Find out!
Across
- 6. m×n matrix M. what is nxn matrix called?
- 7. ignores all but the most “active”
- 9. number of batches needed to complete one epoch
- 10. gradient of a function at saddle point
- 11. measures how good you are
- 12. ranges from ZERO to INFINITY
Down
- 1. layers which learn spatial features in CNN
- 2. low bias makes models to ..........
- 3. assigned prior to training
- 4. opposite of vanish
- 5. Word embedding technique
- 7. solves problems of border effects
- 8. I can improve or decrease the updates