hello,
i want to use LSTM for below data structure. i want to predict next rating of review.
first: business and rating (1-5 stars)
Business_1 = [1, 3, 4, 5, 2, 4, 1, 1, 4, 5, 2, 1] => length = 12
Business_2 = [3, 3, 2, 2, 1, 5, 4] => length = 7
Business_3 = [1, 3] => length = 2
how I can use nn.LSTM when the length of data is different? if I set L= 3 for Business_3 is not ok!!
i have 2 problems.
- set L for LSTM
- different length of data for each business. i want to use one LSTM model for all business.
my second project is like it but with emotion.( 5 categorial emotion = [happy, sad, neutral , angry, fear]
Business_1 =
{review1: happy:0.4, sad:0.1, fear:0.0, angry:0.2, neutral: 0.3
review2: happy:0.0, sad:0.1, fear:0.0, angry:0.1, neutral: 0.7
review3: happy:0.6, sad:0.0, fear:0.0, angry:0.2, neutral: 0.2
}
Business_2=
{review1: happy:0.4, sad:0.1, fear:0.0, angry:0.2, neutral: 0.3
review2: happy:0.3, sad:0.5, fear:0.1, angry:0.1, neutral: 0.0
review3: happy:0.4, sad:0.0, fear:0.1, angry:0.1, neutral: 0.4
}
i want to predict the type of emotion and the amount of it for next review.
thank you very much for your response from now.