Added train-test split

akalai · akalai · commit cd6821aa4f9e · 2022-02-08T17:01:57.000-05:00
diff --git a/.gitignore b/.gitignore
@@ -1,3 +1,5 @@
+.vscode/
+
 # Byte-compiled / optimized / DLL files
 __pycache__/
 *.py[cod]
diff --git a/README.md b/README.md
@@ -241,6 +241,11 @@ You can try out the notebook at (link to be added).
     * Conway's 99-graph problem (**open**)
     * Finding a cycle in the Collatz process (**open**)
 
+## Train-test split
+The file [`split.json`](/notebooks/split.json) contains a suggested train-test split. This split was hand-selected
+by the puzzle authors, who are familiar with all puzzles, so that: there is minimal overlap between related puzzles 
+in the two splits. In particular, for pairs of related puzzles, either both were placed in the training set or the test
+set.
 
 ## Contributing
 
diff --git a/puzzles/split.json b/puzzles/split.json

Original file line number	Diff line number	Diff line change
`@@ -1,3 +1,5 @@`
	`1`	`+.vscode/`
	`2`	`+`
`1`	`3`	`# Byte-compiled / optimized / DLL files`
`2`	`4`	`__pycache__/`
`3`	`5`	`*.py[cod]`