Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,13 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
| 4 |
+
|
| 5 |
+
Contains files for a Transformer model that answers 5,6,etc-digit addition and/or subtractions questions (e.g. 123450-345670=-0123230).
|
| 6 |
+
|
| 7 |
+
The model can have 1, 2 or 3 layers, 3 or 4 attention heads, d-model = 510, d-head = 170.
|
| 8 |
+
The model can be trained to do addition, subtraction or both (aka "mixed").
|
| 9 |
+
An untrained mixed model can be initialised with a previously trained addition model.
|
| 10 |
+
|
| 11 |
+
The file are generated by two CoLabs:
|
| 12 |
+
- One is used to train the model outputing a "pt" file: https://github.com/apartresearch/Verified_addition/blob/main/assets/Accurate_Math_Train.ipynb
|
| 13 |
+
- One is used to analyse the model outputing a "json" file: https://github.com/apartresearch/Verified_addition/blob/main/assets/Accurate_Math_Analyse.ipynb
|