Comments (5)
It's OK.
If you want to use other dataset, use preprocess.py to process the dataset into suitable format and store them in particular directory.
preprocess.py assumes that the format of dataset is "userID movieID rating timestamp(optional)".
It collects every movie each users watched and make lists like
user1 movie1 rating1 movie2 rating2 ...
user2 movie10 rating10 movie11 rating11 ...
...
Then, you can use them to train and test.
from improving-rnn-recommendation-model.
Thanks very much!
Your code little difficult for me, I have some other questions, all these questions LSTM
1, What is the time length of the input?
2, How many LSTM layer stack together? and how many neurons?
3, What is your specific network structure? for example, input_layer ( ? * ? ) --> Embedding layer( ? * ?)---> LSTM layer( ? * ? ) -->Dropout layer ---> Dense layer---> output
4, What is the output format?
5, If the input does not have rating, just have user_id and movie_id, it is ok to train?
6, Do you have code only about LSTM, can you send to me? Email: [email protected]
Thanks!!
from improving-rnn-recommendation-model.
hi, are you still here?
from improving-rnn-recommendation-model.
- maximum input length is set to 30 in default. You can modify with --max_length
- You can set the number of stack with --r_l. For example, 100-50 value means 2 stacked layers, and the dimension of first lstm is 100 and that of second lstm is 59.
- Input layer - Embedding layer - lstm layer - dense layer - output.
There's no dropout layer.
Each input is one-hot form. Which means the size of an input vector is the number of items in dataset.
It looks like i1 = [0, 0, 1, 0, 0 .....]
And total input will be like i1, i2, i3... which mean the movies an user watched.
You can set the size of embedding vector by --r_emb. - The output format is a vector each dimension of which means how possible corresponding movie will be watched next. Its size is the number of movies. The 10 movies that have highest possibility score will be recommended.
- Its ok.
- I can't understand the meaning of 'code only about LSTM' in your question. See recurrent_layers.py for codes related to tensorflow LSTM. If you want to know about raw lstm codes implemented in numpy or codes of lstm in tensorflow, you'd better use google.
from improving-rnn-recommendation-model.
thanks very much
from improving-rnn-recommendation-model.
Related Issues (8)
- got an error during the train HOT 6
- model image
- get 'ValueError: invalid literal for int() with base 10' message HOT 5
- there is no attribute or method named "self.tarining" HOT 3
- suggestions and questions on training the data HOT 2
- May I know the command line parameter when running LSTM_weight_tying(dim:100) HOT 5
- when enabled --l_emb , then testing the model it will raise an excption.
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from improving-rnn-recommendation-model.