Sharing this observation as it initially caused me a small touch of confusion in case may be beneficial to other students. In the lesson4-imdb notebook, the 11th cell derives the lengths of several objects. One of those objects is described in the preceding text as “# tokens in the training set”. Based on the lecture address of this point I believe this item would be better described as “number of tokenized training sets” (of which there is one).
gnavink (Navin Kumar) #2
Yes. the wordings could have been better. Can it also be said as the number of items/samples in the training set …