-
Notifications
You must be signed in to change notification settings - Fork 16
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
adding eval through mxeval #3
Conversation
@@ -1,5 +1,5 @@ | |||
import time | |||
from human_eval.data import write_jsonl, read_problems | |||
from mxeval.data import write_jsonl, read_problems, get_data |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Can you update requirements.txt?
llm_cfg/infer.py
Outdated
@@ -25,10 +25,12 @@ | |||
p.add_argument("--quantize", type=bool, default=True) | |||
p.add_argument("--gpu", type=int, default=1) | |||
p.add_argument("--num_samples", type=int, default=1) | |||
p.add_argument("--language", choices = ["python", "go"], default = "python", help = "language") | |||
p.add_argument("--dataset", choices = ["mbxp", "multi-humaneval", "mathqa-x"], default = "mbxp", help = "dataset") |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
let's just call it it humaneval
and make that the default
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Also, just mathqa
llm_cfg/evaluation_mxeval.py
Outdated
@@ -0,0 +1,35 @@ | |||
import sys |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We already have a file doing this called evaluation.py
, can we just modify that file? That file additionally computes types of errors while doing the evaluation.
You don't need language
, dataset
and k
as arguments. language
and dataset
information is present in the filename.
and all pass@k are computed for no extra cost when n>k
?
No description provided.