S2V
S2V copied to clipboard
lajanugen
ICLR 2018 Quick-Thought vectors
Hi, I am interested in comparing to your QuickThoughts method by evaluating it on the full SentEval benchmark. To do that I need to write something like the following: ```python...
Trained the model using the existing code but when trying to load the model using Skip-thought code: ``` encoder.load_model(configuration.model_config(bidirectional_encoder=False), vocabulary_file=VOCAB_FILE, embedding_matrix_file=EMBEDDING_MATRIX_FILE, checkpoint_path=CHECKPOINT_PATH) ``` It's throwing me exception: NotFoundError (see above...
The instructions are very vague. It's not even clear if this project is self-containing. In `eval.sh`, there is a parameter for the link to the directory of SkipThoughts implementation. Does...
It is very incredible that the version using pre-trained word embedding is worse than the one using random word embedding. I don't know if I had wrong configurations, configurations are...
I'm trying to freeze the Quick-thoughts model but am having trouble identifying the input and output nodes. I believe the output node should be "word_embeddings" but am not 100% sure....
I would like to know what you loss values during the training are. For me the loss start at around 4-5 and then decreases to around 2 after some minutes....
Can you please provide us with the glove.840B.300d.npy file that you used with the pre-trained GloVe word embeddings?
I can successfully run the model on a virtual machine with GPUs but am unable to run the same code on my local computer using a CPU. I am getting...