11 Checkpoint: Signals, CNN and CNN

Content and key concepts

Progress is available with JavaScript. View is not equal to development.
Eight reviews of the mechanism
Threshold: 7 out of 8. without a timer. Balls are given for answers, not for visits. This is a local self-check; the decisions will be shown after dispatch.
1. Prifs16cosine3Hz and 13Hz give...
Coupling samples phase0
Always different samples
Just zeros.
2. In the sliding-window of the conversion core...
Turns over.
Undressed
Mandatoryly normalized to the sum1
3. CNN-convolution usually calculates...
Cross-correllation without a revolution of the stored core
DFT automatically
Cosine similarity mandatory
4. For liner control via DFT, it is necessary...
Enough with zeros.
Always cut to length
Ignore Integrated Phase
5. The length of the full convolution for N4, K2?
Your numerical response
6. First valid control forx=[1.2.3.4],h=[1-1]?
Your numerical response
7. First 2D output H05fixture0×0+1×1+3×2+4×3?
Your numerical response
8. CNN: Cin3,Cout2,kernel3×3,cbias?
Your numerical response
The result will appear after all the answers have been sent.
Research and explain
Compare convolution and correlation on an unsymmetric core, then check the specific product with zero-pading. In CNN, separate fixed kernels feature exponer from the learner1×2filter. Explain why the disappearance of training loss on one line is not an evaluation of geneticisation.
First, shape the hypothesis, then change one parameter. Compare the numbers to the formula; state the extreme case and the limit of applicability. The free conclusion is not evaluated by a dummy automatic score.