Add basic BLEU metric class (#hacktoberfest)

src/evidently/metrics/bleu_metric.py

@@ -0,0 +1,41 @@
+# src/evidently/metrics/bleu_metric.py
+"""
+BLEU Metric for evaluating candidate sentences against reference sentences.
+Beginner-friendly contribution for Hacktoberfest.
+"""
+
+from nltk.translate.bleu_score import sentence_bleu
+
+class BleuMetric:
+    def __init__(self):
+        """Initialize the BLEU metric class."""
+        pass
+
+    def evaluate(self, reference, candidate):
+        """
+        Evaluate the BLEU score.
+
+        Parameters:
+        reference (str): The reference sentence.
+        candidate (str): The candidate sentence to evaluate.
+
+        Returns:
+        float: BLEU score between 0 and 1.
+        """
+        # Convert sentences to lists of words
+        reference_words = reference.split()
+        candidate_words = candidate.split()
+
+        # Compute BLEU score
+        score = sentence_bleu([reference_words], candidate_words)
+
+        return score
+
+
+# Example usage (for maintainers/tests)
+if __name__ == "__main__":
+    reference_sentence = "The cat is on the mat"
+    candidate_sentence = "The cat is on the mat"
+
+    bleu = BleuMetric()
+    print("BLEU Score:", bleu.evaluate(reference_sentence, candidate_sentence))