Omer Arshad

Results 11 comments of Omer Arshad

HFSaveCheckpoint does not work with deepspeed

getting this error while loading a language model ```py RuntimeError: Error(s) in loading state_dict for LanguageModelingTransformer: Missing key(s) in state_dict: "model.lm_head.weight". ``` any leads?

SSLError

any workaround for this?

which attention architecture is used in NER?

well in my experiments attention only models achieve comparable results to LSTM, even got better than LSTM with very less training time

which attention architecture is used in NER?

yes, structure of my model is attention+crf only

It shows only one label result

Did you find a solution to this? I am also facing same issue

will this also work for "https://tfhub.dev/google/universal-sentence-encoder/2"

oops i was talking about "https://tfhub.dev/google/universal-sentence-encoder-large/2" i.e. transformer based

will this also work for "https://tfhub.dev/google/universal-sentence-encoder/2"

Great, So what is the best way to fine tune trainable transformer based USE?

any idea how to use sentence encoder as Siamese architecture?

Can you please guide me where to use above mentioned code ? Below is sniped taken from example , X is first input and X2 is second input with tf.Session(graph=graph)...

The Fragments which form the part of steps, expand according to the amount of widget or data contained in them

yes this happens. please fix it

the problems of word embedding

At that time, they were available on fasttext website