Datasets:
Tasks:
Text Generation
Formats:
parquet
Languages:
English
Size:
10K - 100K
ArXiv:
Tags:
code
License:
fix a typo
Browse files
README.md
CHANGED
@@ -62,7 +62,7 @@ tags:
|
|
62 |
|
63 |
## Introduction
|
64 |
|
65 |
-
This dataset presents the **Java** portion of [RepoBench](https://arxiv.org/abs/2306.03091) v1.1 (ICLR 2024). The data encompasses a collection from GitHub, spanning the period from **October 6th to
|
66 |
|
67 |
## Resources and Links
|
68 |
|
@@ -121,7 +121,7 @@ This dataset presents the **Java** portion of [RepoBench](https://arxiv.org/abs/
|
|
121 |
|
122 |
def construct_prompt(
|
123 |
data: dict,
|
124 |
-
language: str = "
|
125 |
tokenizer= None,
|
126 |
max_token_nums: int = 15800
|
127 |
) -> str:
|
@@ -167,7 +167,7 @@ This dataset presents the **Java** portion of [RepoBench](https://arxiv.org/abs/
|
|
167 |
break
|
168 |
|
169 |
# join the lines back
|
170 |
-
cross_file_prompt = "\n".join(cross_file_prompt_lines[:i
|
171 |
|
172 |
# combine the cross-file prompt and in-file prompt
|
173 |
prompt = cross_file_prompt + in_file_prompt
|
|
|
62 |
|
63 |
## Introduction
|
64 |
|
65 |
+
This dataset presents the **Java** portion of [RepoBench](https://arxiv.org/abs/2306.03091) v1.1 (ICLR 2024). The data encompasses a collection from GitHub, spanning the period from **October 6th to December 31st, 2023**. With a commitment to data integrity, we've implemented a deduplication process based on file content against the Stack v2 dataset (coming soon), aiming to mitigate data leakage and memorization concerns.
|
66 |
|
67 |
## Resources and Links
|
68 |
|
|
|
121 |
|
122 |
def construct_prompt(
|
123 |
data: dict,
|
124 |
+
language: str = "java",
|
125 |
tokenizer= None,
|
126 |
max_token_nums: int = 15800
|
127 |
) -> str:
|
|
|
167 |
break
|
168 |
|
169 |
# join the lines back
|
170 |
+
cross_file_prompt = "\n".join(cross_file_prompt_lines[:i]) + "\n\n"
|
171 |
|
172 |
# combine the cross-file prompt and in-file prompt
|
173 |
prompt = cross_file_prompt + in_file_prompt
|