Hello :) Today is Day 20!
A quick summary of today:
To do NLP, first, you need to tokenize the data and use pad_sequence
Today, I learned about the sequence model.
I tried different models
And another one
Next, I tried making a next-word predictor model
Base model
2nd model
3rd model
Among them, the Base model obtained the highest acuity and lowest loss.
I am still new to NLP, so I need to practice with new data on kaggle.
That is all for today!
See you tomorrow :)