This repository was archived by the owner on Apr 11, 2023. It is now read-only.
File tree Expand file tree Collapse file tree 2 files changed +3
-3
lines changed Expand file tree Collapse file tree 2 files changed +3
-3
lines changed Original file line number Diff line number Diff line change 37
37
38
38
``` bash
39
39
# clone this repository
40
- git clone https://github.com/ml-msr- github/CodeSearchNet.git
40
+ git clone https://github.com/github/CodeSearchNet.git
41
41
cd CodeSearchNet/
42
42
# download data (~3.5GB) from S3; build and run the Docker container
43
43
script/setup
Original file line number Diff line number Diff line change 30
30
This script generates ranking results over the CodeSearchNet corpus for a given model by scoring their relevance
31
31
(using that model) to 99 search queries of the CodeSearchNet Challenge. We use cosine distance between the learned
32
32
representations of the natural language queries and the code, which is stored in jsonlines files with this format:
33
- https://github.com/ml-msr- github/CodeSearchNet#preprocessed-data-format. The 99 challenge queries are located in
34
- this file: https://github.com/ml-msr- github/CodeSearchNet/blob/master/resources/queries.csv.
33
+ https://github.com/github/CodeSearchNet#preprocessed-data-format. The 99 challenge queries are located in
34
+ this file: https://github.com/github/CodeSearchNet/blob/master/resources/queries.csv.
35
35
To download the full CodeSearchNet corpus, see the README at the root of this repository.
36
36
37
37
Note that this script is specific to methods and code in our baseline model and may not generalize to new models.
You can’t perform that action at this time.
0 commit comments