Some tests to better understand compute score

Testing.py

@@ -0,0 +1,18 @@
+import evaluate as ev
+from nDCG import nDCG
+
+metric = nDCG(cache_dir="cache")
+a = [1,2,3,4,5]
+b = [1,2,3,4,5]
+c = [1,2,3,4,0]
+
+
+#metric.add(prediction=a, reference=b)
+metric.add(prediction=c, reference=b)
+metric.add(prediction=c, reference=b)
+metric.add(prediction=c, reference=b)
+print(metric.compute(predictions=[a], references=[b]))
+print(metric.compute(predictions=[a], references=[c]))
+print(metric.compute(predictions=[a], references=[c]))
+print(metric.compute(predictions=[a,a], references=[c,a]))
+print(metric.cache_file_name)

requirements.txt

@@ -1,2 +1,3 @@
-git+https://github.com/huggingface/evaluate@dd35e04844c71068318335e05d8a658200dafe89
-scikit-learn
+scikit-learn
+datasets
+evaluate