Update README.md
Browse files
README.md
CHANGED
|
@@ -14,6 +14,8 @@ In contrast to the ChEBI25-3STAR dataset, not all positive and negative samples
|
|
| 14 |
- For the training split, positive samples are chosen randomly up to a maximum of 200. Negative samples are chosen by distance to the true label via breadth first search. The search iterates until the maximum of 200 samples is reached or the minimum of 25 samples is reached and all samples at the current distance level are included.
|
| 15 |
- For the validation and test split, all positive samples and negative samples that are siblings to positive samples (i.e. that belong to a direct superclass of the true label) are included.
|
| 16 |
|
|
|
|
|
|
|
| 17 |
## Problem Structure
|
| 18 |
Each problem is structured as follows:
|
| 19 |
|
|
|
|
| 14 |
- For the training split, positive samples are chosen randomly up to a maximum of 200. Negative samples are chosen by distance to the true label via breadth first search. The search iterates until the maximum of 200 samples is reached or the minimum of 25 samples is reached and all samples at the current distance level are included.
|
| 15 |
- For the validation and test split, all positive samples and negative samples that are siblings to positive samples (i.e. that belong to a direct superclass of the true label) are included.
|
| 16 |
|
| 17 |
+
The code used to generate these problems can be found on [GitHub](https://github.com/ChEB-AI/chebILP).
|
| 18 |
+
|
| 19 |
## Problem Structure
|
| 20 |
Each problem is structured as follows:
|
| 21 |
|