updated the eval script with the result
Browse files
eval.py
CHANGED
@@ -50,9 +50,12 @@ def log_results(result: Dataset, args: Dict[str, str]):
|
|
50 |
def normalize_text(text: str) -> str:
|
51 |
"""DO ADAPT FOR YOUR USE CASE. this function normalizes the target text."""
|
52 |
|
53 |
-
chars_to_ignore_regex = '[
|
54 |
|
55 |
text = re.sub(chars_to_ignore_regex, "", text.lower())
|
|
|
|
|
|
|
56 |
|
57 |
# In addition, we can normalize the target text, e.g. removing new lines characters etc...
|
58 |
# note that order is important here!
|
|
|
50 |
def normalize_text(text: str) -> str:
|
51 |
"""DO ADAPT FOR YOUR USE CASE. this function normalizes the target text."""
|
52 |
|
53 |
+
chars_to_ignore_regex = '[,?.!-;:""%\'"�\'‘’_,!łńō–—\\\\\\“”\\[\\]]'
|
54 |
|
55 |
text = re.sub(chars_to_ignore_regex, "", text.lower())
|
56 |
+
text = re.sub(r'[‘’´`]', r"'", text)
|
57 |
+
text = re.sub(r'è', r"é", text)
|
58 |
+
text = re.sub(r"(-|' | '| +)", " ", text)
|
59 |
|
60 |
# In addition, we can normalize the target text, e.g. removing new lines characters etc...
|
61 |
# note that order is important here!
|
log_mozilla-foundation_common_voice_7_0_id_test_predictions.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
log_mozilla-foundation_common_voice_7_0_id_test_targets.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
mozilla-foundation_common_voice_7_0_id_test_eval_results.txt
ADDED
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
1 |
+
WER: 0.04492122024807241
|
2 |
+
CER: 0.015773881015293457
|