Skip to content
Projects
Groups
Snippets
Help
This project
Loading...
Sign in / Register
Toggle navigation
codecritic
Overview
Overview
Details
Activity
Cycle Analytics
Repository
Repository
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Charts
Issues
0
Issues
0
List
Board
Labels
Milestones
Merge Requests
0
Merge Requests
0
CI / CD
CI / CD
Pipelines
Jobs
Schedules
Charts
Wiki
Wiki
Snippets
Snippets
Members
Members
Collapse sidebar
Close sidebar
Activity
Graph
Charts
Create a new issue
Jobs
Commits
Issue Boards
Open sidebar
Ziyuan Nan
codecritic
Commits
83e4a159
Commit
83e4a159
authored
Oct 07, 2024
by
nzy
Browse files
Options
Browse Files
Download
Email Patches
Plain Diff
step2 update prefrence pair's format
parent
bd555b17
Hide whitespace changes
Inline
Side-by-side
Showing
2 changed files
with
6 additions
and
4 deletions
+6
-4
step2_prepare_preference_dataset.py
+2
-2
utils_preference_dataset.py
+4
-2
No files found.
step2_prepare_preference_dataset.py
View file @
83e4a159
...
@@ -117,10 +117,10 @@ if __name__ == "__main__":
...
@@ -117,10 +117,10 @@ if __name__ == "__main__":
all_edit_distance_pairs
,
10
*
1000
,
5
,
is_max
=
False
all_edit_distance_pairs
,
10
*
1000
,
5
,
is_max
=
False
)
)
dataset_info
=
mk_dataset_info
(
"apps_min_edit_distance_prefrence"
)
dataset_info
=
mk_dataset_info
(
"apps_min_edit_distance_prefrence"
)
save_json
l
(
save_json
(
metadata
,
cfg
[
"preference_dataset"
][
"min_edit_distance"
][
"metadata_path"
]
metadata
,
cfg
[
"preference_dataset"
][
"min_edit_distance"
][
"metadata_path"
]
)
)
save_json
l
(
save_json
(
preference_pairs
,
preference_pairs
,
cfg
[
"preference_dataset"
][
"min_edit_distance"
][
"preference_dataset_path"
],
cfg
[
"preference_dataset"
][
"min_edit_distance"
][
"preference_dataset_path"
],
)
)
...
...
utils_preference_dataset.py
View file @
83e4a159
...
@@ -19,12 +19,14 @@ def mk_dataset_info(dataset_name):
...
@@ -19,12 +19,14 @@ def mk_dataset_info(dataset_name):
}
}
}
}
# see utils.extract_code
code_template
=
r"```python{}```"
def
mk_preference_pair
(
instruction
,
chosen_code
,
rejected_code
):
def
mk_preference_pair
(
instruction
,
chosen_code
,
rejected_code
):
return
{
return
{
"messages"
:
[
"messages"
:
[
{
"role"
:
"user"
,
"content"
:
instruction
},
{
"role"
:
"user"
,
"content"
:
instruction
},
],
],
"chosen"
:
{
"role"
:
"assistant"
,
"content"
:
c
hosen_code
},
"chosen"
:
{
"role"
:
"assistant"
,
"content"
:
c
ode_template
.
format
(
chosen_code
)
},
"rejected"
:
{
"role"
:
"assistant"
,
"content"
:
rejected_code
},
"rejected"
:
{
"role"
:
"assistant"
,
"content"
:
code_template
.
format
(
rejected_code
)
},
}
}
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment