search

zhongkaifu / RNNSharp

RNNSharp is a toolkit of deep recurrent neural network which is widely used for many different kinds of tasks, such as sequence labeling, sequence-to-sequence and so on. It's written by C# language and based on .NET framework 4.6 or above versions. RNNSharp supports many different types of networks, such as forward and bi-directional network, sequence-to-sequence network, and different types of layers, such as LSTM, Softmax, sampled Softmax and others.
BSD 3-Clause "New" or "Revised" License
285 stars 91 forks source link

(Request) Make code more easy to use #5

Open thelemontree10 opened 8 years ago

thelemontree10 commented 8 years ago

Hello zhongkaifu,

Your framework is very great but there are something I think you should change a little to get your code more easy to use. Can you change your code in the way that it can receive directly a matrix, a jagged array as input, have a train and predict method like here ? (The example)

http://accord-framework.net/docs/html/T_Accord_MachineLearning_VectorMachines_MulticlassSupportVectorMachine.htm

With this change, I think you will make a big milestone.

Regards

zhongkaifu commented 8 years ago

Thanks for your suggestion. I'm thinking about refactoring each neural network layer to abstract interface. If you have any idea about it, please let me know. Thanks.