Update README.md
Browse files
README.md
CHANGED
|
@@ -18,6 +18,9 @@ jm52m is a dataset we created containing 52m Java methods from 52k Java projects
|
|
| 18 |
|
| 19 |
---
|
| 20 |
## jm52m dataset details
|
|
|
|
|
|
|
|
|
|
| 21 |
| Config | Value |
|
| 22 |
| ------- | ------- |
|
| 23 |
|number of tokens | 8,752,695,577|
|
|
@@ -26,5 +29,5 @@ jm52m is a dataset we created containing 52m Java methods from 52k Java projects
|
|
| 26 |
|number of projects | 52,933 |
|
| 27 |
|megabytes after processing |16,695 |
|
| 28 |
|
| 29 |
-
|
| 30 |
|
|
|
|
| 18 |
|
| 19 |
---
|
| 20 |
## jm52m dataset details
|
| 21 |
+
|
| 22 |
+
We provide the size of our dataset in the following table:
|
| 23 |
+
|
| 24 |
| Config | Value |
|
| 25 |
| ------- | ------- |
|
| 26 |
|number of tokens | 8,752,695,577|
|
|
|
|
| 29 |
|number of projects | 52,933 |
|
| 30 |
|megabytes after processing |16,695 |
|
| 31 |
|
| 32 |
+
We tokenize our data using scripts provided in our [github repository](https://github.com/apcl-research/jam/blob/main/data/jam_jm52m/prepare_fc_raw.py).
|
| 33 |
|