Debanjan Chaudhuri (Deep)
Debanjan Chaudhuri (Deep)
try to set the proxy overriding the method
Hey, am just too busy lately with stuff in general, Would you like to be the contributor in this repo? I could do with some help. Also, need to update...
for me the training took 22 mins/epoch and evaluation around a minute your error suggests that you don't have enough ram on the machine you are running on
yes, we have a custom preprocessor in place, can you try changing it to multi-threaded? in preprocess,py I can try to do it too when I have some time.
The error is in the register parameter in pytorch_model.py Removing them the model works fine.
the model keeps the pre-processed data into memory, I don't think decreasing the batch size will help. can you post the full error stack trace?
no, looks like you are running it on cpu, did you use the argument for running it on the gpu?
maybe check your pytorch gpu configuration/version properly with this? In [1]: import torch In [2]: torch.cuda.current_device() Out[2]: 0 In [3]: torch.cuda.device(0) Out[3]: In [4]: torch.cuda.device_count() Out[4]: 1 In [5]: torch.cuda.get_device_name(0)...
cool that you could run it. Can you please change loss.data[0] to loss.item() and provide the results?
provide the model with a question utterance and a set of possible response, the output would be a predicted response.