Spaces:
Sleeping
Sleeping
Made it work with gpu and multi-references and some optimizations
Browse files- requirements.txt +1 -0
- semf1.py +173 -25
requirements.txt
CHANGED
@@ -1,3 +1,4 @@
|
|
1 |
git+https://github.com/huggingface/evaluate@main
|
|
|
2 |
scikit-learn
|
3 |
sentence-transformers
|
|
|
1 |
git+https://github.com/huggingface/evaluate@main
|
2 |
+
nltk
|
3 |
scikit-learn
|
4 |
sentence-transformers
|
semf1.py
CHANGED
@@ -11,20 +11,21 @@
|
|
11 |
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
12 |
# See the License for the specific language governing permissions and
|
13 |
# limitations under the License.
|
14 |
-
# TODO: Add test cases
|
15 |
"""SEM-F1 metric"""
|
16 |
|
17 |
-
|
18 |
import abc
|
19 |
import sys
|
20 |
-
from typing import List, Optional, Tuple
|
21 |
|
22 |
import datasets
|
23 |
import evaluate
|
|
|
24 |
import numpy as np
|
25 |
from numpy.typing import NDArray
|
26 |
from sentence_transformers import SentenceTransformer
|
27 |
from sklearn.metrics.pairwise import cosine_similarity
|
|
|
28 |
|
29 |
_CITATION = """\
|
30 |
@inproceedings{bansal-etal-2022-sem,
|
@@ -103,18 +104,23 @@ class USE(Encoder):
|
|
103 |
|
104 |
|
105 |
class SBertEncoder(Encoder):
|
106 |
-
def __init__(self, model_name: str):
|
107 |
self.model = SentenceTransformer(model_name)
|
|
|
|
|
108 |
|
109 |
def encode(self, prediction: List[str]) -> NDArray:
|
110 |
-
|
|
|
|
|
111 |
|
112 |
|
113 |
-
def _get_encoder(model_name: str):
|
114 |
if model_name == "use":
|
115 |
-
return
|
|
|
116 |
else:
|
117 |
-
return SBertEncoder(model_name)
|
118 |
|
119 |
|
120 |
def _compute_f1(p, r, eps=sys.float_info.epsilon):
|
@@ -140,7 +146,7 @@ class SemF1(evaluate.Metric):
|
|
140 |
_MODEL_TYPE_TO_NAME = {
|
141 |
"pv1": "paraphrase-distilroberta-base-v1",
|
142 |
"stsb": "stsb-roberta-large",
|
143 |
-
"use": "use",
|
144 |
}
|
145 |
|
146 |
def _info(self):
|
@@ -151,19 +157,56 @@ class SemF1(evaluate.Metric):
|
|
151 |
citation=_CITATION,
|
152 |
inputs_description=_KWARGS_DESCRIPTION,
|
153 |
# This defines the format of each prediction and reference
|
154 |
-
features=
|
155 |
-
|
156 |
-
|
157 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
158 |
# # Homepage of the module for documentation
|
159 |
# Additional links to the codebase or references
|
160 |
reference_urls=["https://aclanthology.org/2022.emnlp-main.49/"]
|
161 |
)
|
162 |
|
163 |
def _get_model_name(self, model_type: Optional[str] = None) -> str:
|
164 |
-
# TODO: make it work with USE as well
|
165 |
if model_type is None:
|
166 |
-
model_type = "pv1" # Change it to use
|
167 |
|
168 |
if model_type not in self._MODEL_TYPE_TO_NAME.keys():
|
169 |
raise ValueError(f"Provide a correct model_type.\n"
|
@@ -172,21 +215,126 @@ class SemF1(evaluate.Metric):
|
|
172 |
|
173 |
return self._MODEL_TYPE_TO_NAME[model_type]
|
174 |
|
175 |
-
def
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
176 |
model_name = self._get_model_name(model_type)
|
177 |
-
encoder = _get_encoder(model_name)
|
178 |
|
|
|
179 |
precisions = [0] * len(predictions)
|
180 |
recalls = [0] * len(predictions)
|
181 |
f1_scores = [0] * len(predictions)
|
182 |
|
183 |
-
|
184 |
-
|
185 |
-
|
186 |
-
|
187 |
-
|
188 |
-
|
189 |
-
|
190 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
191 |
|
192 |
return {"precision": precisions, "recall": recalls, "f1": f1_scores}
|
|
|
11 |
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
12 |
# See the License for the specific language governing permissions and
|
13 |
# limitations under the License.
|
14 |
+
# TODO: Add test cases, Provide an option to pass batch size when computing the embeddings
|
15 |
"""SEM-F1 metric"""
|
16 |
|
|
|
17 |
import abc
|
18 |
import sys
|
19 |
+
from typing import List, Optional, Tuple, Union
|
20 |
|
21 |
import datasets
|
22 |
import evaluate
|
23 |
+
import nltk
|
24 |
import numpy as np
|
25 |
from numpy.typing import NDArray
|
26 |
from sentence_transformers import SentenceTransformer
|
27 |
from sklearn.metrics.pairwise import cosine_similarity
|
28 |
+
import torch
|
29 |
|
30 |
_CITATION = """\
|
31 |
@inproceedings{bansal-etal-2022-sem,
|
|
|
104 |
|
105 |
|
106 |
class SBertEncoder(Encoder):
|
107 |
+
def __init__(self, model_name: str, device: Union[str, int], batch_size: int):
|
108 |
self.model = SentenceTransformer(model_name)
|
109 |
+
self.device = device
|
110 |
+
self.batch_size = batch_size
|
111 |
|
112 |
def encode(self, prediction: List[str]) -> NDArray:
|
113 |
+
"""Returns sentence embeddings of dim: Batch x Dim"""
|
114 |
+
# SBert output is always Batch x Dim
|
115 |
+
return self.model.encode(prediction, device=self.device, batch_size=self.batch_size)
|
116 |
|
117 |
|
118 |
+
def _get_encoder(model_name: str, device: Union[str, int], batch_size: int) -> Encoder:
|
119 |
if model_name == "use":
|
120 |
+
return SBertEncoder(model_name, device)
|
121 |
+
# return USE() # TODO: This will change depending on PyTorch USE VS TF USE model
|
122 |
else:
|
123 |
+
return SBertEncoder(model_name, device, batch_size)
|
124 |
|
125 |
|
126 |
def _compute_f1(p, r, eps=sys.float_info.epsilon):
|
|
|
146 |
_MODEL_TYPE_TO_NAME = {
|
147 |
"pv1": "paraphrase-distilroberta-base-v1",
|
148 |
"stsb": "stsb-roberta-large",
|
149 |
+
"use": "sentence-transformers/use-cmlm-multilingual", # TODO: check PyTorch USE VS TF USE
|
150 |
}
|
151 |
|
152 |
def _info(self):
|
|
|
157 |
citation=_CITATION,
|
158 |
inputs_description=_KWARGS_DESCRIPTION,
|
159 |
# This defines the format of each prediction and reference
|
160 |
+
features=[
|
161 |
+
# Multi References: False, Tokenize_Sentences = False
|
162 |
+
datasets.Features(
|
163 |
+
{
|
164 |
+
# predictions: List[List[str]] - List of predictions where prediction is a list of sentences
|
165 |
+
"predictions": datasets.Sequence(datasets.Value("string", id="sequence"), id="predictions"),
|
166 |
+
# references: List[List[str]] - List of references where each reference is a list of sentences
|
167 |
+
"references": datasets.Sequence(datasets.Value("string", id="sequence"), id="references"),
|
168 |
+
}
|
169 |
+
),
|
170 |
+
# Multi References: False, Tokenize_Sentences = True
|
171 |
+
datasets.Features(
|
172 |
+
{
|
173 |
+
# predictions: List[str] - List of predictions
|
174 |
+
"predictions": datasets.Value("string", id="sequence"),
|
175 |
+
# references: List[str] - List of documents
|
176 |
+
"references": datasets.Value("string", id="sequence"),
|
177 |
+
}
|
178 |
+
),
|
179 |
+
# Multi References: True, Tokenize_Sentences = False
|
180 |
+
datasets.Features(
|
181 |
+
{
|
182 |
+
# predictions: List[List[str]] - List of predictions where prediction is a list of sentences
|
183 |
+
"predictions": datasets.Sequence(datasets.Value("string", id="sequence"), id="predictions"),
|
184 |
+
# references: List[List[List[str]]] - List of multi-references.
|
185 |
+
# So each "reference" is also a list (r1, r2, ...).
|
186 |
+
# Further, each ri's are also list of sentences.
|
187 |
+
"references": datasets.Sequence(
|
188 |
+
datasets.Sequence(datasets.Value("string", id="sequence"), id="ref"), id="references"),
|
189 |
+
}
|
190 |
+
),
|
191 |
+
# Multi References: True, Tokenize_Sentences = True
|
192 |
+
datasets.Features(
|
193 |
+
{
|
194 |
+
# predictions: List[str] - List of predictions
|
195 |
+
"predictions": datasets.Value("string", id="sequence"),
|
196 |
+
# references: List[List[List[str]]] - List of multi-references.
|
197 |
+
# So each "reference" is also a list (r1, r2, ...).
|
198 |
+
"references": datasets.Sequence(datasets.Value("string", id="ref"), id="references"),
|
199 |
+
}
|
200 |
+
),
|
201 |
+
],
|
202 |
# # Homepage of the module for documentation
|
203 |
# Additional links to the codebase or references
|
204 |
reference_urls=["https://aclanthology.org/2022.emnlp-main.49/"]
|
205 |
)
|
206 |
|
207 |
def _get_model_name(self, model_type: Optional[str] = None) -> str:
|
|
|
208 |
if model_type is None:
|
209 |
+
model_type = "pv1" # TODO: Change it to use
|
210 |
|
211 |
if model_type not in self._MODEL_TYPE_TO_NAME.keys():
|
212 |
raise ValueError(f"Provide a correct model_type.\n"
|
|
|
215 |
|
216 |
return self._MODEL_TYPE_TO_NAME[model_type]
|
217 |
|
218 |
+
def _download_and_prepare(self, dl_manager):
|
219 |
+
"""Optional: download external resources useful to compute the scores"""
|
220 |
+
import nltk
|
221 |
+
if not nltk.data.find("tokenizers/punkt"):
|
222 |
+
nltk.download("punkt", quiet=True)
|
223 |
+
|
224 |
+
def _compute(
|
225 |
+
self,
|
226 |
+
predictions,
|
227 |
+
references,
|
228 |
+
model_type: Optional[str] = None,
|
229 |
+
tokenize_sentences: bool = True,
|
230 |
+
gpu: Union[bool, int] = False,
|
231 |
+
batch_size: int = 32,
|
232 |
+
):
|
233 |
+
|
234 |
+
# Ensure gpu index is within the range of total available gpus
|
235 |
+
gpu_available = True if torch.cuda.is_available() else False
|
236 |
+
if gpu_available:
|
237 |
+
gpu_count = torch.cuda.device_count()
|
238 |
+
if isinstance(gpu, int) and gpu >= gpu_count:
|
239 |
+
raise ValueError(
|
240 |
+
f"There are {gpu_count} gpus available. Provide the correct gpu index. You provided: {gpu}"
|
241 |
+
)
|
242 |
+
|
243 |
+
# get the device
|
244 |
+
if gpu is False:
|
245 |
+
device = "cpu"
|
246 |
+
elif gpu is True and torch.cuda.is_available():
|
247 |
+
device = 0 # by default run on device 0
|
248 |
+
elif isinstance(gpu, int):
|
249 |
+
device = gpu
|
250 |
+
else: # This will never happen
|
251 |
+
raise ValueError(f"gpu must be bool or int. Provided value: {gpu}")
|
252 |
+
|
253 |
+
# TODO: Also have a check on references to ensure they are also in correct format
|
254 |
+
# Ensure prediction documents are not already tokenized if tokenize_sentences is True
|
255 |
+
if not isinstance(predictions[0], str) and tokenize_sentences:
|
256 |
+
raise ValueError(f"Each prediction/reference should be a document i.e. when tokenize_sentences is True. "
|
257 |
+
f"Currently, each prediction is of type {type(predictions[0])} ")
|
258 |
+
|
259 |
+
# Check single reference or multi-reference case
|
260 |
+
multi_references = False
|
261 |
+
if tokenize_sentences:
|
262 |
+
# references: List[List[reference]]
|
263 |
+
if isinstance(references[0], list) and isinstance(references[0][0], str):
|
264 |
+
multi_references = True
|
265 |
+
else:
|
266 |
+
# references: List[List[List[sentence]]]
|
267 |
+
if (
|
268 |
+
isinstance(references[0], list) and
|
269 |
+
isinstance(references[0][0], list) and
|
270 |
+
isinstance(references[0][0][0], str)
|
271 |
+
):
|
272 |
+
multi_references = True
|
273 |
+
|
274 |
+
# Get the encoder model
|
275 |
model_name = self._get_model_name(model_type)
|
276 |
+
encoder = _get_encoder(model_name, device=device)
|
277 |
|
278 |
+
# Init output scores
|
279 |
precisions = [0] * len(predictions)
|
280 |
recalls = [0] * len(predictions)
|
281 |
f1_scores = [0] * len(predictions)
|
282 |
|
283 |
+
# Compute Score in case of single reference
|
284 |
+
if not multi_references:
|
285 |
+
for idx, (pred, ref) in enumerate(zip(predictions, references)):
|
286 |
+
|
287 |
+
# Sentence Tokenize prediction and reference
|
288 |
+
if tokenize_sentences:
|
289 |
+
ref = nltk.tokenize.sent_tokenize(ref) # List[str]
|
290 |
+
pred = nltk.tokenize.sent_tokenize(pred) # List[str]
|
291 |
+
|
292 |
+
pred_sent_count = len(pred)
|
293 |
+
embeddings = encoder.encode(pred + ref)
|
294 |
+
pred_embeddings = embeddings[:pred_sent_count]
|
295 |
+
ref_embeddings = embeddings[pred_sent_count:]
|
296 |
+
|
297 |
+
p, r = _compute_cosine_similarity(pred_embeddings, ref_embeddings)
|
298 |
+
f1 = _compute_f1(p, r)
|
299 |
+
precisions[idx] = p
|
300 |
+
recalls[idx] = r
|
301 |
+
f1_scores[idx] = f1
|
302 |
+
|
303 |
+
else:
|
304 |
+
# Compute Score in case of multiple reference
|
305 |
+
for idx, (pred, refs) in enumerate(zip(predictions, references)):
|
306 |
+
# Sentence Tokenize prediction and reference
|
307 |
+
if tokenize_sentences:
|
308 |
+
refs = [nltk.tokenize.sent_tokenize(ref) for ref in refs] # List[List[str]]
|
309 |
+
pred = nltk.tokenize.sent_tokenize(pred) # List[str]
|
310 |
+
|
311 |
+
ref_count = len(refs)
|
312 |
+
pred_sent_count = len(pred)
|
313 |
+
ref_sent_counts = [0] + [len(ref) for ref in refs]
|
314 |
+
cumsum_ref_sent_counts = np.cumsum(ref_sent_counts)
|
315 |
+
|
316 |
+
all_sentences = pred + sum(refs, [])
|
317 |
+
embeddings = encoder.encode(all_sentences)
|
318 |
+
pred_embeddings = embeddings[:pred_sent_count]
|
319 |
+
ref_embeddings = [
|
320 |
+
embeddings[pred_sent_count + cumsum_ref_sent_counts[c_idx]:
|
321 |
+
pred_sent_count + cumsum_ref_sent_counts[c_idx + 1]]
|
322 |
+
for c_idx in range(ref_count)
|
323 |
+
]
|
324 |
+
# pred_embeddings = encoder.encode(pred)
|
325 |
+
# ref_embeddings = [encoder.encode(refs) for ref in refs]
|
326 |
+
|
327 |
+
# Precision: Concatenate all the sentences in all the references
|
328 |
+
concat_ref_embeddings = np.concatenate(ref_embeddings, axis=0)
|
329 |
+
p, _ = _compute_cosine_similarity(pred_embeddings, concat_ref_embeddings)
|
330 |
+
|
331 |
+
# Recall: Compute individually for each reference
|
332 |
+
scores = [_compute_cosine_similarity(r_embeds, pred_embeddings) for r_embeds in ref_embeddings]
|
333 |
+
r = np.mean([r_scores for (r_scores, _) in scores]).item()
|
334 |
+
|
335 |
+
f1 = _compute_f1(p, r)
|
336 |
+
precisions[idx] = p # TODO: check why idx says invalid type
|
337 |
+
recalls[idx] = r
|
338 |
+
f1_scores[idx] = f1
|
339 |
|
340 |
return {"precision": precisions, "recall": recalls, "f1": f1_scores}
|