Commit
•
69ef466
1
Parent(s):
a782098
Add SetFit model
Browse files- README.md +57 -187
- config.json +8 -8
- config_sentence_transformers.json +2 -2
- config_setfit.json +2 -2
- model.safetensors +1 -1
- model_head.pkl +2 -2
- tokenizer.json +2 -1
README.md
CHANGED
@@ -5,71 +5,43 @@ tags:
|
|
5 |
- sentence-transformers
|
6 |
- text-classification
|
7 |
- generated_from_setfit_trainer
|
|
|
8 |
metrics:
|
9 |
- accuracy
|
10 |
widget:
|
11 |
-
- text:
|
12 |
-
|
13 |
-
|
14 |
-
|
15 |
-
|
16 |
-
|
17 |
-
|
18 |
-
|
19 |
-
|
20 |
-
|
21 |
-
|
22 |
-
- text:
|
23 |
-
|
24 |
-
|
25 |
-
|
26 |
-
la
|
27 |
-
|
28 |
-
|
29 |
-
|
30 |
-
|
31 |
-
|
32 |
-
- text:
|
33 |
-
|
34 |
-
|
35 |
-
|
36 |
-
|
37 |
-
que seria bona idea fer un petit passeig a l'Avinguda Dr Fleming, la part
|
38 |
-
per sobre Passeig de Ronda. Actualment hi ha 4 carrils (2 centrals i 2
|
39 |
-
laterals) inicialment imagino que pel Camp d'Esports (tot i que no hi ha
|
40 |
-
massa aforament) si els centrals fossin peatonals s'aconseguiria conectar la
|
41 |
-
part de Ciutat Jardí i zona Ricard Vinyes amb un passeig agradable. Al
|
42 |
-
carrer hi ha 3/4 locals buits i potser es es motivaria a fer negocis (tipus
|
43 |
-
cafeteria o comerç de proximitat). A Lleida ens falten zones agradables on
|
44 |
-
caminar i hi hauria una bona oportunitat en aquesta avinguda. Moltes gràcies
|
45 |
-
per llegir-nos!!! Helena i Òscar
|
46 |
-
- text: >-
|
47 |
-
Ticket guardería escola bressol Municipal Valiet : Bon dia, Necessitaria
|
48 |
-
saber si la gestio de l’escola bresssol Vailet la porteu centralitzadament.
|
49 |
-
La meva empresa fa un pagament mensual de 73€ a l’escola en nom de la meva
|
50 |
-
filla que dita escola afirma no haver rebut. Gràcies
|
51 |
pipeline_tag: text-classification
|
52 |
inference: true
|
53 |
-
base_model: ibaucells/RoBERTa-ca-CaWikiTC
|
54 |
-
model-index:
|
55 |
-
- name: SetFit with ibaucells/RoBERTa-ca-CaWikiTC
|
56 |
-
results:
|
57 |
-
- task:
|
58 |
-
type: text-classification
|
59 |
-
name: Text Classification
|
60 |
-
dataset:
|
61 |
-
name: Unknown
|
62 |
-
type: unknown
|
63 |
-
split: test
|
64 |
-
metrics:
|
65 |
-
- type: accuracy
|
66 |
-
value: 0.4235294117647059
|
67 |
-
name: Accuracy
|
68 |
---
|
69 |
|
70 |
-
# SetFit with
|
71 |
|
72 |
-
This is a [SetFit](https://github.com/huggingface/setfit) model that can be used for Text Classification. This SetFit model uses [
|
73 |
|
74 |
The model has been trained using an efficient few-shot learning technique that involves:
|
75 |
|
@@ -80,7 +52,7 @@ The model has been trained using an efficient few-shot learning technique that i
|
|
80 |
|
81 |
### Model Description
|
82 |
- **Model Type:** SetFit
|
83 |
-
- **Sentence Transformer body:** [
|
84 |
- **Classification head:** a [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance
|
85 |
- **Maximum Sequence Length:** 512 tokens
|
86 |
- **Number of Classes:** 17 classes
|
@@ -95,32 +67,25 @@ The model has been trained using an efficient few-shot learning technique that i
|
|
95 |
- **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
|
96 |
|
97 |
### Model Labels
|
98 |
-
| Label | Examples
|
99 |
-
|
100 |
-
| 0 | <ul><li>'Aquest article tracta sobre Aigües'</li><li>'Aquest article tracta sobre Aigües'</li><li>'Aquest article tracta sobre Aigües'</li></ul>
|
101 |
-
| 1 | <ul><li>'Aquest article tracta sobre
|
102 |
-
| 2 | <ul><li>'Aquest article tracta sobre Cultura'</li><li>'Aquest article tracta sobre Cultura'</li><li>'Aquest article tracta sobre Cultura'</li></ul>
|
103 |
-
| 3 | <ul><li>'Aquest article tracta sobre Economia'</li><li>'Aquest article tracta sobre Economia'</li><li>'Aquest article tracta sobre Economia'</li></ul>
|
104 |
-
| 4 | <ul><li>'Aquest article tracta sobre Educació'</li><li>'Aquest article tracta sobre Educació'</li><li>'Aquest article tracta sobre Educació'</li></ul>
|
105 |
-
| 5 | <ul><li>'Aquest article tracta sobre Enllumenat'</li><li>'Aquest article tracta sobre Enllumenat'</li><li>'Aquest article tracta sobre Enllumenat'</li></ul>
|
106 |
-
| 6 | <ul><li>'Aquest article tracta sobre Esports'</li><li>'Aquest article tracta sobre Esports'</li><li>'Aquest article tracta sobre Esports'</li></ul>
|
107 |
-
| 7 | <ul><li>'Aquest article tracta sobre Habitatge'</li><li>'Aquest article tracta sobre Habitatge'</li><li>'Aquest article tracta sobre Habitatge'</li></ul>
|
108 |
-
| 8 | <ul><li>'Aquest article tracta sobre Horta'</li><li>'Aquest article tracta sobre Horta'</li><li>'Aquest article tracta sobre Horta'</li></ul>
|
109 |
-
| 9 | <ul><li>'Aquest article tracta sobre Medi ambient'</li><li>'Aquest article tracta sobre Medi ambient'</li><li>'Aquest article tracta sobre Medi ambient'</li></ul>
|
110 |
-
| 10 | <ul><li>'Aquest article tracta sobre
|
111 |
-
| 11 | <ul><li>'Aquest article tracta sobre
|
112 |
-
| 12 | <ul><li>'Aquest article tracta sobre
|
113 |
-
| 13 | <ul><li>'Aquest article tracta sobre
|
114 |
-
| 14 | <ul><li>'Aquest article tracta sobre
|
115 |
-
| 15 | <ul><li>'Aquest article tracta sobre
|
116 |
-
| 16 | <ul><li>'Aquest article tracta sobre
|
117 |
-
|
118 |
-
## Evaluation
|
119 |
-
|
120 |
-
### Metrics
|
121 |
-
| Label | Accuracy |
|
122 |
-
|:--------|:---------|
|
123 |
-
| **all** | 0.4235 |
|
124 |
|
125 |
## Uses
|
126 |
|
@@ -140,7 +105,7 @@ from setfit import SetFitModel
|
|
140 |
# Download from the 🤗 Hub
|
141 |
model = SetFitModel.from_pretrained("adriansanz/test8")
|
142 |
# Run inference
|
143 |
-
preds = model("
|
144 |
```
|
145 |
|
146 |
<!--
|
@@ -172,7 +137,7 @@ preds = model("Mateicula estiu de petits : Bona tarda, Un cop adjudicada la pla
|
|
172 |
### Training Set Metrics
|
173 |
| Training set | Min | Median | Max |
|
174 |
|:-------------|:----|:-------|:----|
|
175 |
-
| Word count | 5 | 5.
|
176 |
|
177 |
| Label | Training Sample Count |
|
178 |
|:------|:----------------------|
|
@@ -195,8 +160,8 @@ preds = model("Mateicula estiu de petits : Bona tarda, Un cop adjudicada la pla
|
|
195 |
| 16 | 8 |
|
196 |
|
197 |
### Training Hyperparameters
|
198 |
-
- batch_size: (
|
199 |
-
- num_epochs: (1,
|
200 |
- max_steps: -1
|
201 |
- sampling_strategy: oversampling
|
202 |
- body_learning_rate: (2e-05, 1e-05)
|
@@ -211,109 +176,14 @@ preds = model("Mateicula estiu de petits : Bona tarda, Un cop adjudicada la pla
|
|
211 |
- eval_max_steps: -1
|
212 |
- load_best_model_at_end: False
|
213 |
|
214 |
-
### Training Results
|
215 |
-
| Epoch | Step | Training Loss | Validation Loss |
|
216 |
-
|:-------:|:--------:|:-------------:|:---------------:|
|
217 |
-
| 0.0005 | 1 | 0.4258 | - |
|
218 |
-
| 0.0230 | 50 | 0.201 | - |
|
219 |
-
| 0.0460 | 100 | 0.194 | - |
|
220 |
-
| 0.0689 | 150 | 0.237 | - |
|
221 |
-
| 0.0919 | 200 | 0.1165 | - |
|
222 |
-
| 0.1149 | 250 | 0.0622 | - |
|
223 |
-
| 0.1379 | 300 | 0.0904 | - |
|
224 |
-
| 0.1608 | 350 | 0.0045 | - |
|
225 |
-
| 0.1838 | 400 | 0.0188 | - |
|
226 |
-
| 0.2068 | 450 | 0.0025 | - |
|
227 |
-
| 0.2298 | 500 | 0.0017 | - |
|
228 |
-
| 0.2528 | 550 | 0.0014 | - |
|
229 |
-
| 0.2757 | 600 | 0.0013 | - |
|
230 |
-
| 0.2987 | 650 | 0.0014 | - |
|
231 |
-
| 0.3217 | 700 | 0.0027 | - |
|
232 |
-
| 0.3447 | 750 | 0.0014 | - |
|
233 |
-
| 0.3676 | 800 | 0.0007 | - |
|
234 |
-
| 0.3906 | 850 | 0.0014 | - |
|
235 |
-
| 0.4136 | 900 | 0.0011 | - |
|
236 |
-
| 0.4366 | 950 | 0.0011 | - |
|
237 |
-
| 0.4596 | 1000 | 0.0017 | - |
|
238 |
-
| 0.4825 | 1050 | 0.0007 | - |
|
239 |
-
| 0.5055 | 1100 | 0.001 | - |
|
240 |
-
| 0.5285 | 1150 | 0.0008 | - |
|
241 |
-
| 0.5515 | 1200 | 0.0005 | - |
|
242 |
-
| 0.5744 | 1250 | 0.0009 | - |
|
243 |
-
| 0.5974 | 1300 | 0.0008 | - |
|
244 |
-
| 0.6204 | 1350 | 0.0013 | - |
|
245 |
-
| 0.6434 | 1400 | 0.0008 | - |
|
246 |
-
| 0.6664 | 1450 | 0.001 | - |
|
247 |
-
| 0.6893 | 1500 | 0.0006 | - |
|
248 |
-
| 0.7123 | 1550 | 0.0008 | - |
|
249 |
-
| 0.7353 | 1600 | 0.0006 | - |
|
250 |
-
| 0.7583 | 1650 | 0.0005 | - |
|
251 |
-
| 0.7812 | 1700 | 0.0006 | - |
|
252 |
-
| 0.8042 | 1750 | 0.0006 | - |
|
253 |
-
| 0.8272 | 1800 | 0.001 | - |
|
254 |
-
| 0.8502 | 1850 | 0.0005 | - |
|
255 |
-
| 0.8732 | 1900 | 0.0007 | - |
|
256 |
-
| 0.8961 | 1950 | 0.0009 | - |
|
257 |
-
| 0.9191 | 2000 | 0.0005 | - |
|
258 |
-
| 0.9421 | 2050 | 0.0005 | - |
|
259 |
-
| 0.9651 | 2100 | 0.0005 | - |
|
260 |
-
| 0.9881 | 2150 | 0.0005 | - |
|
261 |
-
| **1.0** | **2176** | **-** | **0.2021** |
|
262 |
-
| 0.0005 | 1 | 0.0004 | - |
|
263 |
-
| 0.0230 | 50 | 0.0006 | - |
|
264 |
-
| 0.0460 | 100 | 0.0005 | - |
|
265 |
-
| 0.0689 | 150 | 0.0005 | - |
|
266 |
-
| 0.0919 | 200 | 0.0004 | - |
|
267 |
-
| 0.1149 | 250 | 0.0005 | - |
|
268 |
-
| 0.1379 | 300 | 0.0004 | - |
|
269 |
-
| 0.1608 | 350 | 0.0018 | - |
|
270 |
-
| 0.1838 | 400 | 0.0005 | - |
|
271 |
-
| 0.2068 | 450 | 0.0003 | - |
|
272 |
-
| 0.2298 | 500 | 0.0003 | - |
|
273 |
-
| 0.2528 | 550 | 0.0003 | - |
|
274 |
-
| 0.2757 | 600 | 0.0002 | - |
|
275 |
-
| 0.2987 | 650 | 0.0003 | - |
|
276 |
-
| 0.3217 | 700 | 0.0004 | - |
|
277 |
-
| 0.3447 | 750 | 0.0002 | - |
|
278 |
-
| 0.3676 | 800 | 0.0002 | - |
|
279 |
-
| 0.3906 | 850 | 0.0006 | - |
|
280 |
-
| 0.4136 | 900 | 0.0002 | - |
|
281 |
-
| 0.4366 | 950 | 0.0002 | - |
|
282 |
-
| 0.4596 | 1000 | 0.0004 | - |
|
283 |
-
| 0.4825 | 1050 | 0.0004 | - |
|
284 |
-
| 0.5055 | 1100 | 0.0003 | - |
|
285 |
-
| 0.5285 | 1150 | 0.0002 | - |
|
286 |
-
| 0.5515 | 1200 | 0.0002 | - |
|
287 |
-
| 0.5744 | 1250 | 0.0002 | - |
|
288 |
-
| 0.5974 | 1300 | 0.0002 | - |
|
289 |
-
| 0.6204 | 1350 | 0.0003 | - |
|
290 |
-
| 0.6434 | 1400 | 0.0001 | - |
|
291 |
-
| 0.6664 | 1450 | 0.0003 | - |
|
292 |
-
| 0.6893 | 1500 | 0.0002 | - |
|
293 |
-
| 0.7123 | 1550 | 0.0003 | - |
|
294 |
-
| 0.7353 | 1600 | 0.0002 | - |
|
295 |
-
| 0.7583 | 1650 | 0.0002 | - |
|
296 |
-
| 0.7812 | 1700 | 0.0002 | - |
|
297 |
-
| 0.8042 | 1750 | 0.0002 | - |
|
298 |
-
| 0.8272 | 1800 | 0.0002 | - |
|
299 |
-
| 0.8502 | 1850 | 0.0002 | - |
|
300 |
-
| 0.8732 | 1900 | 0.0003 | - |
|
301 |
-
| 0.8961 | 1950 | 0.0003 | - |
|
302 |
-
| 0.9191 | 2000 | 0.0003 | - |
|
303 |
-
| 0.9421 | 2050 | 0.0002 | - |
|
304 |
-
| 0.9651 | 2100 | 0.0002 | - |
|
305 |
-
| 0.9881 | 2150 | 0.0002 | - |
|
306 |
-
| **1.0** | **2176** | **-** | **0.1685** |
|
307 |
-
|
308 |
-
* The bold row denotes the saved checkpoint.
|
309 |
### Framework Versions
|
310 |
- Python: 3.10.12
|
311 |
- SetFit: 1.0.3
|
312 |
-
- Sentence Transformers: 2.
|
313 |
-
- Transformers: 4.
|
314 |
- PyTorch: 2.2.1+cu121
|
315 |
-
- Datasets: 2.
|
316 |
-
- Tokenizers: 0.
|
317 |
|
318 |
## Citation
|
319 |
|
|
|
5 |
- sentence-transformers
|
6 |
- text-classification
|
7 |
- generated_from_setfit_trainer
|
8 |
+
base_model: adriansanz/halfine
|
9 |
metrics:
|
10 |
- accuracy
|
11 |
widget:
|
12 |
+
- text: 'partits de futbol en vía pública que generen sorrolls i desperfectes : soc
|
13 |
+
una veina de la plaça victor mateu i moles. durant tot l’any el jovent juguen
|
14 |
+
a fútbol en la part asfaltada de la plaça i disparen les pilotes contra la tanca
|
15 |
+
del recinte que correspon als portals 1, 2 i 3 de dita plaça. aquests partits
|
16 |
+
s’allarguen fins passades les 12 de la nit i a més provoquen l’enfonsament de
|
17 |
+
la tanca metàl•lica. avisada la guàrdia urbana en diverses ocasions en les que
|
18 |
+
han fet cas omís. aquests fets pertorben el descans dels veïns a més dels desperfectes
|
19 |
+
que ocasionen i que haurem de pagar entre els veïns de la comunitat. aquests fets
|
20 |
+
s’estan produint des de la remodelació de la plaça. gràcies per la vostra atenció.'
|
21 |
+
- text: 'una bombeta fosa en una farola : al carrer antoni agusti al nº 9 hi ha una
|
22 |
+
farola amb una bombeta fosa fa dies que i está'
|
23 |
+
- text: 'vehicle circulant amb itv caducada : el vehicle marca hundai civic de color
|
24 |
+
verd provist de plaques de matricula 2449 cyj circula pels carrers dels voltants
|
25 |
+
del arnau de vilanova amb la itv caducada del març del 23 per tant aquest vehicle
|
26 |
+
cal que passi la inspeccio tecnica corresponent per a garantir la seguretat de
|
27 |
+
la resta d''usuaris de la zona....gracies.....apercebin al conductor o propietari
|
28 |
+
al mes aviat possible.....'
|
29 |
+
- text: 'devolució fiances : bona tarda, voldria saber el procediment a seguir per
|
30 |
+
sol•licitar la devolució d’unes fiances que es varen constituir per realitzar
|
31 |
+
unes obres, les quals ja han acabat. no sé si depèn d''urbanisme, via pública,
|
32 |
+
tramitacions... disculpeu. atentament marc castelló'
|
33 |
+
- text: 'senyal vertical d,obligacio de circulacion giratoria despenjada : la senyal
|
34 |
+
vertical circular de obligacio de circulacio giratoria situada a la glorieta de
|
35 |
+
dr.fleming amb avinguda 11 de setembre,,,,,davant centre penitenciari ,,,roman
|
36 |
+
despenjada al terra,,,,caldria tornar la a penjar en el seu lloc......gracies....revisin
|
37 |
+
la zona.....'
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
38 |
pipeline_tag: text-classification
|
39 |
inference: true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
40 |
---
|
41 |
|
42 |
+
# SetFit with adriansanz/halfine
|
43 |
|
44 |
+
This is a [SetFit](https://github.com/huggingface/setfit) model that can be used for Text Classification. This SetFit model uses [adriansanz/halfine](https://huggingface.co/adriansanz/halfine) as the Sentence Transformer embedding model. A [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance is used for classification.
|
45 |
|
46 |
The model has been trained using an efficient few-shot learning technique that involves:
|
47 |
|
|
|
52 |
|
53 |
### Model Description
|
54 |
- **Model Type:** SetFit
|
55 |
+
- **Sentence Transformer body:** [adriansanz/halfine](https://huggingface.co/adriansanz/halfine)
|
56 |
- **Classification head:** a [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance
|
57 |
- **Maximum Sequence Length:** 512 tokens
|
58 |
- **Number of Classes:** 17 classes
|
|
|
67 |
- **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
|
68 |
|
69 |
### Model Labels
|
70 |
+
| Label | Examples |
|
71 |
+
|:------|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
|
72 |
+
| 0 | <ul><li>'Aquest article tracta sobre Aigües'</li><li>'Aquest article tracta sobre Aigües'</li><li>'Aquest article tracta sobre Aigües'</li></ul> |
|
73 |
+
| 1 | <ul><li>'Aquest article tracta sobre Consum, comerç i mercats'</li><li>'Aquest article tracta sobre Consum, comerç i mercats'</li><li>'Aquest article tracta sobre Consum, comerç i mercats'</li></ul> |
|
74 |
+
| 2 | <ul><li>'Aquest article tracta sobre Cultura'</li><li>'Aquest article tracta sobre Cultura'</li><li>'Aquest article tracta sobre Cultura'</li></ul> |
|
75 |
+
| 3 | <ul><li>'Aquest article tracta sobre Economia'</li><li>'Aquest article tracta sobre Economia'</li><li>'Aquest article tracta sobre Economia'</li></ul> |
|
76 |
+
| 4 | <ul><li>'Aquest article tracta sobre Educació'</li><li>'Aquest article tracta sobre Educació'</li><li>'Aquest article tracta sobre Educació'</li></ul> |
|
77 |
+
| 5 | <ul><li>'Aquest article tracta sobre Enllumenat públic'</li><li>'Aquest article tracta sobre Enllumenat públic'</li><li>'Aquest article tracta sobre Enllumenat públic'</li></ul> |
|
78 |
+
| 6 | <ul><li>'Aquest article tracta sobre Esports'</li><li>'Aquest article tracta sobre Esports'</li><li>'Aquest article tracta sobre Esports'</li></ul> |
|
79 |
+
| 7 | <ul><li>'Aquest article tracta sobre Habitatge'</li><li>'Aquest article tracta sobre Habitatge'</li><li>'Aquest article tracta sobre Habitatge'</li></ul> |
|
80 |
+
| 8 | <ul><li>'Aquest article tracta sobre Horta'</li><li>'Aquest article tracta sobre Horta'</li><li>'Aquest article tracta sobre Horta'</li></ul> |
|
81 |
+
| 9 | <ul><li>'Aquest article tracta sobre Medi ambient'</li><li>'Aquest article tracta sobre Medi ambient'</li><li>'Aquest article tracta sobre Medi ambient'</li></ul> |
|
82 |
+
| 10 | <ul><li>'Aquest article tracta sobre Neteja de la via pública'</li><li>'Aquest article tracta sobre Neteja de la via pública'</li><li>'Aquest article tracta sobre Neteja de la via pública'</li></ul> |
|
83 |
+
| 11 | <ul><li>'Aquest article tracta sobre Salut pública i Cementiri'</li><li>'Aquest article tracta sobre Salut pública i Cementiri'</li><li>'Aquest article tracta sobre Salut pública i Cementiri'</li></ul> |
|
84 |
+
| 12 | <ul><li>'Aquest article tracta sobre Seguretat'</li><li>'Aquest article tracta sobre Seguretat'</li><li>'Aquest article tracta sobre Seguretat'</li></ul> |
|
85 |
+
| 13 | <ul><li>'Aquest article tracta sobre Serveis socials'</li><li>'Aquest article tracta sobre Serveis socials'</li><li>'Aquest article tracta sobre Serveis socials'</li></ul> |
|
86 |
+
| 14 | <ul><li>'Aquest article tracta sobre Tramitacions'</li><li>'Aquest article tracta sobre Tramitacions'</li><li>'Aquest article tracta sobre Tramitacions'</li></ul> |
|
87 |
+
| 15 | <ul><li>'Aquest article tracta sobre Urbanisme'</li><li>'Aquest article tracta sobre Urbanisme'</li><li>'Aquest article tracta sobre Urbanisme'</li></ul> |
|
88 |
+
| 16 | <ul><li>'Aquest article tracta sobre Via pública i mobilitat'</li><li>'Aquest article tracta sobre Via pública i mobilitat'</li><li>'Aquest article tracta sobre Via pública i mobilitat'</li></ul> |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
89 |
|
90 |
## Uses
|
91 |
|
|
|
105 |
# Download from the 🤗 Hub
|
106 |
model = SetFitModel.from_pretrained("adriansanz/test8")
|
107 |
# Run inference
|
108 |
+
preds = model("una bombeta fosa en una farola : al carrer antoni agusti al nº 9 hi ha una farola amb una bombeta fosa fa dies que i está")
|
109 |
```
|
110 |
|
111 |
<!--
|
|
|
137 |
### Training Set Metrics
|
138 |
| Training set | Min | Median | Max |
|
139 |
|:-------------|:----|:-------|:----|
|
140 |
+
| Word count | 5 | 5.9412 | 9 |
|
141 |
|
142 |
| Label | Training Sample Count |
|
143 |
|:------|:----------------------|
|
|
|
160 |
| 16 | 8 |
|
161 |
|
162 |
### Training Hyperparameters
|
163 |
+
- batch_size: (16, 2)
|
164 |
+
- num_epochs: (1, 16)
|
165 |
- max_steps: -1
|
166 |
- sampling_strategy: oversampling
|
167 |
- body_learning_rate: (2e-05, 1e-05)
|
|
|
176 |
- eval_max_steps: -1
|
177 |
- load_best_model_at_end: False
|
178 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
179 |
### Framework Versions
|
180 |
- Python: 3.10.12
|
181 |
- SetFit: 1.0.3
|
182 |
+
- Sentence Transformers: 2.7.0
|
183 |
+
- Transformers: 4.40.1
|
184 |
- PyTorch: 2.2.1+cu121
|
185 |
+
- Datasets: 2.19.1
|
186 |
+
- Tokenizers: 0.19.1
|
187 |
|
188 |
## Citation
|
189 |
|
config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "
|
3 |
"architectures": [
|
4 |
"RobertaModel"
|
5 |
],
|
@@ -13,16 +13,16 @@
|
|
13 |
"hidden_dropout_prob": 0.1,
|
14 |
"hidden_size": 768,
|
15 |
"id2label": {
|
16 |
-
"0": "
|
17 |
-
"1": "
|
18 |
-
"2": "
|
19 |
},
|
20 |
"initializer_range": 0.02,
|
21 |
"intermediate_size": 3072,
|
22 |
"label2id": {
|
23 |
-
"
|
24 |
-
"
|
25 |
-
"
|
26 |
},
|
27 |
"layer_norm_eps": 1e-05,
|
28 |
"max_position_embeddings": 514,
|
@@ -33,7 +33,7 @@
|
|
33 |
"position_embedding_type": "absolute",
|
34 |
"problem_type": "single_label_classification",
|
35 |
"torch_dtype": "float32",
|
36 |
-
"transformers_version": "4.
|
37 |
"type_vocab_size": 1,
|
38 |
"use_cache": true,
|
39 |
"vocab_size": 50262
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "adriansanz/halfine",
|
3 |
"architectures": [
|
4 |
"RobertaModel"
|
5 |
],
|
|
|
13 |
"hidden_dropout_prob": 0.1,
|
14 |
"hidden_size": 768,
|
15 |
"id2label": {
|
16 |
+
"0": "ENTAILMENT",
|
17 |
+
"1": "NEUTRAL",
|
18 |
+
"2": "CONTRADICTION"
|
19 |
},
|
20 |
"initializer_range": 0.02,
|
21 |
"intermediate_size": 3072,
|
22 |
"label2id": {
|
23 |
+
"CONTRADICTION": 2,
|
24 |
+
"ENTAILMENT": 0,
|
25 |
+
"NEUTRAL": 1
|
26 |
},
|
27 |
"layer_norm_eps": 1e-05,
|
28 |
"max_position_embeddings": 514,
|
|
|
33 |
"position_embedding_type": "absolute",
|
34 |
"problem_type": "single_label_classification",
|
35 |
"torch_dtype": "float32",
|
36 |
+
"transformers_version": "4.40.1",
|
37 |
"type_vocab_size": 1,
|
38 |
"use_cache": true,
|
39 |
"vocab_size": 50262
|
config_sentence_transformers.json
CHANGED
@@ -1,7 +1,7 @@
|
|
1 |
{
|
2 |
"__version__": {
|
3 |
-
"sentence_transformers": "2.
|
4 |
-
"transformers": "4.
|
5 |
"pytorch": "2.2.1+cu121"
|
6 |
},
|
7 |
"prompts": {},
|
|
|
1 |
{
|
2 |
"__version__": {
|
3 |
+
"sentence_transformers": "2.7.0",
|
4 |
+
"transformers": "4.40.1",
|
5 |
"pytorch": "2.2.1+cu121"
|
6 |
},
|
7 |
"prompts": {},
|
config_setfit.json
CHANGED
@@ -1,4 +1,4 @@
|
|
1 |
{
|
2 |
-
"
|
3 |
-
"
|
4 |
}
|
|
|
1 |
{
|
2 |
+
"normalize_embeddings": false,
|
3 |
+
"labels": null
|
4 |
}
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 498595688
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:098e7b6967b5a260cdb0d4676b3f3e81f7ed25d9d77b80d62ce76710dce500e2
|
3 |
size 498595688
|
model_head.pkl
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:95a2349eb3f96eb70d9185b2f02af4ffabe28566b0bbeda0ad56312e9e6879cb
|
3 |
+
size 324
|
tokenizer.json
CHANGED
@@ -3,7 +3,7 @@
|
|
3 |
"truncation": {
|
4 |
"direction": "Right",
|
5 |
"max_length": 512,
|
6 |
-
"strategy": "
|
7 |
"stride": 0
|
8 |
},
|
9 |
"padding": {
|
@@ -95,6 +95,7 @@
|
|
95 |
"end_of_word_suffix": "",
|
96 |
"fuse_unk": false,
|
97 |
"byte_fallback": false,
|
|
|
98 |
"vocab": {
|
99 |
"<s>": 0,
|
100 |
"<pad>": 1,
|
|
|
3 |
"truncation": {
|
4 |
"direction": "Right",
|
5 |
"max_length": 512,
|
6 |
+
"strategy": "OnlyFirst",
|
7 |
"stride": 0
|
8 |
},
|
9 |
"padding": {
|
|
|
95 |
"end_of_word_suffix": "",
|
96 |
"fuse_unk": false,
|
97 |
"byte_fallback": false,
|
98 |
+
"ignore_merges": false,
|
99 |
"vocab": {
|
100 |
"<s>": 0,
|
101 |
"<pad>": 1,
|