Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
GPyT is a GPT2 model trained from scratch (not fine tuned) on Python code from Github. Overall, it was ~80GB of pure Python code, the current GPyT model is a mere 2 epochs through this data, so it may benefit greatly from continued training and/or fine-tuning.
|
| 2 |
|
| 3 |
Newlines are replaced by `<N>`
|
|
@@ -31,3 +41,6 @@ This should give you something like:
|
|
| 31 |
`def do_something():<N> print("Hello")<N>`
|
| 32 |
|
| 33 |
...which is what the model is expecting as input.
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language: "Python"
|
| 3 |
+
tags:
|
| 4 |
+
- Code
|
| 5 |
+
- GPyT
|
| 6 |
+
- code generator
|
| 7 |
+
license: "MIT"
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
|
| 11 |
GPyT is a GPT2 model trained from scratch (not fine tuned) on Python code from Github. Overall, it was ~80GB of pure Python code, the current GPyT model is a mere 2 epochs through this data, so it may benefit greatly from continued training and/or fine-tuning.
|
| 12 |
|
| 13 |
Newlines are replaced by `<N>`
|
|
|
|
| 41 |
`def do_something():<N> print("Hello")<N>`
|
| 42 |
|
| 43 |
...which is what the model is expecting as input.
|
| 44 |
+
|
| 45 |
+
|
| 46 |
+
|