kuljeet98 commited on
Commit
1f850dd
1 Parent(s): 472d7a5

Upload 11 files

Browse files
Files changed (2) hide show
  1. README.md +93 -37
  2. pytorch_model.bin +1 -1
README.md CHANGED
@@ -9,35 +9,34 @@ tags:
9
  - sentence-similarity
10
  - feature-extraction
11
  - generated_from_trainer
12
- - dataset_size:1043
13
  - loss:CosineSimilarityLoss
14
  widget:
15
- - source_sentence: vice president customer quality
16
  sentences:
17
- - executive committee senior vice president corp development strategy bd&l & mergers
18
- and acquisitions
19
- - vice president & chief financial officer medical division
20
- - vice president
21
- - source_sentence: chief officer of human resources
22
  sentences:
23
- - chief human resources officer
24
- - director healthcare investment banking
25
- - consultor en desarrollo empresarial y capacitador part time
26
- - source_sentence: gerente general en safs de tamaño medio
27
  sentences:
28
- - director ejecutivo
29
- - ' president respiratory interventions operating unit'
30
- - senior facilities engineering manager
31
- - source_sentence: chief technology and digital officer
32
  sentences:
33
- - vice president product international
34
- - executive director digital applications
35
- - senior vice president operations
36
- - source_sentence: vice president of accounting
37
  sentences:
38
- - director of product management
39
- - regional vice president of operations
40
- - purchasing coordinator hv battery and powertrain
41
  ---
42
 
43
  # SentenceTransformer based on sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
@@ -89,9 +88,9 @@ from sentence_transformers import SentenceTransformer
89
  model = SentenceTransformer("sentence_transformers_model_id")
90
  # Run inference
91
  sentences = [
92
- 'vice president of accounting',
93
- 'purchasing coordinator hv battery and powertrain',
94
- 'regional vice president of operations',
95
  ]
96
  embeddings = model.encode(sentences)
97
  print(embeddings.shape)
@@ -146,19 +145,19 @@ You can finetune this model on your own dataset.
146
  #### Unnamed Dataset
147
 
148
 
149
- * Size: 1,043 training samples
150
  * Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>label</code>
151
  * Approximate statistics based on the first 1000 samples:
152
- | | sentence_0 | sentence_1 | label |
153
- |:--------|:---------------------------------------------------------------------------------|:---------------------------------------------------------------------------------|:---------------------------------------------------------------|
154
- | type | string | string | float |
155
- | details | <ul><li>min: 3 tokens</li><li>mean: 6.62 tokens</li><li>max: 12 tokens</li></ul> | <ul><li>min: 3 tokens</li><li>mean: 7.77 tokens</li><li>max: 21 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.13</li><li>max: 1.0</li></ul> |
156
  * Samples:
157
- | sentence_0 | sentence_1 | label |
158
- |:---------------------------------------------|:---------------------------------------------------------------------------|:-----------------|
159
- | <code>vice president quality</code> | <code>director merger & acquisition mergers and acquisitions office</code> | <code>0.0</code> |
160
- | <code>chief executive officer</code> | <code>vice president defense business unit</code> | <code>0.0</code> |
161
- | <code>vice president customer quality</code> | <code>director major & planned gifts regions hospital foundation</code> | <code>0.0</code> |
162
  * Loss: [<code>CosineSimilarityLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosinesimilarityloss) with these parameters:
163
  ```json
164
  {
@@ -171,7 +170,7 @@ You can finetune this model on your own dataset.
171
 
172
  - `per_device_train_batch_size`: 16
173
  - `per_device_eval_batch_size`: 16
174
- - `num_train_epochs`: 1
175
  - `multi_dataset_batch_sampler`: round_robin
176
 
177
  #### All Hyperparameters
@@ -193,7 +192,7 @@ You can finetune this model on your own dataset.
193
  - `adam_beta2`: 0.999
194
  - `adam_epsilon`: 1e-08
195
  - `max_grad_norm`: 1
196
- - `num_train_epochs`: 1
197
  - `max_steps`: -1
198
  - `lr_scheduler_type`: linear
199
  - `lr_scheduler_kwargs`: {}
@@ -288,6 +287,63 @@ You can finetune this model on your own dataset.
288
 
289
  </details>
290
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
291
  ### Framework Versions
292
  - Python: 3.8.5
293
  - Sentence Transformers: 3.0.1
 
9
  - sentence-similarity
10
  - feature-extraction
11
  - generated_from_trainer
12
+ - dataset_size:8408
13
  - loss:CosineSimilarityLoss
14
  widget:
15
+ - source_sentence: president
16
  sentences:
17
+ - assistante de banque priv e banco santander rio
18
+ - worldwide executive vice president corindus a siemens healthineers company
19
+ - soporte t cnico superior
20
+ - source_sentence: chief business strategy officer
 
21
  sentences:
22
+ - sub jefe
23
+ - analista senior recursos humanos sales staff and logistics
24
+ - subgerente sostenibilidad y hseq
25
+ - source_sentence: gerente de planificación
26
  sentences:
27
+ - analista de soporte web
28
+ - director
29
+ - gestion calidad
30
+ - source_sentence: global human resources leader
31
  sentences:
32
+ - director manufacturing engineering
33
+ - quality specialist
34
+ - asesoramiento para comprar inmuebles en uruguay paraguay espa a y usa
35
+ - source_sentence: commercial manager
36
  sentences:
37
+ - jefe de turno planta envasado de vinos
38
+ - gerente de operaciones
39
+ - vice president of finance americas
40
  ---
41
 
42
  # SentenceTransformer based on sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
 
88
  model = SentenceTransformer("sentence_transformers_model_id")
89
  # Run inference
90
  sentences = [
91
+ 'commercial manager',
92
+ 'gerente de operaciones',
93
+ 'vice president of finance americas',
94
  ]
95
  embeddings = model.encode(sentences)
96
  print(embeddings.shape)
 
145
  #### Unnamed Dataset
146
 
147
 
148
+ * Size: 8,408 training samples
149
  * Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>label</code>
150
  * Approximate statistics based on the first 1000 samples:
151
+ | | sentence_0 | sentence_1 | label |
152
+ |:--------|:--------------------------------------------------------------------------------|:---------------------------------------------------------------------------------|:---------------------------------------------------------------|
153
+ | type | string | string | float |
154
+ | details | <ul><li>min: 3 tokens</li><li>mean: 6.2 tokens</li><li>max: 12 tokens</li></ul> | <ul><li>min: 3 tokens</li><li>mean: 7.75 tokens</li><li>max: 21 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.06</li><li>max: 1.0</li></ul> |
155
  * Samples:
156
+ | sentence_0 | sentence_1 | label |
157
+ |:----------------------------------------|:------------------------------------------------------------------------------|:-----------------|
158
+ | <code>strategic planning manager</code> | <code>senior brand manager uap southern cone & personal care cdm chile</code> | <code>0.0</code> |
159
+ | <code>director de planificacion</code> | <code>key account manager tiendas paris</code> | <code>0.0</code> |
160
+ | <code>gerente general</code> | <code>analista de cobranza</code> | <code>0.0</code> |
161
  * Loss: [<code>CosineSimilarityLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosinesimilarityloss) with these parameters:
162
  ```json
163
  {
 
170
 
171
  - `per_device_train_batch_size`: 16
172
  - `per_device_eval_batch_size`: 16
173
+ - `num_train_epochs`: 50
174
  - `multi_dataset_batch_sampler`: round_robin
175
 
176
  #### All Hyperparameters
 
192
  - `adam_beta2`: 0.999
193
  - `adam_epsilon`: 1e-08
194
  - `max_grad_norm`: 1
195
+ - `num_train_epochs`: 50
196
  - `max_steps`: -1
197
  - `lr_scheduler_type`: linear
198
  - `lr_scheduler_kwargs`: {}
 
287
 
288
  </details>
289
 
290
+ ### Training Logs
291
+ | Epoch | Step | Training Loss |
292
+ |:-------:|:-----:|:-------------:|
293
+ | 0.9506 | 500 | 0.0434 |
294
+ | 1.9011 | 1000 | 0.0135 |
295
+ | 2.8517 | 1500 | 0.0072 |
296
+ | 3.8023 | 2000 | 0.0056 |
297
+ | 4.7529 | 2500 | 0.0044 |
298
+ | 5.7034 | 3000 | 0.0038 |
299
+ | 6.6540 | 3500 | 0.0034 |
300
+ | 7.6046 | 4000 | 0.0032 |
301
+ | 8.5551 | 4500 | 0.0029 |
302
+ | 9.5057 | 5000 | 0.0028 |
303
+ | 10.4563 | 5500 | 0.0026 |
304
+ | 11.4068 | 6000 | 0.0025 |
305
+ | 12.3574 | 6500 | 0.0026 |
306
+ | 13.3080 | 7000 | 0.0023 |
307
+ | 14.2586 | 7500 | 0.0023 |
308
+ | 15.2091 | 8000 | 0.0023 |
309
+ | 16.1597 | 8500 | 0.0022 |
310
+ | 17.1103 | 9000 | 0.0021 |
311
+ | 18.0608 | 9500 | 0.0019 |
312
+ | 19.0114 | 10000 | 0.0021 |
313
+ | 19.9620 | 10500 | 0.0019 |
314
+ | 20.9125 | 11000 | 0.0019 |
315
+ | 21.8631 | 11500 | 0.0016 |
316
+ | 22.8137 | 12000 | 0.0018 |
317
+ | 23.7643 | 12500 | 0.0018 |
318
+ | 24.7148 | 13000 | 0.0018 |
319
+ | 25.6654 | 13500 | 0.0016 |
320
+ | 26.6160 | 14000 | 0.0017 |
321
+ | 27.5665 | 14500 | 0.0016 |
322
+ | 28.5171 | 15000 | 0.0016 |
323
+ | 29.4677 | 15500 | 0.0016 |
324
+ | 30.4183 | 16000 | 0.0016 |
325
+ | 31.3688 | 16500 | 0.0019 |
326
+ | 32.3194 | 17000 | 0.0018 |
327
+ | 33.2700 | 17500 | 0.0017 |
328
+ | 34.2205 | 18000 | 0.0016 |
329
+ | 35.1711 | 18500 | 0.0016 |
330
+ | 36.1217 | 19000 | 0.0016 |
331
+ | 37.0722 | 19500 | 0.0015 |
332
+ | 38.0228 | 20000 | 0.0012 |
333
+ | 38.9734 | 20500 | 0.0015 |
334
+ | 39.9240 | 21000 | 0.0015 |
335
+ | 40.8745 | 21500 | 0.0013 |
336
+ | 41.8251 | 22000 | 0.0014 |
337
+ | 42.7757 | 22500 | 0.0014 |
338
+ | 43.7262 | 23000 | 0.0014 |
339
+ | 44.6768 | 23500 | 0.0013 |
340
+ | 45.6274 | 24000 | 0.0012 |
341
+ | 46.5779 | 24500 | 0.0014 |
342
+ | 47.5285 | 25000 | 0.0012 |
343
+ | 48.4791 | 25500 | 0.0013 |
344
+ | 49.4297 | 26000 | 0.0013 |
345
+
346
+
347
  ### Framework Versions
348
  - Python: 3.8.5
349
  - Sentence Transformers: 3.0.1
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:816e595448b22d534934bd93f8b0e1346023f3f8c86da7b779543569ca34e9c4
3
  size 470682214
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b69a4fffcc06b0364d9c0218b2133c7f68d5baaf76631b46ef8a023f61430847
3
  size 470682214