JordanTallon commited on
Commit
2728b49
1 Parent(s): 2a48c96

Push model using huggingface_hub.

Browse files
1_Pooling/config.json CHANGED
@@ -3,5 +3,7 @@
3
  "pooling_mode_cls_token": true,
4
  "pooling_mode_mean_tokens": false,
5
  "pooling_mode_max_tokens": false,
6
- "pooling_mode_mean_sqrt_len_tokens": false
 
 
7
  }
 
3
  "pooling_mode_cls_token": true,
4
  "pooling_mode_mean_tokens": false,
5
  "pooling_mode_max_tokens": false,
6
+ "pooling_mode_mean_sqrt_len_tokens": false,
7
+ "pooling_mode_weightedmean_tokens": false,
8
+ "pooling_mode_lasttoken": false
9
  }
README.md CHANGED
@@ -6,27 +6,25 @@ tags:
6
  - text-classification
7
  - generated_from_setfit_trainer
8
  metrics:
9
- - f1
10
  - precision
11
  - recall
12
- - accuracy
13
  widget:
14
- - text: Since the start of the coronavirus outbreak, Trump has hampered efforts to
15
- slow the virus’s spread and encouraged Americans’ restlessness under quarantine.
16
- - text: ' It has to be particularly described what he is looking for said Asha Rangappa
17
- who was a counter intelligence agent for the FBI and now a Yale Law School professor
18
- A judge isn t going to sign off some sort of blanket warrant that tells Facebook
19
- to turn over everything '
20
- - text: 'Now in response to these very serious crises it seems to me that we have
21
- two choices First we can throw up our hands in despair We can say I am not going
22
- to get involved '
23
- - text: Over the past week, activists, some of who are believed to be affiliated with
24
- Black Lives Matter have rioted across the country following the death of George
25
- Floyd in police custody, wreaking havoc and destruction against America’s towns,
26
- cities, and local communities. 
27
- - text: Working-class Americans, like those who make up the majority of South Bend
28
- residents, have secured the largest wage hikes in the nation compared to all other
29
- economic demographic groups — a direct result of Trump tightening the labor market.
30
  pipeline_tag: text-classification
31
  inference: true
32
  base_model: BAAI/bge-small-en-v1.5
@@ -41,18 +39,18 @@ model-index:
41
  type: unknown
42
  split: test
43
  metrics:
44
- - type: f1
45
- value: 0.6952861952861953
46
- name: F1
47
  - type: precision
48
- value: 0.6952861952861953
49
  name: Precision
50
  - type: recall
51
- value: 0.6952861952861953
52
  name: Recall
53
- - type: accuracy
54
- value: 0.6952861952861953
55
- name: Accuracy
56
  ---
57
 
58
  # SetFit with BAAI/bge-small-en-v1.5
@@ -83,18 +81,18 @@ The model has been trained using an efficient few-shot learning technique that i
83
  - **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
84
 
85
  ### Model Labels
86
- | Label | Examples |
87
- |:-------|:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
88
- | center | <ul><li>'A leading economist who vouched for Democratic presidential candidate Elizabeth Warren’s healthcare reform plan told Reuters on Thursday he doubts its staggering cost can be fully covered alongside her other government programs.'</li><li>'U.S. President Donald Trump is doing well and is very healthy, White House adviser Kellyanne Conway told Fox News on Thursday, after a U.S. military official who worked at the White House was found to have been infected with the novel coronavirus.'</li><li>'Alabama has the most restrictive abortion law in the U.S., banning abortion at any stage of pregnancy and for any reason, including in cases of rape and incest.'</li></ul> |
89
- | left | <ul><li>'Meet the shadowy accountants who do Trump’s taxes and help him seem richer than he is'</li><li>'When did vaccines become politicized? Amid a measles outbreak, suddenly Republicans support anti-vaxxers.'</li><li>'Last summer, the Republican White House announced plans to roll back the tougher standards, making it easier for the automotive industry to sell less efficient vehicles that pollute more.'</li></ul> |
90
- | right | <ul><li>'Joe Biden told Wall Street donors to his campaign that he planned to reverse most of President Donald Trump’s tax cuts.'</li><li>'For far too many on the left, chaos is the point. Destruction is the goal. They prefer the unknown madness that lies ahead to whatever is still managing to (barely) hold us together in the present.'</li><li>'Cuba’s health ministry initially vowed an investigation into Paloma Dominguez Caballero’s death; last week, state media published a report essentially absolving the government of any wrongdoing, categorically stating that nothing was wrong with the vaccine Dominguez received.'</li></ul> |
91
 
92
  ## Evaluation
93
 
94
  ### Metrics
95
- | Label | F1 | Precision | Recall | Accuracy |
96
- |:--------|:-------|:----------|:-------|:---------|
97
- | **all** | 0.6953 | 0.6953 | 0.6953 | 0.6953 |
98
 
99
  ## Uses
100
 
@@ -114,7 +112,7 @@ from setfit import SetFitModel
114
  # Download from the 🤗 Hub
115
  model = SetFitModel.from_pretrained("JordanTallon/Unifeed")
116
  # Run inference
117
- preds = model("Since the start of the coronavirus outbreak, Trump has hampered efforts to slow the virus’s spread and encouraged Americans’ restlessness under quarantine.")
118
  ```
119
 
120
  <!--
@@ -146,223 +144,540 @@ preds = model("Since the start of the coronavirus outbreak, Trump has hampered e
146
  ### Training Set Metrics
147
  | Training set | Min | Median | Max |
148
  |:-------------|:----|:--------|:----|
149
- | Word count | 6 | 33.1655 | 86 |
150
 
151
  | Label | Training Sample Count |
152
  |:-------|:----------------------|
153
- | center | 802 |
154
- | left | 784 |
155
- | right | 788 |
156
 
157
  ### Training Hyperparameters
158
  - batch_size: (32, 32)
159
- - num_epochs: (3, 3)
160
  - max_steps: -1
161
  - sampling_strategy: oversampling
162
- - num_iterations: 20
163
- - body_learning_rate: (2e-05, 2e-05)
164
- - head_learning_rate: 2e-05
165
  - loss: CosineSimilarityLoss
166
  - distance_metric: cosine_distance
167
  - margin: 0.25
168
  - end_to_end: False
169
- - use_amp: False
170
  - warmup_proportion: 0.1
171
- - seed: 42
 
172
  - eval_max_steps: -1
173
- - load_best_model_at_end: False
174
 
175
  ### Training Results
176
- | Epoch | Step | Training Loss | Validation Loss |
177
- |:------:|:----:|:-------------:|:---------------:|
178
- | 0.0003 | 1 | 0.2552 | - |
179
- | 0.0168 | 50 | 0.2613 | - |
180
- | 0.0337 | 100 | 0.2653 | - |
181
- | 0.0505 | 150 | 0.2574 | - |
182
- | 0.0674 | 200 | 0.2455 | - |
183
- | 0.0842 | 250 | 0.2583 | - |
184
- | 0.1011 | 300 | 0.2736 | - |
185
- | 0.1179 | 350 | 0.2341 | - |
186
- | 0.1348 | 400 | 0.2524 | - |
187
- | 0.1516 | 450 | 0.2429 | - |
188
- | 0.1685 | 500 | 0.2579 | - |
189
- | 0.1853 | 550 | 0.2363 | - |
190
- | 0.2022 | 600 | 0.2789 | - |
191
- | 0.2190 | 650 | 0.186 | - |
192
- | 0.2358 | 700 | 0.2425 | - |
193
- | 0.2527 | 750 | 0.1963 | - |
194
- | 0.2695 | 800 | 0.1858 | - |
195
- | 0.2864 | 850 | 0.1499 | - |
196
- | 0.3032 | 900 | 0.2219 | - |
197
- | 0.3201 | 950 | 0.1376 | - |
198
- | 0.3369 | 1000 | 0.1115 | - |
199
- | 0.3538 | 1050 | 0.1205 | - |
200
- | 0.3706 | 1100 | 0.1398 | - |
201
- | 0.3875 | 1150 | 0.1585 | - |
202
- | 0.4043 | 1200 | 0.1328 | - |
203
- | 0.4212 | 1250 | 0.0954 | - |
204
- | 0.4380 | 1300 | 0.0707 | - |
205
- | 0.4549 | 1350 | 0.2214 | - |
206
- | 0.4717 | 1400 | 0.1351 | - |
207
- | 0.4885 | 1450 | 0.1249 | - |
208
- | 0.5054 | 1500 | 0.1656 | - |
209
- | 0.5222 | 1550 | 0.1573 | - |
210
- | 0.5391 | 1600 | 0.1103 | - |
211
- | 0.5559 | 1650 | 0.0787 | - |
212
- | 0.5728 | 1700 | 0.126 | - |
213
- | 0.5896 | 1750 | 0.0876 | - |
214
- | 0.6065 | 1800 | 0.1687 | - |
215
- | 0.6233 | 1850 | 0.1319 | - |
216
- | 0.6402 | 1900 | 0.0815 | - |
217
- | 0.6570 | 1950 | 0.09 | - |
218
- | 0.6739 | 2000 | 0.0471 | - |
219
- | 0.6907 | 2050 | 0.1032 | - |
220
- | 0.7075 | 2100 | 0.0858 | - |
221
- | 0.7244 | 2150 | 0.0859 | - |
222
- | 0.7412 | 2200 | 0.0946 | - |
223
- | 0.7581 | 2250 | 0.0618 | - |
224
- | 0.7749 | 2300 | 0.0233 | - |
225
- | 0.7918 | 2350 | 0.0148 | - |
226
- | 0.8086 | 2400 | 0.0367 | - |
227
- | 0.8255 | 2450 | 0.0111 | - |
228
- | 0.8423 | 2500 | 0.0034 | - |
229
- | 0.8592 | 2550 | 0.0174 | - |
230
- | 0.8760 | 2600 | 0.0304 | - |
231
- | 0.8929 | 2650 | 0.0303 | - |
232
- | 0.9097 | 2700 | 0.0031 | - |
233
- | 0.9265 | 2750 | 0.0058 | - |
234
- | 0.9434 | 2800 | 0.0034 | - |
235
- | 0.9602 | 2850 | 0.0011 | - |
236
- | 0.9771 | 2900 | 0.0013 | - |
237
- | 0.9939 | 2950 | 0.0296 | - |
238
- | 1.0108 | 3000 | 0.0008 | - |
239
- | 1.0276 | 3050 | 0.0189 | - |
240
- | 1.0445 | 3100 | 0.0295 | - |
241
- | 1.0613 | 3150 | 0.0276 | - |
242
- | 1.0782 | 3200 | 0.0008 | - |
243
- | 1.0950 | 3250 | 0.0008 | - |
244
- | 1.1119 | 3300 | 0.0009 | - |
245
- | 1.1287 | 3350 | 0.0009 | - |
246
- | 1.1456 | 3400 | 0.0008 | - |
247
- | 1.1624 | 3450 | 0.0099 | - |
248
- | 1.1792 | 3500 | 0.0009 | - |
249
- | 1.1961 | 3550 | 0.0299 | - |
250
- | 1.2129 | 3600 | 0.0007 | - |
251
- | 1.2298 | 3650 | 0.001 | - |
252
- | 1.2466 | 3700 | 0.0009 | - |
253
- | 1.2635 | 3750 | 0.0008 | - |
254
- | 1.2803 | 3800 | 0.001 | - |
255
- | 1.2972 | 3850 | 0.0009 | - |
256
- | 1.3140 | 3900 | 0.0008 | - |
257
- | 1.3309 | 3950 | 0.0007 | - |
258
- | 1.3477 | 4000 | 0.0007 | - |
259
- | 1.3646 | 4050 | 0.03 | - |
260
- | 1.3814 | 4100 | 0.0008 | - |
261
- | 1.3982 | 4150 | 0.0012 | - |
262
- | 1.4151 | 4200 | 0.0292 | - |
263
- | 1.4319 | 4250 | 0.0006 | - |
264
- | 1.4488 | 4300 | 0.0007 | - |
265
- | 1.4656 | 4350 | 0.0006 | - |
266
- | 1.4825 | 4400 | 0.0007 | - |
267
- | 1.4993 | 4450 | 0.0008 | - |
268
- | 1.5162 | 4500 | 0.0008 | - |
269
- | 1.5330 | 4550 | 0.0015 | - |
270
- | 1.5499 | 4600 | 0.0032 | - |
271
- | 1.5667 | 4650 | 0.0015 | - |
272
- | 1.5836 | 4700 | 0.0006 | - |
273
- | 1.6004 | 4750 | 0.0006 | - |
274
- | 1.6173 | 4800 | 0.0021 | - |
275
- | 1.6341 | 4850 | 0.0013 | - |
276
- | 1.6509 | 4900 | 0.0006 | - |
277
- | 1.6678 | 4950 | 0.0006 | - |
278
- | 1.6846 | 5000 | 0.0013 | - |
279
- | 1.7015 | 5050 | 0.0006 | - |
280
- | 1.7183 | 5100 | 0.0007 | - |
281
- | 1.7352 | 5150 | 0.0005 | - |
282
- | 1.7520 | 5200 | 0.0005 | - |
283
- | 1.7689 | 5250 | 0.0006 | - |
284
- | 1.7857 | 5300 | 0.0005 | - |
285
- | 1.8026 | 5350 | 0.0005 | - |
286
- | 1.8194 | 5400 | 0.0005 | - |
287
- | 1.8363 | 5450 | 0.0004 | - |
288
- | 1.8531 | 5500 | 0.0066 | - |
289
- | 1.8699 | 5550 | 0.0005 | - |
290
- | 1.8868 | 5600 | 0.0006 | - |
291
- | 1.9036 | 5650 | 0.0005 | - |
292
- | 1.9205 | 5700 | 0.0005 | - |
293
- | 1.9373 | 5750 | 0.0014 | - |
294
- | 1.9542 | 5800 | 0.0006 | - |
295
- | 1.9710 | 5850 | 0.0004 | - |
296
- | 1.9879 | 5900 | 0.0006 | - |
297
- | 2.0047 | 5950 | 0.0005 | - |
298
- | 2.0216 | 6000 | 0.0006 | - |
299
- | 2.0384 | 6050 | 0.0005 | - |
300
- | 2.0553 | 6100 | 0.0004 | - |
301
- | 2.0721 | 6150 | 0.0012 | - |
302
- | 2.0889 | 6200 | 0.0004 | - |
303
- | 2.1058 | 6250 | 0.0005 | - |
304
- | 2.1226 | 6300 | 0.0004 | - |
305
- | 2.1395 | 6350 | 0.0005 | - |
306
- | 2.1563 | 6400 | 0.0005 | - |
307
- | 2.1732 | 6450 | 0.0005 | - |
308
- | 2.1900 | 6500 | 0.0004 | - |
309
- | 2.2069 | 6550 | 0.0004 | - |
310
- | 2.2237 | 6600 | 0.0005 | - |
311
- | 2.2406 | 6650 | 0.0004 | - |
312
- | 2.2574 | 6700 | 0.0005 | - |
313
- | 2.2743 | 6750 | 0.0004 | - |
314
- | 2.2911 | 6800 | 0.0005 | - |
315
- | 2.3080 | 6850 | 0.0007 | - |
316
- | 2.3248 | 6900 | 0.0004 | - |
317
- | 2.3416 | 6950 | 0.0018 | - |
318
- | 2.3585 | 7000 | 0.0004 | - |
319
- | 2.3753 | 7050 | 0.0004 | - |
320
- | 2.3922 | 7100 | 0.0004 | - |
321
- | 2.4090 | 7150 | 0.0004 | - |
322
- | 2.4259 | 7200 | 0.0004 | - |
323
- | 2.4427 | 7250 | 0.0005 | - |
324
- | 2.4596 | 7300 | 0.0004 | - |
325
- | 2.4764 | 7350 | 0.0005 | - |
326
- | 2.4933 | 7400 | 0.0012 | - |
327
- | 2.5101 | 7450 | 0.0026 | - |
328
- | 2.5270 | 7500 | 0.0004 | - |
329
- | 2.5438 | 7550 | 0.0003 | - |
330
- | 2.5606 | 7600 | 0.0004 | - |
331
- | 2.5775 | 7650 | 0.0004 | - |
332
- | 2.5943 | 7700 | 0.0004 | - |
333
- | 2.6112 | 7750 | 0.0004 | - |
334
- | 2.6280 | 7800 | 0.0004 | - |
335
- | 2.6449 | 7850 | 0.0004 | - |
336
- | 2.6617 | 7900 | 0.0004 | - |
337
- | 2.6786 | 7950 | 0.0003 | - |
338
- | 2.6954 | 8000 | 0.0004 | - |
339
- | 2.7123 | 8050 | 0.0004 | - |
340
- | 2.7291 | 8100 | 0.0004 | - |
341
- | 2.7460 | 8150 | 0.0004 | - |
342
- | 2.7628 | 8200 | 0.0004 | - |
343
- | 2.7796 | 8250 | 0.0004 | - |
344
- | 2.7965 | 8300 | 0.0005 | - |
345
- | 2.8133 | 8350 | 0.0004 | - |
346
- | 2.8302 | 8400 | 0.0004 | - |
347
- | 2.8470 | 8450 | 0.0004 | - |
348
- | 2.8639 | 8500 | 0.0004 | - |
349
- | 2.8807 | 8550 | 0.0004 | - |
350
- | 2.8976 | 8600 | 0.0004 | - |
351
- | 2.9144 | 8650 | 0.0004 | - |
352
- | 2.9313 | 8700 | 0.0004 | - |
353
- | 2.9481 | 8750 | 0.0004 | - |
354
- | 2.9650 | 8800 | 0.0004 | - |
355
- | 2.9818 | 8850 | 0.0004 | - |
356
- | 2.9987 | 8900 | 0.0003 | - |
357
-
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
358
  ### Framework Versions
359
  - Python: 3.10.12
360
  - SetFit: 1.0.3
361
- - Sentence Transformers: 2.2.2
362
- - Transformers: 4.35.2
363
  - PyTorch: 2.1.0+cu121
364
- - Datasets: 2.16.1
365
- - Tokenizers: 0.15.1
366
 
367
  ## Citation
368
 
 
6
  - text-classification
7
  - generated_from_setfit_trainer
8
  metrics:
9
+ - accuracy
10
  - precision
11
  - recall
12
+ - f1
13
  widget:
14
+ - text: A Black man, Floyd died in police custody May 25 after a Minneapolis cop kneeled
15
+ on his neck for more than eight minutes.
16
+ - text: 'Now Modi has made international headlines for yet another similarity: He’s
17
+ constructing a massive wall but unlike Trump’s goal of keeping immigrants out,
18
+ Modi’s wall was built to hide the country’s poverty from the gold-plated American
19
+ president.'
20
+ - text: Billionaire Democrat presidential hopeful Mike Bloomberg is a staunch proponent
21
+ of gun control for America with one caveat–he gets to spend his days surrounded
22
+ by good guys with guns to keep him safe.
23
+ - text: The number of women behind the camera on Hollywood movies jumped to record
24
+ levels in 2019, with 12 directing top-grossing films including “Frozen II,” “Captain
25
+ Marvel” and “Hustlers,” two studies showed on Thursday.
26
+ - text: The hearing comes a day after the Democrat-led House held a hearing to discuss
27
+ the alleged threat of white nationalist terrorism to the country.
 
 
28
  pipeline_tag: text-classification
29
  inference: true
30
  base_model: BAAI/bge-small-en-v1.5
 
39
  type: unknown
40
  split: test
41
  metrics:
42
+ - type: accuracy
43
+ value: 0.7010135135135135
44
+ name: Accuracy
45
  - type: precision
46
+ value: 0.7024038067625294
47
  name: Precision
48
  - type: recall
49
+ value: 0.7010135135135135
50
  name: Recall
51
+ - type: f1
52
+ value: 0.7015820127453647
53
+ name: F1
54
  ---
55
 
56
  # SetFit with BAAI/bge-small-en-v1.5
 
81
  - **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
82
 
83
  ### Model Labels
84
+ | Label | Examples |
85
+ |:-------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
86
+ | left | <ul><li>'Tennessee has an annual sales tax-free holiday weekend that\xa0begins\xa0on the last Friday of July.\xa0'</li><li>'In what could be construed as an act of treason,\xa0President Trump recently ordered such\xa0paramilitary groups and right-wing thugs\xa0to take up arms and to threaten Democratic-led state governments such as Michigan\'s in order to force them to "reopen" their state.'</li><li>'Trump, not surprisingly, used the speech as an opportunity to attack former President Barack Obama, claiming that he did nothing to promote criminal justice reform when he was in office.\xa0'</li></ul> |
87
+ | right | <ul><li>'In the Joe Biden-Bernie Sanders “Unity” platform, Democrats are vowing to provide free, American taxpayer-funded health care to illegal aliens who are able to enroll in former President Obama’s Deferred Action for Childhood Arrivals (DACA) program.'</li><li>'The new numbers from Gallup are an unwelcome sight for Democrats after kicking off the week with a disaster caucus in Iowa who and simultaneously anticipating a Trump acquittal in the Senate. Trump will also now have the opportunity to shine in his newfound approval in Tuesday night’s address to the nation while Democrats are in disarray.'</li><li>'Though Trump has successfully increased wages by four percent over the last 12 months for America’s blue collar and working class by decreasing foreign competition through a crackdown on illegal immigration, experts have warned that those wage hikes will not continue heading into the 2020 election should current illegal immigration levels keep rising at record levels.'</li></ul> |
88
+ | center | <ul><li>'LeBron James shares thoughts on his Los Angeles house getting vandalized pic twitter com 4RFLK42xhu'</li><li>'O’Rourke, a native of the U.S.-Mexican border town El Paso, has blasted Trump’s use of tariffs as a “huge mistake” and has vowed to suspend them on his first day in office.'</li><li>'Here are a few people we will be reminding you about in every article that pertains to a film they re tied to '</li></ul> |
89
 
90
  ## Evaluation
91
 
92
  ### Metrics
93
+ | Label | Accuracy | Precision | Recall | F1 |
94
+ |:--------|:---------|:----------|:-------|:-------|
95
+ | **all** | 0.7010 | 0.7024 | 0.7010 | 0.7016 |
96
 
97
  ## Uses
98
 
 
112
  # Download from the 🤗 Hub
113
  model = SetFitModel.from_pretrained("JordanTallon/Unifeed")
114
  # Run inference
115
+ preds = model("A Black man, Floyd died in police custody May 25 after a Minneapolis cop kneeled on his neck for more than eight minutes.")
116
  ```
117
 
118
  <!--
 
144
  ### Training Set Metrics
145
  | Training set | Min | Median | Max |
146
  |:-------------|:----|:--------|:----|
147
+ | Word count | 9 | 32.9560 | 90 |
148
 
149
  | Label | Training Sample Count |
150
  |:-------|:----------------------|
151
+ | center | 777 |
152
+ | left | 780 |
153
+ | right | 808 |
154
 
155
  ### Training Hyperparameters
156
  - batch_size: (32, 32)
157
+ - num_epochs: (200, 200)
158
  - max_steps: -1
159
  - sampling_strategy: oversampling
160
+ - num_iterations: 1
161
+ - body_learning_rate: (2e-05, 1e-05)
162
+ - head_learning_rate: 0.01
163
  - loss: CosineSimilarityLoss
164
  - distance_metric: cosine_distance
165
  - margin: 0.25
166
  - end_to_end: False
167
+ - use_amp: True
168
  - warmup_proportion: 0.1
169
+ - seed: 326
170
+ - run_name: unifeed_bias_training
171
  - eval_max_steps: -1
172
+ - load_best_model_at_end: True
173
 
174
  ### Training Results
175
+ | Epoch | Step | Training Loss | Validation Loss |
176
+ |:--------:|:--------:|:-------------:|:---------------:|
177
+ | 0.0002 | 1 | 0.2486 | - |
178
+ | 1.0 | 4878 | 0.0092 | 0.308 |
179
+ | 2.0 | 9756 | 0.0004 | 0.3228 |
180
+ | 3.0 | 14634 | 0.0002 | 0.3326 |
181
+ | 4.0 | 19512 | 0.0002 | 0.3191 |
182
+ | 5.0 | 24390 | 0.0001 | 0.3279 |
183
+ | 6.0 | 29268 | 0.0001 | 0.3384 |
184
+ | 7.0 | 34146 | 0.0001 | 0.3311 |
185
+ | 8.0 | 39024 | 0.0001 | 0.3316 |
186
+ | 0.0068 | 1 | 0.0007 | - |
187
+ | 1.0 | 148 | 0.0006 | 0.3042 |
188
+ | 2.0 | 296 | 0.0006 | 0.3352 |
189
+ | 3.0 | 444 | 0.0382 | 0.3059 |
190
+ | 4.0 | 592 | 0.0022 | 0.3055 |
191
+ | 5.0 | 740 | 0.0044 | 0.3034 |
192
+ | 6.0 | 888 | 0.0006 | 0.3185 |
193
+ | 7.0 | 1036 | 0.0005 | 0.3066 |
194
+ | 8.0 | 1184 | 0.0008 | 0.3196 |
195
+ | 9.0 | 1332 | 0.0004 | 0.326 |
196
+ | 10.0 | 1480 | 0.0004 | 0.352 |
197
+ | 11.0 | 1628 | 0.0005 | 0.3122 |
198
+ | 12.0 | 1776 | 0.0003 | 0.3268 |
199
+ | 13.0 | 1924 | 0.0004 | 0.2928 |
200
+ | 14.0 | 2072 | 0.0004 | 0.3148 |
201
+ | 15.0 | 2220 | 0.0003 | 0.3153 |
202
+ | 16.0 | 2368 | 0.0004 | 0.3385 |
203
+ | 17.0 | 2516 | 0.0004 | 0.3107 |
204
+ | 18.0 | 2664 | 0.0004 | 0.3225 |
205
+ | 19.0 | 2812 | 0.0003 | 0.3073 |
206
+ | 20.0 | 2960 | 0.0003 | 0.316 |
207
+ | 21.0 | 3108 | 0.0003 | 0.3053 |
208
+ | 22.0 | 3256 | 0.0004 | 0.3227 |
209
+ | 23.0 | 3404 | 0.0004 | 0.3099 |
210
+ | 24.0 | 3552 | 0.0003 | 0.3043 |
211
+ | 25.0 | 3700 | 0.0003 | 0.3316 |
212
+ | 0.0034 | 1 | 0.0004 | - |
213
+ | 1.0 | 296 | 0.0003 | 0.3321 |
214
+ | 2.0 | 592 | 0.0016 | 0.3202 |
215
+ | 3.0 | 888 | 0.0005 | 0.3376 |
216
+ | 4.0 | 1184 | 0.0004 | 0.3167 |
217
+ | 5.0 | 1480 | 0.0003 | 0.3342 |
218
+ | 6.0 | 1776 | 0.0003 | 0.3183 |
219
+ | 7.0 | 2072 | 0.0003 | 0.3086 |
220
+ | 8.0 | 2368 | 0.0003 | 0.312 |
221
+ | 9.0 | 2664 | 0.0003 | 0.3169 |
222
+ | 10.0 | 2960 | 0.0003 | 0.3317 |
223
+ | 11.0 | 3256 | 0.0004 | 0.3126 |
224
+ | 12.0 | 3552 | 0.0003 | 0.3003 |
225
+ | 13.0 | 3848 | 0.0003 | 0.3119 |
226
+ | 14.0 | 4144 | 0.0003 | 0.316 |
227
+ | 15.0 | 4440 | 0.0002 | 0.3183 |
228
+ | 16.0 | 4736 | 0.0003 | 0.313 |
229
+ | 17.0 | 5032 | 0.0003 | 0.3187 |
230
+ | 18.0 | 5328 | 0.0002 | 0.3295 |
231
+ | 19.0 | 5624 | 0.0002 | 0.3487 |
232
+ | 20.0 | 5920 | 0.0003 | 0.3458 |
233
+ | 21.0 | 6216 | 0.0002 | 0.331 |
234
+ | 22.0 | 6512 | 0.0002 | 0.3499 |
235
+ | 23.0 | 6808 | 0.0003 | 0.3296 |
236
+ | 24.0 | 7104 | 0.0003 | 0.3097 |
237
+ | 25.0 | 7400 | 0.0003 | 0.3197 |
238
+ | 0.0068 | 1 | 0.0003 | - |
239
+ | 1.0 | 148 | 0.0003 | 0.3219 |
240
+ | 2.0 | 296 | 0.0003 | 0.3185 |
241
+ | 3.0 | 444 | 0.0003 | 0.3114 |
242
+ | 4.0 | 592 | 0.0003 | 0.2989 |
243
+ | 5.0 | 740 | 0.0003 | 0.335 |
244
+ | 6.0 | 888 | 0.0004 | 0.3132 |
245
+ | 7.0 | 1036 | 0.0003 | 0.3264 |
246
+ | 8.0 | 1184 | 0.0004 | 0.3461 |
247
+ | 9.0 | 1332 | 0.0002 | 0.3185 |
248
+ | 10.0 | 1480 | 0.0002 | 0.3336 |
249
+ | 11.0 | 1628 | 0.0003 | 0.3282 |
250
+ | 12.0 | 1776 | 0.0003 | 0.3206 |
251
+ | 13.0 | 1924 | 0.0002 | 0.3303 |
252
+ | 14.0 | 2072 | 0.0002 | 0.3362 |
253
+ | 15.0 | 2220 | 0.0002 | 0.3382 |
254
+ | 16.0 | 2368 | 0.0002 | 0.3241 |
255
+ | 17.0 | 2516 | 0.0002 | 0.3303 |
256
+ | 18.0 | 2664 | 0.0002 | 0.3301 |
257
+ | 19.0 | 2812 | 0.0002 | 0.319 |
258
+ | 20.0 | 2960 | 0.0002 | 0.3304 |
259
+ | 21.0 | 3108 | 0.0002 | 0.3379 |
260
+ | 22.0 | 3256 | 0.0002 | 0.3424 |
261
+ | 23.0 | 3404 | 0.0002 | 0.3273 |
262
+ | 24.0 | 3552 | 0.0002 | 0.3213 |
263
+ | 25.0 | 3700 | 0.0002 | 0.3191 |
264
+ | 0.0068 | 1 | 0.0003 | - |
265
+ | 1.0 | 148 | 0.0003 | 0.3245 |
266
+ | 2.0 | 296 | 0.0002 | 0.3148 |
267
+ | 3.0 | 444 | 0.0002 | 0.3174 |
268
+ | 4.0 | 592 | 0.0003 | 0.3242 |
269
+ | 5.0 | 740 | 0.0003 | 0.3352 |
270
+ | 6.0 | 888 | 0.0003 | 0.3112 |
271
+ | 7.0 | 1036 | 0.0003 | 0.3204 |
272
+ | 8.0 | 1184 | 0.0003 | 0.3734 |
273
+ | 9.0 | 1332 | 0.0002 | 0.3383 |
274
+ | 10.0 | 1480 | 0.0003 | 0.3272 |
275
+ | 11.0 | 1628 | 0.0002 | 0.3106 |
276
+ | 12.0 | 1776 | 0.0003 | 0.3307 |
277
+ | 13.0 | 1924 | 0.0003 | 0.3359 |
278
+ | 14.0 | 2072 | 0.0002 | 0.3264 |
279
+ | 15.0 | 2220 | 0.0002 | 0.3254 |
280
+ | 16.0 | 2368 | 0.0002 | 0.3349 |
281
+ | 17.0 | 2516 | 0.0132 | 0.3399 |
282
+ | 18.0 | 2664 | 0.0002 | 0.343 |
283
+ | 19.0 | 2812 | 0.0002 | 0.3306 |
284
+ | 20.0 | 2960 | 0.0002 | 0.3472 |
285
+ | 21.0 | 3108 | 0.0002 | 0.3234 |
286
+ | 22.0 | 3256 | 0.002 | 0.3281 |
287
+ | 23.0 | 3404 | 0.0002 | 0.3289 |
288
+ | 24.0 | 3552 | 0.0002 | 0.2974 |
289
+ | 25.0 | 3700 | 0.0002 | 0.3153 |
290
+ | 26.0 | 3848 | 0.0002 | 0.3273 |
291
+ | 27.0 | 3996 | 0.0002 | 0.313 |
292
+ | 28.0 | 4144 | 0.0002 | 0.3303 |
293
+ | 29.0 | 4292 | 0.0002 | 0.3106 |
294
+ | 30.0 | 4440 | 0.0002 | 0.3155 |
295
+ | 31.0 | 4588 | 0.0002 | 0.3553 |
296
+ | 32.0 | 4736 | 0.0002 | 0.3039 |
297
+ | 33.0 | 4884 | 0.0001 | 0.3133 |
298
+ | 34.0 | 5032 | 0.0002 | 0.3323 |
299
+ | 35.0 | 5180 | 0.0002 | 0.3264 |
300
+ | 36.0 | 5328 | 0.0002 | 0.3133 |
301
+ | 37.0 | 5476 | 0.0002 | 0.3308 |
302
+ | 38.0 | 5624 | 0.0002 | 0.3137 |
303
+ | 39.0 | 5772 | 0.0002 | 0.3062 |
304
+ | 40.0 | 5920 | 0.0002 | 0.3438 |
305
+ | 41.0 | 6068 | 0.0002 | 0.3426 |
306
+ | 42.0 | 6216 | 0.0002 | 0.326 |
307
+ | 43.0 | 6364 | 0.0002 | 0.322 |
308
+ | 44.0 | 6512 | 0.0002 | 0.3202 |
309
+ | 45.0 | 6660 | 0.0002 | 0.3253 |
310
+ | 46.0 | 6808 | 0.0002 | 0.3272 |
311
+ | 47.0 | 6956 | 0.0002 | 0.3258 |
312
+ | 48.0 | 7104 | 0.0002 | 0.3252 |
313
+ | 49.0 | 7252 | 0.0002 | 0.3233 |
314
+ | 50.0 | 7400 | 0.0002 | 0.3234 |
315
+ | 0.0135 | 1 | 0.0002 | - |
316
+ | 1.0 | 74 | 0.0002 | - |
317
+ | 0.0068 | 1 | 0.0002 | - |
318
+ | 1.0 | 148 | 0.0002 | 0.3036 |
319
+ | 2.0 | 296 | 0.0002 | 0.3555 |
320
+ | 3.0 | 444 | 0.0002 | 0.3331 |
321
+ | 4.0 | 592 | 0.0002 | 0.3086 |
322
+ | 5.0 | 740 | 0.0002 | 0.3036 |
323
+ | 6.0 | 888 | 0.0002 | 0.3217 |
324
+ | 7.0 | 1036 | 0.0002 | 0.3416 |
325
+ | 8.0 | 1184 | 0.0002 | 0.3309 |
326
+ | 9.0 | 1332 | 0.0002 | 0.3424 |
327
+ | 10.0 | 1480 | 0.0003 | 0.3655 |
328
+ | 11.0 | 1628 | 0.0002 | 0.3042 |
329
+ | 12.0 | 1776 | 0.0019 | 0.326 |
330
+ | 13.0 | 1924 | 0.0002 | 0.3161 |
331
+ | 14.0 | 2072 | 0.0002 | 0.3286 |
332
+ | 15.0 | 2220 | 0.0002 | 0.3563 |
333
+ | 16.0 | 2368 | 0.0002 | 0.326 |
334
+ | 17.0 | 2516 | 0.0002 | 0.3114 |
335
+ | 18.0 | 2664 | 0.0002 | 0.3366 |
336
+ | 19.0 | 2812 | 0.0002 | 0.329 |
337
+ | 20.0 | 2960 | 0.0002 | 0.3217 |
338
+ | 21.0 | 3108 | 0.0002 | 0.325 |
339
+ | 22.0 | 3256 | 0.0002 | 0.3243 |
340
+ | 23.0 | 3404 | 0.0002 | 0.3341 |
341
+ | 24.0 | 3552 | 0.0002 | 0.3237 |
342
+ | 25.0 | 3700 | 0.0002 | 0.3433 |
343
+ | 26.0 | 3848 | 0.0002 | 0.3196 |
344
+ | 27.0 | 3996 | 0.0001 | 0.3372 |
345
+ | 28.0 | 4144 | 0.0001 | 0.3191 |
346
+ | 29.0 | 4292 | 0.0001 | 0.328 |
347
+ | 30.0 | 4440 | 0.0002 | 0.3416 |
348
+ | 31.0 | 4588 | 0.0002 | 0.3132 |
349
+ | 32.0 | 4736 | 0.0002 | 0.3429 |
350
+ | 33.0 | 4884 | 0.0002 | 0.336 |
351
+ | 34.0 | 5032 | 0.0002 | 0.3507 |
352
+ | 35.0 | 5180 | 0.0001 | 0.3483 |
353
+ | 36.0 | 5328 | 0.0002 | 0.3325 |
354
+ | 37.0 | 5476 | 0.0001 | 0.3406 |
355
+ | 38.0 | 5624 | 0.0003 | 0.3538 |
356
+ | 39.0 | 5772 | 0.0002 | 0.3422 |
357
+ | 40.0 | 5920 | 0.0002 | 0.3359 |
358
+ | 41.0 | 6068 | 0.0002 | 0.3252 |
359
+ | 42.0 | 6216 | 0.0002 | 0.326 |
360
+ | 43.0 | 6364 | 0.0002 | 0.3613 |
361
+ | 44.0 | 6512 | 0.0001 | 0.332 |
362
+ | 45.0 | 6660 | 0.0002 | 0.3295 |
363
+ | 46.0 | 6808 | 0.0002 | 0.3265 |
364
+ | 47.0 | 6956 | 0.0002 | 0.2982 |
365
+ | 48.0 | 7104 | 0.0002 | 0.3017 |
366
+ | 49.0 | 7252 | 0.0001 | 0.309 |
367
+ | 50.0 | 7400 | 0.0001 | 0.3199 |
368
+ | 51.0 | 7548 | 0.0001 | 0.325 |
369
+ | 52.0 | 7696 | 0.0002 | 0.3222 |
370
+ | 53.0 | 7844 | 0.0001 | 0.3189 |
371
+ | 54.0 | 7992 | 0.0001 | 0.3329 |
372
+ | 55.0 | 8140 | 0.0001 | 0.3272 |
373
+ | 56.0 | 8288 | 0.0001 | 0.3292 |
374
+ | 57.0 | 8436 | 0.0001 | 0.3283 |
375
+ | 58.0 | 8584 | 0.0001 | 0.3301 |
376
+ | 59.0 | 8732 | 0.0001 | 0.3334 |
377
+ | 60.0 | 8880 | 0.0001 | 0.3144 |
378
+ | 61.0 | 9028 | 0.0002 | 0.3487 |
379
+ | 62.0 | 9176 | 0.0002 | 0.3602 |
380
+ | **63.0** | **9324** | **0.0001** | **0.3056** |
381
+ | 64.0 | 9472 | 0.0001 | 0.3415 |
382
+ | 65.0 | 9620 | 0.0002 | 0.3299 |
383
+ | 66.0 | 9768 | 0.0001 | 0.3254 |
384
+ | 67.0 | 9916 | 0.0001 | 0.3396 |
385
+ | 68.0 | 10064 | 0.0001 | 0.3501 |
386
+ | 69.0 | 10212 | 0.0001 | 0.3275 |
387
+ | 70.0 | 10360 | 0.0001 | 0.34 |
388
+ | 71.0 | 10508 | 0.0001 | 0.3351 |
389
+ | 72.0 | 10656 | 0.0001 | 0.3367 |
390
+ | 73.0 | 10804 | 0.0001 | 0.3548 |
391
+ | 74.0 | 10952 | 0.0001 | 0.33 |
392
+ | 75.0 | 11100 | 0.0001 | 0.3259 |
393
+ | 76.0 | 11248 | 0.0002 | 0.3283 |
394
+ | 77.0 | 11396 | 0.0001 | 0.3214 |
395
+ | 78.0 | 11544 | 0.0001 | 0.324 |
396
+ | 79.0 | 11692 | 0.0001 | 0.3247 |
397
+ | 80.0 | 11840 | 0.0001 | 0.3347 |
398
+ | 81.0 | 11988 | 0.0001 | 0.3292 |
399
+ | 82.0 | 12136 | 0.0002 | 0.3568 |
400
+ | 83.0 | 12284 | 0.0001 | 0.324 |
401
+ | 84.0 | 12432 | 0.0001 | 0.3245 |
402
+ | 85.0 | 12580 | 0.0001 | 0.3368 |
403
+ | 86.0 | 12728 | 0.0001 | 0.3372 |
404
+ | 87.0 | 12876 | 0.0001 | 0.3432 |
405
+ | 88.0 | 13024 | 0.0001 | 0.3048 |
406
+ | 89.0 | 13172 | 0.0001 | 0.3395 |
407
+ | 90.0 | 13320 | 0.0001 | 0.3204 |
408
+ | 91.0 | 13468 | 0.0001 | 0.3122 |
409
+ | 92.0 | 13616 | 0.0001 | 0.3372 |
410
+ | 93.0 | 13764 | 0.0001 | 0.3306 |
411
+ | 94.0 | 13912 | 0.0001 | 0.3362 |
412
+ | 95.0 | 14060 | 0.0001 | 0.3386 |
413
+ | 96.0 | 14208 | 0.0001 | 0.3198 |
414
+ | 97.0 | 14356 | 0.0001 | 0.3176 |
415
+ | 98.0 | 14504 | 0.0001 | 0.3604 |
416
+ | 99.0 | 14652 | 0.0001 | 0.3507 |
417
+ | 100.0 | 14800 | 0.0001 | 0.3272 |
418
+ | 0.0023 | 1 | 0.0001 | - |
419
+ | 1.0 | 444 | 0.0002 | 0.3295 |
420
+ | 2.0 | 888 | 0.0001 | 0.3144 |
421
+ | 3.0 | 1332 | 0.0001 | 0.3213 |
422
+ | 4.0 | 1776 | 0.0001 | 0.3362 |
423
+ | 5.0 | 2220 | 0.0001 | 0.3398 |
424
+ | 6.0 | 2664 | 0.0001 | 0.3385 |
425
+ | 7.0 | 3108 | 0.0002 | 0.3406 |
426
+ | 8.0 | 3552 | 0.0001 | 0.3253 |
427
+ | 9.0 | 3996 | 0.0001 | 0.3253 |
428
+ | 10.0 | 4440 | 0.0001 | 0.3119 |
429
+ | 11.0 | 4884 | 0.0001 | 0.3204 |
430
+ | 12.0 | 5328 | 0.0001 | 0.3387 |
431
+ | 13.0 | 5772 | 0.0001 | 0.3387 |
432
+ | 14.0 | 6216 | 0.0001 | 0.3584 |
433
+ | 15.0 | 6660 | 0.0001 | 0.3548 |
434
+ | 16.0 | 7104 | 0.0001 | 0.3314 |
435
+ | 17.0 | 7548 | 0.0001 | 0.3335 |
436
+ | 18.0 | 7992 | 0.0001 | 0.3325 |
437
+ | 19.0 | 8436 | 0.0001 | 0.3545 |
438
+ | 20.0 | 8880 | 0.0001 | 0.3456 |
439
+ | **21.0** | **9324** | **0.0001** | **0.3532** |
440
+ | 22.0 | 9768 | 0.0001 | 0.3524 |
441
+ | 23.0 | 10212 | 0.0001 | 0.352 |
442
+ | 24.0 | 10656 | 0.0001 | 0.3502 |
443
+ | 25.0 | 11100 | 0.0 | 0.3275 |
444
+ | 0.0034 | 1 | 0.0001 | - |
445
+ | 1.0 | 296 | 0.0001 | 0.3209 |
446
+ | 2.0 | 592 | 0.0001 | 0.3265 |
447
+ | 3.0 | 888 | 0.0001 | 0.3414 |
448
+ | 4.0 | 1184 | 0.0001 | 0.3314 |
449
+ | 5.0 | 1480 | 0.0002 | 0.3498 |
450
+ | 6.0 | 1776 | 0.0001 | 0.337 |
451
+ | 7.0 | 2072 | 0.0001 | 0.3347 |
452
+ | 8.0 | 2368 | 0.0001 | 0.3494 |
453
+ | 9.0 | 2664 | 0.0001 | 0.3326 |
454
+ | 10.0 | 2960 | 0.0001 | 0.3259 |
455
+ | 11.0 | 3256 | 0.0002 | 0.3443 |
456
+ | 12.0 | 3552 | 0.0001 | 0.3431 |
457
+ | 13.0 | 3848 | 0.0001 | 0.324 |
458
+ | 14.0 | 4144 | 0.0001 | 0.3339 |
459
+ | 15.0 | 4440 | 0.0001 | 0.3255 |
460
+ | 16.0 | 4736 | 0.0001 | 0.3379 |
461
+ | 17.0 | 5032 | 0.0001 | 0.3285 |
462
+ | 18.0 | 5328 | 0.0001 | 0.3362 |
463
+ | 19.0 | 5624 | 0.0001 | 0.3319 |
464
+ | 20.0 | 5920 | 0.0001 | 0.3456 |
465
+ | 21.0 | 6216 | 0.0001 | 0.329 |
466
+ | 22.0 | 6512 | 0.0001 | 0.3386 |
467
+ | 23.0 | 6808 | 0.0001 | 0.3278 |
468
+ | 24.0 | 7104 | 0.0001 | 0.3078 |
469
+ | 25.0 | 7400 | 0.0001 | 0.3155 |
470
+ | 0.0068 | 1 | 0.0001 | - |
471
+ | 1.0 | 148 | 0.0001 | 0.3225 |
472
+ | 2.0 | 296 | 0.0001 | 0.3526 |
473
+ | 3.0 | 444 | 0.0001 | 0.3265 |
474
+ | 4.0 | 592 | 0.0001 | 0.3206 |
475
+ | 5.0 | 740 | 0.0001 | 0.3126 |
476
+ | 6.0 | 888 | 0.0001 | 0.3306 |
477
+ | 7.0 | 1036 | 0.0001 | 0.3189 |
478
+ | 8.0 | 1184 | 0.0001 | 0.3246 |
479
+ | 9.0 | 1332 | 0.0001 | 0.3346 |
480
+ | 10.0 | 1480 | 0.0001 | 0.3528 |
481
+ | 11.0 | 1628 | 0.0001 | 0.3204 |
482
+ | 12.0 | 1776 | 0.0001 | 0.34 |
483
+ | 13.0 | 1924 | 0.0001 | 0.3291 |
484
+ | 14.0 | 2072 | 0.0001 | 0.3444 |
485
+ | 15.0 | 2220 | 0.0001 | 0.339 |
486
+ | 16.0 | 2368 | 0.0001 | 0.3533 |
487
+ | 17.0 | 2516 | 0.0001 | 0.3288 |
488
+ | 18.0 | 2664 | 0.0001 | 0.3475 |
489
+ | 19.0 | 2812 | 0.0001 | 0.3464 |
490
+ | 20.0 | 2960 | 0.0001 | 0.3351 |
491
+ | 21.0 | 3108 | 0.0001 | 0.3421 |
492
+ | 22.0 | 3256 | 0.0001 | 0.3351 |
493
+ | 23.0 | 3404 | 0.0001 | 0.3416 |
494
+ | 24.0 | 3552 | 0.0001 | 0.3414 |
495
+ | 25.0 | 3700 | 0.0001 | 0.3433 |
496
+ | 26.0 | 3848 | 0.0001 | 0.3339 |
497
+ | 27.0 | 3996 | 0.0001 | 0.35 |
498
+ | 28.0 | 4144 | 0.0001 | 0.3215 |
499
+ | 29.0 | 4292 | 0.0001 | 0.3278 |
500
+ | 30.0 | 4440 | 0.0001 | 0.3508 |
501
+ | 31.0 | 4588 | 0.0001 | 0.3356 |
502
+ | 32.0 | 4736 | 0.0001 | 0.3617 |
503
+ | 33.0 | 4884 | 0.0001 | 0.3368 |
504
+ | 34.0 | 5032 | 0.0001 | 0.3551 |
505
+ | 35.0 | 5180 | 0.0001 | 0.3582 |
506
+ | 36.0 | 5328 | 0.0001 | 0.333 |
507
+ | 37.0 | 5476 | 0.0 | 0.3461 |
508
+ | 38.0 | 5624 | 0.0001 | 0.3515 |
509
+ | 39.0 | 5772 | 0.0001 | 0.3601 |
510
+ | 40.0 | 5920 | 0.0001 | 0.347 |
511
+ | 41.0 | 6068 | 0.0001 | 0.3444 |
512
+ | 42.0 | 6216 | 0.0 | 0.3609 |
513
+ | 43.0 | 6364 | 0.0 | 0.3432 |
514
+ | 44.0 | 6512 | 0.0 | 0.3526 |
515
+ | 45.0 | 6660 | 0.0 | 0.3382 |
516
+ | 46.0 | 6808 | 0.0 | 0.353 |
517
+ | 47.0 | 6956 | 0.0001 | 0.3374 |
518
+ | 48.0 | 7104 | 0.0001 | 0.327 |
519
+ | 49.0 | 7252 | 0.0001 | 0.3202 |
520
+ | 50.0 | 7400 | 0.0 | 0.3386 |
521
+ | 51.0 | 7548 | 0.0001 | 0.3501 |
522
+ | 52.0 | 7696 | 0.0002 | 0.3341 |
523
+ | 53.0 | 7844 | 0.0001 | 0.3024 |
524
+ | 54.0 | 7992 | 0.0001 | 0.3456 |
525
+ | 55.0 | 8140 | 0.0 | 0.3323 |
526
+ | 56.0 | 8288 | 0.0 | 0.3259 |
527
+ | 57.0 | 8436 | 0.0 | 0.3246 |
528
+ | 58.0 | 8584 | 0.0 | 0.3341 |
529
+ | 59.0 | 8732 | 0.0 | 0.3347 |
530
+ | 60.0 | 8880 | 0.0 | 0.322 |
531
+ | 61.0 | 9028 | 0.0001 | 0.3323 |
532
+ | 62.0 | 9176 | 0.0 | 0.3471 |
533
+ | **63.0** | **9324** | **0.0001** | **0.2913** |
534
+ | 64.0 | 9472 | 0.0 | 0.3144 |
535
+ | 65.0 | 9620 | 0.0001 | 0.3184 |
536
+ | 66.0 | 9768 | 0.0 | 0.3251 |
537
+ | 67.0 | 9916 | 0.0001 | 0.3342 |
538
+ | 68.0 | 10064 | 0.0 | 0.3486 |
539
+ | 69.0 | 10212 | 0.0 | 0.3381 |
540
+ | 70.0 | 10360 | 0.0 | 0.3161 |
541
+ | 71.0 | 10508 | 0.0 | 0.3036 |
542
+ | 72.0 | 10656 | 0.0 | 0.3141 |
543
+ | 73.0 | 10804 | 0.0 | 0.3307 |
544
+ | 74.0 | 10952 | 0.0 | 0.3153 |
545
+ | 75.0 | 11100 | 0.0 | 0.3016 |
546
+ | 76.0 | 11248 | 0.0001 | 0.3321 |
547
+ | 77.0 | 11396 | 0.0001 | 0.3194 |
548
+ | 78.0 | 11544 | 0.0001 | 0.3496 |
549
+ | 79.0 | 11692 | 0.0 | 0.3218 |
550
+ | 80.0 | 11840 | 0.0 | 0.3251 |
551
+ | 81.0 | 11988 | 0.0 | 0.3468 |
552
+ | 82.0 | 12136 | 0.0 | 0.3803 |
553
+ | 83.0 | 12284 | 0.0 | 0.3354 |
554
+ | 84.0 | 12432 | 0.0 | 0.351 |
555
+ | 85.0 | 12580 | 0.0 | 0.3231 |
556
+ | 86.0 | 12728 | 0.0 | 0.3027 |
557
+ | 87.0 | 12876 | 0.0 | 0.3309 |
558
+ | 88.0 | 13024 | 0.0 | 0.3194 |
559
+ | 89.0 | 13172 | 0.0 | 0.3611 |
560
+ | 90.0 | 13320 | 0.0 | 0.3288 |
561
+ | 91.0 | 13468 | 0.0 | 0.3261 |
562
+ | 92.0 | 13616 | 0.0 | 0.3268 |
563
+ | 93.0 | 13764 | 0.0 | 0.3433 |
564
+ | 94.0 | 13912 | 0.0 | 0.3438 |
565
+ | 95.0 | 14060 | 0.0 | 0.3288 |
566
+ | 96.0 | 14208 | 0.0 | 0.3263 |
567
+ | 97.0 | 14356 | 0.0 | 0.3331 |
568
+ | 98.0 | 14504 | 0.0 | 0.3334 |
569
+ | 99.0 | 14652 | 0.0 | 0.319 |
570
+ | 100.0 | 14800 | 0.0 | 0.3033 |
571
+ | 101.0 | 14948 | 0.0001 | 0.3051 |
572
+ | 102.0 | 15096 | 0.0 | 0.3321 |
573
+ | 103.0 | 15244 | 0.0 | 0.3181 |
574
+ | 104.0 | 15392 | 0.0 | 0.2943 |
575
+ | 105.0 | 15540 | 0.0 | 0.3137 |
576
+ | 106.0 | 15688 | 0.0 | 0.3111 |
577
+ | 107.0 | 15836 | 0.0 | 0.2968 |
578
+ | 108.0 | 15984 | 0.0 | 0.3072 |
579
+ | 109.0 | 16132 | 0.0 | 0.3154 |
580
+ | 110.0 | 16280 | 0.0001 | 0.3211 |
581
+ | 111.0 | 16428 | 0.0 | 0.2974 |
582
+ | 112.0 | 16576 | 0.0 | 0.3057 |
583
+ | 113.0 | 16724 | 0.0 | 0.296 |
584
+ | 114.0 | 16872 | 0.0 | 0.3104 |
585
+ | 115.0 | 17020 | 0.0 | 0.3029 |
586
+ | 116.0 | 17168 | 0.0 | 0.329 |
587
+ | 117.0 | 17316 | 0.0 | 0.3275 |
588
+ | 118.0 | 17464 | 0.0 | 0.3343 |
589
+ | 119.0 | 17612 | 0.0 | 0.3168 |
590
+ | 120.0 | 17760 | 0.0 | 0.3208 |
591
+ | 121.0 | 17908 | 0.0 | 0.2973 |
592
+ | 122.0 | 18056 | 0.0 | 0.3121 |
593
+ | 123.0 | 18204 | 0.0 | 0.3049 |
594
+ | 124.0 | 18352 | 0.0 | 0.3079 |
595
+ | 125.0 | 18500 | 0.0 | 0.2994 |
596
+ | 126.0 | 18648 | 0.0 | 0.3189 |
597
+ | 127.0 | 18796 | 0.0 | 0.3255 |
598
+ | 128.0 | 18944 | 0.0 | 0.3111 |
599
+ | 129.0 | 19092 | 0.0 | 0.3182 |
600
+ | 130.0 | 19240 | 0.0 | 0.356 |
601
+ | 131.0 | 19388 | 0.0 | 0.3299 |
602
+ | 132.0 | 19536 | 0.0 | 0.3308 |
603
+ | 133.0 | 19684 | 0.0 | 0.3379 |
604
+ | 134.0 | 19832 | 0.0 | 0.3233 |
605
+ | 135.0 | 19980 | 0.0 | 0.327 |
606
+ | 136.0 | 20128 | 0.0 | 0.318 |
607
+ | 137.0 | 20276 | 0.0 | 0.2937 |
608
+ | 138.0 | 20424 | 0.0 | 0.3039 |
609
+ | 139.0 | 20572 | 0.0 | 0.3367 |
610
+ | 140.0 | 20720 | 0.0 | 0.3185 |
611
+ | 141.0 | 20868 | 0.0 | 0.3441 |
612
+ | 142.0 | 21016 | 0.0 | 0.3055 |
613
+ | 143.0 | 21164 | 0.0 | 0.3202 |
614
+ | 144.0 | 21312 | 0.0 | 0.3144 |
615
+ | 145.0 | 21460 | 0.0 | 0.3304 |
616
+ | 146.0 | 21608 | 0.0 | 0.3165 |
617
+ | 147.0 | 21756 | 0.0 | 0.309 |
618
+ | 148.0 | 21904 | 0.0 | 0.3086 |
619
+ | 149.0 | 22052 | 0.0 | 0.2987 |
620
+ | 150.0 | 22200 | 0.0 | 0.3198 |
621
+ | 151.0 | 22348 | 0.0 | 0.3372 |
622
+ | 152.0 | 22496 | 0.0 | 0.3156 |
623
+ | 153.0 | 22644 | 0.0 | 0.3206 |
624
+ | 154.0 | 22792 | 0.0 | 0.322 |
625
+ | 155.0 | 22940 | 0.0 | 0.3445 |
626
+ | 156.0 | 23088 | 0.0 | 0.3183 |
627
+ | 157.0 | 23236 | 0.0 | 0.3203 |
628
+ | 158.0 | 23384 | 0.0 | 0.3337 |
629
+ | 159.0 | 23532 | 0.0 | 0.3245 |
630
+ | 160.0 | 23680 | 0.0 | 0.3068 |
631
+ | 161.0 | 23828 | 0.0 | 0.3199 |
632
+ | 162.0 | 23976 | 0.0 | 0.3308 |
633
+ | 163.0 | 24124 | 0.0 | 0.3446 |
634
+ | 164.0 | 24272 | 0.0 | 0.341 |
635
+ | 165.0 | 24420 | 0.0 | 0.3155 |
636
+ | 166.0 | 24568 | 0.0 | 0.3306 |
637
+ | 167.0 | 24716 | 0.0 | 0.3422 |
638
+ | 168.0 | 24864 | 0.0 | 0.336 |
639
+ | 169.0 | 25012 | 0.0 | 0.3271 |
640
+ | 170.0 | 25160 | 0.0 | 0.3062 |
641
+ | 171.0 | 25308 | 0.0 | 0.305 |
642
+ | 172.0 | 25456 | 0.0 | 0.3047 |
643
+ | 173.0 | 25604 | 0.0 | 0.3281 |
644
+ | 174.0 | 25752 | 0.0 | 0.3059 |
645
+ | 175.0 | 25900 | 0.0 | 0.2993 |
646
+ | 176.0 | 26048 | 0.0 | 0.3206 |
647
+ | 177.0 | 26196 | 0.0 | 0.3274 |
648
+ | 178.0 | 26344 | 0.0 | 0.3249 |
649
+ | 179.0 | 26492 | 0.0 | 0.3049 |
650
+ | 180.0 | 26640 | 0.0 | 0.3131 |
651
+ | 181.0 | 26788 | 0.0 | 0.3119 |
652
+ | 182.0 | 26936 | 0.0 | 0.3457 |
653
+ | 183.0 | 27084 | 0.0 | 0.3242 |
654
+ | 184.0 | 27232 | 0.0 | 0.3006 |
655
+ | 185.0 | 27380 | 0.0 | 0.3054 |
656
+ | 186.0 | 27528 | 0.0 | 0.3135 |
657
+ | 187.0 | 27676 | 0.0 | 0.3102 |
658
+ | 188.0 | 27824 | 0.0 | 0.3394 |
659
+ | 189.0 | 27972 | 0.0 | 0.3256 |
660
+ | 190.0 | 28120 | 0.0 | 0.2973 |
661
+ | 191.0 | 28268 | 0.0 | 0.3124 |
662
+ | 192.0 | 28416 | 0.0 | 0.321 |
663
+ | 193.0 | 28564 | 0.0 | 0.3332 |
664
+ | 194.0 | 28712 | 0.0 | 0.3136 |
665
+ | 195.0 | 28860 | 0.0 | 0.32 |
666
+ | 196.0 | 29008 | 0.0 | 0.3486 |
667
+ | 197.0 | 29156 | 0.0 | 0.3259 |
668
+ | 198.0 | 29304 | 0.0 | 0.3134 |
669
+ | 199.0 | 29452 | 0.0 | 0.3437 |
670
+ | 200.0 | 29600 | 0.0 | 0.3029 |
671
+
672
+ * The bold row denotes the saved checkpoint.
673
  ### Framework Versions
674
  - Python: 3.10.12
675
  - SetFit: 1.0.3
676
+ - Sentence Transformers: 2.3.1
677
+ - Transformers: 4.37.2
678
  - PyTorch: 2.1.0+cu121
679
+ - Datasets: 2.17.1
680
+ - Tokenizers: 0.15.2
681
 
682
  ## Citation
683
 
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "/root/.cache/torch/sentence_transformers/BAAI_bge-small-en-v1.5/",
3
  "architectures": [
4
  "BertModel"
5
  ],
@@ -24,7 +24,7 @@
24
  "pad_token_id": 0,
25
  "position_embedding_type": "absolute",
26
  "torch_dtype": "float32",
27
- "transformers_version": "4.35.2",
28
  "type_vocab_size": 2,
29
  "use_cache": true,
30
  "vocab_size": 30522
 
1
  {
2
+ "_name_or_path": "path/to/your/output_dir/step_9324",
3
  "architectures": [
4
  "BertModel"
5
  ],
 
24
  "pad_token_id": 0,
25
  "position_embedding_type": "absolute",
26
  "torch_dtype": "float32",
27
+ "transformers_version": "4.37.2",
28
  "type_vocab_size": 2,
29
  "use_cache": true,
30
  "vocab_size": 30522
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:34bc18b710d5938ff734c02b30b212c5f3972732c0a9ef6f687fead5bbc7db99
3
  size 133462128
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:05f550ffe917a9861e23ffed9d358160ca551629d1839ed53233b503100fc7c8
3
  size 133462128
model_head.pkl CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3237c01d5595ea25eac4e75aeeb2f5a07cea99b9d189cd60cdcf10d6a28b6430
3
  size 10143
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:135b8e639a759ff3525a6cf26ae9df097df6e01baec87a377f8d1d8f5245e68a
3
  size 10143
special_tokens_map.json CHANGED
@@ -1,7 +1,37 @@
1
  {
2
- "cls_token": "[CLS]",
3
- "mask_token": "[MASK]",
4
- "pad_token": "[PAD]",
5
- "sep_token": "[SEP]",
6
- "unk_token": "[UNK]"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7
  }
 
1
  {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
  }
tokenizer_config.json CHANGED
@@ -46,12 +46,19 @@
46
  "do_basic_tokenize": true,
47
  "do_lower_case": true,
48
  "mask_token": "[MASK]",
 
49
  "model_max_length": 512,
50
  "never_split": null,
 
51
  "pad_token": "[PAD]",
 
 
52
  "sep_token": "[SEP]",
 
53
  "strip_accents": null,
54
  "tokenize_chinese_chars": true,
55
  "tokenizer_class": "BertTokenizer",
 
 
56
  "unk_token": "[UNK]"
57
  }
 
46
  "do_basic_tokenize": true,
47
  "do_lower_case": true,
48
  "mask_token": "[MASK]",
49
+ "max_length": 512,
50
  "model_max_length": 512,
51
  "never_split": null,
52
+ "pad_to_multiple_of": null,
53
  "pad_token": "[PAD]",
54
+ "pad_token_type_id": 0,
55
+ "padding_side": "right",
56
  "sep_token": "[SEP]",
57
+ "stride": 0,
58
  "strip_accents": null,
59
  "tokenize_chinese_chars": true,
60
  "tokenizer_class": "BertTokenizer",
61
+ "truncation_side": "right",
62
+ "truncation_strategy": "longest_first",
63
  "unk_token": "[UNK]"
64
  }