Spaces:
Runtime error
Runtime error
davebulaval
commited on
Commit
•
495f38b
1
Parent(s):
ba092fc
remove device handling
Browse files- meaningbert.py +3 -5
meaningbert.py
CHANGED
@@ -67,7 +67,6 @@ MeaningBERT metric for assessing meaning preservation between sentences.
|
|
67 |
Args:
|
68 |
predictions (list of str): Predictions sentences.
|
69 |
references (list of str): References sentences (same number of element as predictions).
|
70 |
-
device (str): Device to use for model inference. By default, set to "cuda".
|
71 |
|
72 |
Returns:
|
73 |
score: the meaning score between two sentences in alist format respecting the order of the predictions and
|
@@ -78,7 +77,7 @@ Examples:
|
|
78 |
|
79 |
>>> references = ["hello there", "general kenobi"]
|
80 |
>>> predictions = ["hello there", "general kenobi"]
|
81 |
-
>>> meaning_bert = evaluate.load("davebulaval/meaningbert"
|
82 |
>>> results = meaning_bert.compute(predictions=predictions, references=references)
|
83 |
"""
|
84 |
|
@@ -113,7 +112,6 @@ class MeaningBERT(evaluate.Metric):
|
|
113 |
self,
|
114 |
predictions: List,
|
115 |
references: List,
|
116 |
-
device: str = "cuda",
|
117 |
) -> Dict:
|
118 |
assert len(references) == len(
|
119 |
predictions
|
@@ -125,7 +123,7 @@ class MeaningBERT(evaluate.Metric):
|
|
125 |
|
126 |
# We load the MeaningBERT pretrained model
|
127 |
scorer = AutoModelForSequenceClassification.from_pretrained(
|
128 |
-
"davebulaval/MeaningBERT"
|
129 |
)
|
130 |
scorer.eval()
|
131 |
|
@@ -140,7 +138,7 @@ class MeaningBERT(evaluate.Metric):
|
|
140 |
truncation=True,
|
141 |
padding=True,
|
142 |
return_tensors="pt",
|
143 |
-
)
|
144 |
|
145 |
with filter_logging_context():
|
146 |
# We process the text
|
|
|
67 |
Args:
|
68 |
predictions (list of str): Predictions sentences.
|
69 |
references (list of str): References sentences (same number of element as predictions).
|
|
|
70 |
|
71 |
Returns:
|
72 |
score: the meaning score between two sentences in alist format respecting the order of the predictions and
|
|
|
77 |
|
78 |
>>> references = ["hello there", "general kenobi"]
|
79 |
>>> predictions = ["hello there", "general kenobi"]
|
80 |
+
>>> meaning_bert = evaluate.load("davebulaval/meaningbert")
|
81 |
>>> results = meaning_bert.compute(predictions=predictions, references=references)
|
82 |
"""
|
83 |
|
|
|
112 |
self,
|
113 |
predictions: List,
|
114 |
references: List,
|
|
|
115 |
) -> Dict:
|
116 |
assert len(references) == len(
|
117 |
predictions
|
|
|
123 |
|
124 |
# We load the MeaningBERT pretrained model
|
125 |
scorer = AutoModelForSequenceClassification.from_pretrained(
|
126 |
+
"davebulaval/MeaningBERT"
|
127 |
)
|
128 |
scorer.eval()
|
129 |
|
|
|
138 |
truncation=True,
|
139 |
padding=True,
|
140 |
return_tensors="pt",
|
141 |
+
)
|
142 |
|
143 |
with filter_logging_context():
|
144 |
# We process the text
|