Closed varun-tandon closed 3 years ago
Hi @varun-tandon
Thanks for the question! So basically, we use EASSE for evaluation, see this line: https://github.com/facebookresearch/access/blob/7b61fbf0bad665798d662e0a90d2a0e451367df6/access/evaluation/general.py#L8
According to EASSE readme, you can use your custom test set by specifying it in the command line with --test_set custom --orig_sents_path {path_to_source} --refs_sents_paths {paths_to_references} --sys_sents_path {path_to_prediction}
.
You can also do it programmatically with the method evaluate_system_output
which takes arguments with the same name.
Tell me if that works!
Thanks so much @louismartin! I'll try this out and let you know how it goes 😀
Hi @louismartin your suggestion worked perfectly. Thanks for the help!
Hi @louismartin, thanks so much for providing such a usable resource and responding to issues. I really appreciate it!
So sorry if this is a simple question: what/where is the
turkcorpus_{phase}_legacy
, which is referenced inaccess/evaluation/general.py
?For some context, I'm trying to modify
scripts/evaluate.py
to evaluate on some of my own data. To do so, I modifiedgeneral.py
inaccess/evaluation
, specifically allowing for a directory parameter forget_prediction_on_turkcorpus
andget_prediction_on_turkcorpus
.I don't quite understand the first parameter to
evaluate_system_output
, which you have set tof'turkcorpus_{phase}_legacy'
. I attempted to replace this with my own .simple file, but when I try this I get the following error:I've also tried looking for a directory with the
turkcorpus_test_legacy
name, but to no avail.Thanks so much for the help, and please let me know if there is any additional information I can provide to clarify my problem.