How will I train the model and get the pickle files

[archana3194]

IN WHICH FOLDER I NEED TO KEEP THE WIKI-DBPEDIA_TRAIN.PY @YueLiu

[lalppaul]

Could you please answer this?

[YueLiu]

Hi @archana3194 @lalppaul , there were two tf generated files you got after training the model. because they were too large so I could not upload them. you should be able to get them once you ran the end-end training. I suggest you try to use seq2seq/s2s_preprocess.py to generate the .pkl file. there s a local path needed for pretrained word2vec. Then you may try baselines/main.py and see if your input dimension is valid. (you might need to swap the loss.data[0] to loss.item() in main.py due to the recent update in pytorch)

[YueLiu]

and of course, you need to prepare your training corpus following this pattern

[lalppaul]

Thanks for the response. I ran the s2s_preprocess.py file and generated the 20180405.pkl file. After that i tried running the main.py from baselines and found out that i was missing train, test and valid pkl files. Am i missing any steps in between. Also, could you please provide the training corpus? Thanks in advance!!

[petrLorenc]

@YueLiu I have the same problem - could you please update the instruction how to run your code?

[mitchelldehaven]

Has anyone gotten this code working that could give the steps required to get it training?