search

aditya1503 / Siamese-LSTM

Siamese Recurrent Neural network with LSTM for evaluating semantic similarity between sentences.
200 stars 68 forks source link

example1.py - bad similarity results #5

Open keotic opened 8 years ago

keotic commented 8 years ago

Running the example on the given model yields similarity score of around 2 in a scale from 1-5 between the 2 very similar sentences given

aditya1503 commented 8 years ago

"A truly wise man" and "he is smart" have similarity around 2.3 on a scale of 1-5.

rap9430 commented 8 years ago

My result is 3.15449548

liuenda commented 8 years ago

@rap9430 Hi. my results is around 2.7 after careful training. Could do please tell me how can you obtain such high score? Thank you!

jnAILab commented 7 years ago

@rap9430 Could do please tell me how can you obtain such high score? Thank you!

AsmaShaukat commented 6 years ago

Hello everyone. could you find reason of unequal similarity measures on sample examples by running example1.py. I need reported similarity score in paper (A. Siamese Recurrent Architectures for Learning Sentence Similarity) as a main part of my project. kindly help me in figuring out this issue. My obtained similarities using pre-provided weights are
similarityscores