floating point error in encoding

[satarupaguha11]

I have successfully trained skipthoughts from scratch on my own data. Now given new data X, I want to get its vectors representation. So i do the following

vectors = skipthoughts.encode(model, X)

But this gives me the following error:

floating point exception (core dumped)

I have also set the THEANO_FLAGS floatX to float32. Any help will be appreciated.

[lovekesh-thakur]

Not an expert but most likely you're getting memory issues. i think you are running out of memory.

[polyrhythmatic]

Having the same issue on a GTX Titan X and 32 gigs of ram with a rather small sentence corpus (around 37k sentences). Any idea what the problem could be? I'm not getting any information on the error.

Flags I am using

THEANO_FLAGS=floatX=float32,device=gpu,nvcc.flags=-D_FORCE_INLINES

I can't run more than 1000 sentences at at a time. Any idea what is happening here?

[clovercoser]

X should be a list, not string. You may forget '[ ]'.

[pcallier]

We got this behavior sometimes from inadvertantly sending in empty strings (after overzealous preprocessing). It was always hard to pinpoint so we're not sure we fixed it but that might be something to check.

[Spider101]

I got the same error. Turns out one of the sentences in the list was just whitespace, i.e. " ". So, after the internal preprocessing in the skipthoughts.py script, it came up to be an empty string which might have been the reason for the error. Hope this helps!

[pavanpankaj]

Traceback (most recent call last): File "train.py", line 8, in vectors=encoder.encode(sent_tok) File "/local_home/rpr/code/skipthoughts.py", line 102, in encode return encode(self._model, X, use_norm, verbose, batch_size, use_eos) File "/local_home/rpr/code/skipthoughts.py", line 132, in encode for minibatch in range(numbatches):

please help me

[pavanpankaj]

3 128 Traceback (most recent call last): File "train.py", line 8, in vectors=encoder.encode(sent_tok) File "/local_home/rpr/code/skipthoughts.py", line 102, in encode return encode(self._model, X, use_norm, verbose, batch_size, use_eos) File "/local_home/rpr/code/skipthoughts.py", line 132, in encode for minibatch in range(numbatches):

I thought only 3 words in dictionary are found. In batches it is going on and the batch size is 128. only 3 in a batch found so 3/128 is float range(float) is an error it is showing