add WER calculation tutorial

Author: x4nth055

README.md

@@ -60,6 +60,7 @@ This is a repository of all the tutorials of [The Python Code](https://www.thepy
         - [Tokenization, Stemming, and Lemmatization in Python](https://www.thepythoncode.com/article/tokenization-stemming-and-lemmatization-in-python). ([code](machine-learning/nlp/tokenization-stemming-lemmatization))
         - [How to Fine Tune BERT for Semantic Textual Similarity using Transformers in Python](https://www.thepythoncode.com/article/finetune-bert-for-semantic-textual-similarity-in-python). ([code](machine-learning/nlp/semantic-textual-similarity))
         - [How to Calculate the BLEU Score in Python](https://www.thepythoncode.com/article/bleu-score-in-python). ([code](machine-learning/nlp/bleu-score))
+        - [Word Error Rate in Python](https://www.thepythoncode.com/article/calculate-word-error-rate-in-python). ([code](machine-learning/nlp/wer-score))
     - ### [Computer Vision](https://www.thepythoncode.com/topic/computer-vision)
         - [How to Detect Human Faces in Python using OpenCV](https://www.thepythoncode.com/article/detect-faces-opencv-python). ([code](machine-learning/face_detection))
         - [How to Make an Image Classifier in Python using TensorFlow and Keras](https://www.thepythoncode.com/article/image-classification-keras-python). ([code](machine-learning/image-classifier))

machine-learning/nlp/wer-score/README.md

@@ -0,0 +1,6 @@
+# [Word Error Rate in Python](https://www.thepythoncode.com/article/calculate-word-error-rate-in-python)
+- `pip install -r requirements.txt`
+- `wer_basic.py` is the basic implementation of WER algorithm.
+- `wer_accurate.py` is the accurate implementation of WER algorithm.
+- `wer_jiwer.py` is the implementation of WER algorithm using [jiwer](https://pypi.org/project/jiwer/).
+- `wer_evaluate.py` is the implementation of WER algorithm using [evaluate](https://pypi.org/project/evaluate/).

machine-learning/nlp/wer-score/requirements.txt

@@ -0,0 +1,3 @@
+numpy
+jiwer
+evaluate

machine-learning/nlp/wer-score/wer_accurate.py

@@ -0,0 +1,44 @@
+import numpy as np
+
+def calculate_wer(reference, hypothesis):
+    # Split the reference and hypothesis sentences into words
+    ref_words = reference.split()
+    hyp_words = hypothesis.split()
+    # Initialize a matrix with size |ref_words|+1 x |hyp_words|+1
+    # The extra row and column are for the case when one of the strings is empty
+    d = np.zeros((len(ref_words) + 1, len(hyp_words) + 1))
+    # The number of operations for an empty hypothesis to become the reference
+    # is just the number of words in the reference (i.e., deleting all words)
+    for i in range(len(ref_words) + 1):
+        d[i, 0] = i
+    # The number of operations for an empty reference to become the hypothesis
+    # is just the number of words in the hypothesis (i.e., inserting all words)
+    for j in range(len(hyp_words) + 1):
+        d[0, j] = j
+    # Iterate over the words in the reference and hypothesis
+    for i in range(1, len(ref_words) + 1):
+        for j in range(1, len(hyp_words) + 1):
+            # If the current words are the same, no operation is needed
+            # So we just take the previous minimum number of operations
+            if ref_words[i - 1] == hyp_words[j - 1]:
+                d[i, j] = d[i - 1, j - 1]
+            else:
+                # If the words are different, we consider three operations:
+                # substitution, insertion, and deletion
+                # And we take the minimum of these three possibilities
+                substitution = d[i - 1, j - 1] + 1
+                insertion = d[i, j - 1] + 1
+                deletion = d[i - 1, j] + 1
+                d[i, j] = min(substitution, insertion, deletion)
+    # The minimum number of operations to transform the hypothesis into the reference
+    # is in the bottom-right cell of the matrix
+    # We divide this by the number of words in the reference to get the WER
+    wer = d[len(ref_words), len(hyp_words)] / len(ref_words)
+    return wer
+
+
+
+if __name__ == "__main__":
+    reference = "The cat is sleeping on the mat."
+    hypothesis = "The cat is playing on mat."
+    print(calculate_wer(reference, hypothesis))

machine-learning/nlp/wer-score/wer_basic.py

@@ -0,0 +1,21 @@
+def calculate_wer(reference, hypothesis):
+	ref_words = reference.split()
+	hyp_words = hypothesis.split()
+ 
+	# Counting the number of substitutions, deletions, and insertions
+	substitutions = sum(1 for ref, hyp in zip(ref_words, hyp_words) if ref != hyp)
+	deletions = len(ref_words) - len(hyp_words)
+	insertions = len(hyp_words) - len(ref_words)
+ 
+	# Total number of words in the reference text
+	total_words = len(ref_words)
+ 
+	# Calculating the Word Error Rate (WER)
+	wer = (substitutions + deletions + insertions) / total_words
+	return wer
+
+
+if __name__ == "__main__":
+    reference = "the cat sat on the mat"
+    hypothesis = "the cat mat"
+    print(calculate_wer(reference, hypothesis))

machine-learning/nlp/wer-score/wer_evaluate.py

@@ -0,0 +1,9 @@
+import evaluate
+
+wer = evaluate.load("wer")
+
+# reference = "the cat sat on the mat"
+# hypothesis = "the cat mat"
+reference = "The cat is sleeping on the mat."
+hypothesis = "The cat is playing on mat."
+print(wer.compute(references=[reference], predictions=[hypothesis]))

machine-learning/nlp/wer-score/wer_jiwer.py

@@ -0,0 +1,8 @@
+from jiwer import wer
+
+if __name__ == "__main__":
+    # reference = "the cat sat on the mat"
+    # hypothesis = "the cat mat"
+    reference = "The cat is sleeping on the mat."
+    hypothesis = "The cat is playing on mat."
+    print(wer(reference, hypothesis))