetric(column_name="trip_distance", quantile=0.5), ColumnQuantileMetric(column_name="passenger_count", quantile=0.5), ColumnMissingValuesMetric(column_name="fare_amount"), ColumnDriftMetric(column_name="fare_amount"), ColumnDriftMetric(column_name="trip_distance"), ColumnDriftMetric(column_name="passenger_count") | 1 |
| InRangeValueCount(column="fare_amount", left= -259, right=0) | 1 |
| RowCount | 1 |
| QuantileValue, ValueDrift | 1 |
| MaxValue(column='fare_amount') | 1 |
| DriftedColumnsCount | 1 |
| ColumnDriftMetric("fare_amount") | 1 |
| QuantileValue(column="trip_distance", quantile=0.95), | 1 |
| ColumnQuantileMetric(column_name='fare_amount', quantile=0.5), ColumnMissingValuesMetric(column_name='fare_amount') | 1 |
| assuming this is asking to add QuantileValue(column=fare_amount,quantile=0.5) as well as an additional metric. Another useful metric could be evidently.metrics.MeanValue on trip_distance (MeanValue(column=trip_distance)) | 1 |
| MinValue | 1 |
| UniqueValueCount for categorical columns. This is to ensure that onehotencoding always has the same number of columns generated. If a new value of PULocationId or DOLocationId exists, the model must be retrained. | 1 |
| StdValue(column='fare_amount') | 1 |
| StdValue() | 1 |
| MaxValue(column='fare_amount') - maximum fare amount | 1 |
| DatasetMissingValueCount() | 1 |
| I chose the QuantileValue metric on the column fare_amount with quantile=0.5. | 1 |
| R2Score | 1 |
| QuantileValue + MissingValueCount | 1 |
| RMSE | 1 |
| DriftedColumnsCount() | 1 |
| MeanValue metric | 1 |
| MissingValueCount(column='tip_amount') | 1 |
| InRangeValue | 1 |
| I have added ColumnValueRangeMetric to see how many values fall in and out of a certain range (fare_amount between 0 and 25) | 1 |
| ColumnQuantileMetric(column_name="passenger_count", quantile=0.50) # median passengers | 1 |
| ColumnValueRangeMetric(column_name='fare_amount', left=1, right=200) | 1 |
| StdValue of prediction | 1 |
| StdValue(column='prediction') | 1 |
| rmse | 1 |
| ValueDrift() | 1 |
| MeanValue('trip_distance') | 1 |
| MedianValue() | 1 |
| Metric: dataset summary (e.g. almost duplicated features = 1). Quantile value for fare_amount: 13.5. | 1 |
| DatasetSummaryMetric() | 1 |
| ColumnMissingValuesMetric for the "trip_distance" column, and quantile=0.5 | 1 |
| ValueDrift(column="prediction"), DriftedColumnsCount(), MissingValueCount(column="prediction"), QuantileValue(column="fare_amount", quantile=0.5), RowCount(), EmptyRowsCount() | 1 |
| I chose ColumnQuantileMetric with quantile = 0.5 for the fare_amount column. | 1 |
| MeanValue | 1 |
| 13.5 | 1 |
3. Monitoring
169 / 207 correct (81.6%)
1
10
6 (2.9%)
2
12.5
15 (7.2%)
3
14.2
169 (81.6%)
4
14.8
14 (6.8%)
4. Dashboard
177 / 207 correct (85.5%)
1
project_folder (05-monitoring)
2 (1.0%)
2
project_folder/config (05-monitoring/config)
23 (11.1%)
3
project_folder/dashboards (05-monitoring/dashboards)
177 (85.5%)
4
project_folder/data (05-monitoring/data)
2 (1.0%)
Calculated: 14 July 2025, 17:19