Spaces:
Running
on
CPU Upgrade
Running
on
CPU Upgrade
Add new external models (used for German clustering)
#9
by
slvnwhrl
- opened
app.py
CHANGED
@@ -151,6 +151,12 @@ EXTERNAL_MODELS = [
|
|
151 |
"allenai-specter",
|
152 |
"bert-base-uncased",
|
153 |
"contriever-base-msmarco",
|
|
|
|
|
|
|
|
|
|
|
|
|
154 |
"glove.6B.300d",
|
155 |
"gtr-t5-base",
|
156 |
"gtr-t5-large",
|
@@ -175,6 +181,8 @@ EXTERNAL_MODELS = [
|
|
175 |
"text-search-curie-001",
|
176 |
"text-search-davinci-001",
|
177 |
"unsup-simcse-bert-base-uncased",
|
|
|
|
|
178 |
]
|
179 |
EXTERNAL_MODEL_TO_LINK = {
|
180 |
"LASER2": "https://github.com/facebookresearch/LASER",
|
@@ -215,6 +223,14 @@ EXTERNAL_MODEL_TO_LINK = {
|
|
215 |
"all-mpnet-base-v2": "https://huggingface.co/sentence-transformers/all-mpnet-base-v2",
|
216 |
"paraphrase-multilingual-mpnet-base-v2": "https://huggingface.co/sentence-transformers/paraphrase-multilingual-mpnet-base-v2",
|
217 |
"paraphrase-multilingual-MiniLM-L12-v2": "https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
218 |
}
|
219 |
|
220 |
EXTERNAL_MODEL_TO_DIM = {
|
@@ -226,6 +242,12 @@ EXTERNAL_MODEL_TO_DIM = {
|
|
226 |
"allenai-specter": 768,
|
227 |
"bert-base-uncased": 768,
|
228 |
"contriever-base-msmarco": 768,
|
|
|
|
|
|
|
|
|
|
|
|
|
229 |
"glove.6B.300d": 300,
|
230 |
"gtr-t5-base": 768,
|
231 |
"gtr-t5-large": 768,
|
@@ -256,6 +278,9 @@ EXTERNAL_MODEL_TO_DIM = {
|
|
256 |
"text-search-davinci-001": 12288,
|
257 |
|
258 |
"unsup-simcse-bert-base-uncased": 768,
|
|
|
|
|
|
|
259 |
}
|
260 |
|
261 |
MODELS_TO_SKIP = {
|
|
|
151 |
"allenai-specter",
|
152 |
"bert-base-uncased",
|
153 |
"contriever-base-msmarco",
|
154 |
+
"cross-en-de-roberta-sentence-transformer",
|
155 |
+
"gbert-base",
|
156 |
+
"gbert-large",
|
157 |
+
"gelectra-base",
|
158 |
+
"gelectra-large",
|
159 |
+
"gottbert-base",
|
160 |
"glove.6B.300d",
|
161 |
"gtr-t5-base",
|
162 |
"gtr-t5-large",
|
|
|
181 |
"text-search-curie-001",
|
182 |
"text-search-davinci-001",
|
183 |
"unsup-simcse-bert-base-uncased",
|
184 |
+
"use-cmlm-multilingual",
|
185 |
+
"xlm-roberta-large",
|
186 |
]
|
187 |
EXTERNAL_MODEL_TO_LINK = {
|
188 |
"LASER2": "https://github.com/facebookresearch/LASER",
|
|
|
223 |
"all-mpnet-base-v2": "https://huggingface.co/sentence-transformers/all-mpnet-base-v2",
|
224 |
"paraphrase-multilingual-mpnet-base-v2": "https://huggingface.co/sentence-transformers/paraphrase-multilingual-mpnet-base-v2",
|
225 |
"paraphrase-multilingual-MiniLM-L12-v2": "https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2",
|
226 |
+
"cross-en-de-roberta-sentence-transformer",
|
227 |
+
"gbert-base": "https://huggingface.co/deepset/gbert-base",
|
228 |
+
"gbert-large": "https://huggingface.co/deepset/gbert-large",
|
229 |
+
"gelectra-base": "https://huggingface.co/deepset/gelectra-base",
|
230 |
+
"gelectra-large": "https://huggingface.co/deepset/gelectra-large",
|
231 |
+
"gottbert-base": "https://huggingface.co/uklfr/gottbert-base",
|
232 |
+
"use-cmlm-multilingual": "https://huggingface.co/sentence-transformers/use-cmlm-multilingual",
|
233 |
+
"xlm-roberta-large": "https://huggingface.co/xlm-roberta-large",
|
234 |
}
|
235 |
|
236 |
EXTERNAL_MODEL_TO_DIM = {
|
|
|
242 |
"allenai-specter": 768,
|
243 |
"bert-base-uncased": 768,
|
244 |
"contriever-base-msmarco": 768,
|
245 |
+
"cross-en-de-roberta-sentence-transformer": 0,
|
246 |
+
"gbert-base": 768,
|
247 |
+
"gbert-large": 1024,
|
248 |
+
"gelectra-base": 768,
|
249 |
+
"gelectra-large": 1024,
|
250 |
+
"gottbert-base": 768,
|
251 |
"glove.6B.300d": 300,
|
252 |
"gtr-t5-base": 768,
|
253 |
"gtr-t5-large": 768,
|
|
|
278 |
"text-search-davinci-001": 12288,
|
279 |
|
280 |
"unsup-simcse-bert-base-uncased": 768,
|
281 |
+
|
282 |
+
"use-cmlm-multilingual": 768,
|
283 |
+
"xlm-roberta-large": 1024,
|
284 |
}
|
285 |
|
286 |
MODELS_TO_SKIP = {
|