Sde203 build experience
Key Topics
- Container (
docker
) - Testing (Unit Test, Integtaion Test, E2E Test, Load Test)
- Runbook - https://www.pagerduty.com/resources/learn/what-is-a-runbook/
Model Bias Wape
Lane | OD pair | Predicted price | Actual price |
---|---|---|---|
1 | LAX to JFK | $400 | $420 |
2 | ORD to SFO | $600 | $550 |
3 | DFW to LGA | $700 | $750 |
Eve: Hey Adam, do you know how to evaluate the accuracy of a linear regression model?
Adam: Yeah, one way is to use metrics like WAPE and bias in percentage.
Eve: WAPE? What's that?
Adam: WAPE stands for Weighted Absolute Percentage Error. It's a metric used to measure the accuracy of a model's predictions in terms of percentage error.
Eve: Oh, I see. Can you explain how to calculate it?
Adam: Sure. You take the absolute percentage error for each prediction and then take the weighted average of those errors using the actual values as the weights.
Eve: That makes sense. And what about bias in percentage?
Adam: Bias in percentage is another metric used to evaluate the accuracy of a model. It measures the average deviation of the predicted values from the actual values as a percentage of the average actual value.
Eve: Interesting. How do you calculate bias in percentage?
Adam: You take the difference between the average predicted value and the average actual value, divide it by the average actual value, and then multiply by 100 to express it as a percentage.
Eve: Thanks for explaining that, Adam. I'll keep those metrics in mind when evaluating linear regression models in the future.
Python memory_profiler
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 |
|
Python pandas CSV/TSV
Read large csv from s3 use smart_open
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 |
|
Read json from env and handle logic as pandas dataframe
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 |
|