def after_ep_func(epoch):
out_fp = comb_paths(run_hp.output_path,
'ep%d_%s' % (epoch, run_hp.checkpoint_full_fn))
imodel.save_state(out_fp)
gen_folder_path = comb_paths(run_hp.output_path, "output_ep%d" % epoch)
summ_eval(output_folder=gen_folder_path,
data_pipeline=eval_data_pipeline,
eval_data_source=eval_dev_data_source,
summ_gen_func=partial(idev.summ_generator,
summ_post_proc=summ_post_proc),
rev_formatter_func=idev.format_revs,
avg_rouge=True,
sent_splitter=run_hp.sent_split_func,
analytics_func=run_hp.analytics_func)
gen_seqs(data_sources=gen_data_sources,
output_folder=gen_folder_path,
gen_func=partial(idev.gen_and_save_summs))
epochs=run_hp.epochs,
after_epoch_func=after_ep_func)
logger.info("Total time elapsed %f (s) " % (time() - start))
imodel.save_state(
comb_paths(run_hp.output_path, run_hp.checkpoint_full_fn))
# AFTER TRAINING PROCEDURES #
gen_folder_path = comb_paths(run_hp.output_path, "output")
summ_eval(output_folder=gen_folder_path,
eval_data_source=eval_test_data_source,
summ_gen_func=partial(idev.summ_generator,
summ_post_proc=summ_post_proc),
data_pipeline=eval_data_pipeline,
rev_formatter_func=idev.format_revs,
avg_rouge=True,
sent_splitter=run_hp.sent_split_func,
analytics_func=run_hp.analytics_func)
gen_seqs(data_sources=gen_data_sources,
output_folder=gen_folder_path,
gen_func=partial(idev.gen_and_save_summs))
else:
# inference procedure where summaries are generated for reviews in CSV
# files
infer_bsz = parser_args.infer_batch_size
infer_inp_file_path = parser_args.infer_input_file_path
out_file_name = get_file_name(infer_inp_file_path)
infer_out_file_path = comb_paths(run_hp.output_path,