Let A be an embedding matrix, and let o 4567 O_4567, end subscript be a one-hot vector corresponding to word 4567. Then to get the embedding of word 4567, why don’t we call A ∗ o 4567 in Python?
None of the answers are correct: calling the Python snippet as described above is fine. (selected this) → But the grader tells this is a wrong answer
I see that the above formula is correct what am I missing?
It looks like you are asking about a quiz question. I’ll need a bit more info to figure out what you are asking, but a couple of points here:
- Quiz questions are treated the same way as solution source code, meaning that we aren’t supposed to show the answers in public. So if that’s what you’re asking, we’ll need to switch to a DM thread.
- As mentioned above, I’m not really clear on what the question really is, but note that the operation between an embedding matrix and a one hot vector should be a dot product style multiply, right? Remember that Prof Ng always uses the notation “*” for elementwise multiply. But it’s been a couple of years since I watched the lectures for C5 W2, so it’s entirely possible I’m just missing your point.
