Hugging Face's logo

twright8
/
setfit-oversample-labels-lobbying

Text Classification
setfit
Safetensors
sentence-transformers
bert
generated_from_setfit_trainer
Eval Results
Model card Files Community
twright8 commited on
Commit
7b04920
•
1 Parent(s): 8117533

Push model using huggingface_hub.

Browse files
Files changed (4) hide show
  1. README.md +94 -36
  2. config.json +1 -1
  3. model.safetensors +1 -1
  4. model_head.pkl +1 -1
README.md CHANGED
@@ -18,7 +18,7 @@ widget:
18
  - text: To discuss Northern Powerhouse Rail and HS2
19
  - text: To discuss food security
20
  - text: Electricity market
21
- inference: True
22
  model-index:
23
  - name: SetFit
24
  results:
@@ -31,10 +31,10 @@ model-index:
31
  split: test
32
  metrics:
33
  - type: f1
34
- value: 0.92
35
  name: F1
36
  - type: accuracy
37
- value: 0.9658119658119658
38
  name: Accuracy
39
  ---
40
 
@@ -68,9 +68,9 @@ The model has been trained using an efficient few-shot learning technique that i
68
  ## Evaluation
69
 
70
  ### Metrics
71
- | Label | F1 | Accuracy |
72
- |:--------|:-----|:---------|
73
- | **all** | 0.92 | 0.9658 |
74
 
75
  ## Uses
76
 
@@ -122,15 +122,15 @@ preds = model("Electricity market")
122
  ### Training Set Metrics
123
  | Training set | Min | Median | Max |
124
  |:-------------|:----|:--------|:----|
125
- | Word count | 2 | 26.1406 | 153 |
126
 
127
  ### Training Hyperparameters
128
  - batch_size: (16, 2)
129
- - num_epochs: (4, 9)
130
  - max_steps: -1
131
  - sampling_strategy: oversampling
132
- - body_learning_rate: (1.0797496673911536e-05, 3.457046714445997e-05)
133
- - head_learning_rate: 0.0004470582121407239
134
  - loss: CoSENTLoss
135
  - distance_metric: cosine_distance
136
  - margin: 0.25
@@ -142,32 +142,90 @@ preds = model("Electricity market")
142
  - load_best_model_at_end: True
143
 
144
  ### Training Results
145
- | Epoch | Step | Training Loss | Validation Loss |
146
- |:-------:|:-------:|:-------------:|:---------------:|
147
- | 0.0040 | 1 | 19.1843 | - |
148
- | 0.2024 | 50 | 11.3434 | - |
149
- | 0.4049 | 100 | 9.3116 | - |
150
- | 0.6073 | 150 | 2.7233 | - |
151
- | 0.8097 | 200 | 1.5662 | - |
152
- | **1.0** | **247** | **-** | **14.3603** |
153
- | 1.0121 | 250 | 0.0159 | - |
154
- | 1.2146 | 300 | 0.0135 | - |
155
- | 1.4170 | 350 | 0.0003 | - |
156
- | 1.6194 | 400 | 0.0002 | - |
157
- | 1.8219 | 450 | 0.0007 | - |
158
- | 2.0 | 494 | - | 16.8205 |
159
- | 2.0243 | 500 | 0.0023 | - |
160
- | 2.2267 | 550 | 0.0004 | - |
161
- | 2.4291 | 600 | 0.0001 | - |
162
- | 2.6316 | 650 | 0.0 | - |
163
- | 2.8340 | 700 | 0.0003 | - |
164
- | 3.0 | 741 | - | 15.2312 |
165
- | 3.0364 | 750 | 0.0 | - |
166
- | 3.2389 | 800 | 3.1257 | - |
167
- | 3.4413 | 850 | 0.0001 | - |
168
- | 3.6437 | 900 | 0.0002 | - |
169
- | 3.8462 | 950 | 0.0139 | - |
170
- | 4.0 | 988 | - | 14.4995 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
171
 
172
  * The bold row denotes the saved checkpoint.
173
  ### Framework Versions
 
18
  - text: To discuss Northern Powerhouse Rail and HS2
19
  - text: To discuss food security
20
  - text: Electricity market
21
+ inference: false
22
  model-index:
23
  - name: SetFit
24
  results:
 
31
  split: test
32
  metrics:
33
  - type: f1
34
+ value: 0.9056603773584904
35
  name: F1
36
  - type: accuracy
37
+ value: 0.9572649572649573
38
  name: Accuracy
39
  ---
40
 
 
68
  ## Evaluation
69
 
70
  ### Metrics
71
+ | Label | F1 | Accuracy |
72
+ |:--------|:-------|:---------|
73
+ | **all** | 0.9057 | 0.9573 |
74
 
75
  ## Uses
76
 
 
122
  ### Training Set Metrics
123
  | Training set | Min | Median | Max |
124
  |:-------------|:----|:--------|:----|
125
+ | Word count | 2 | 21.5644 | 153 |
126
 
127
  ### Training Hyperparameters
128
  - batch_size: (16, 2)
129
+ - num_epochs: (6, 9)
130
  - max_steps: -1
131
  - sampling_strategy: oversampling
132
+ - body_learning_rate: (7.928034854554858e-06, 2.7001088851580374e-05)
133
+ - head_learning_rate: 0.009321171293151879
134
  - loss: CoSENTLoss
135
  - distance_metric: cosine_distance
136
  - margin: 0.25
 
142
  - load_best_model_at_end: True
143
 
144
  ### Training Results
145
+ | Epoch | Step | Training Loss | Validation Loss |
146
+ |:-------:|:--------:|:-------------:|:---------------:|
147
+ | 0.0018 | 1 | 8.669 | - |
148
+ | 0.0880 | 50 | 8.6617 | - |
149
+ | 0.1761 | 100 | 12.5549 | - |
150
+ | 0.2641 | 150 | 3.1895 | - |
151
+ | 0.3521 | 200 | 16.3181 | - |
152
+ | 0.4401 | 250 | 0.7513 | - |
153
+ | 0.5282 | 300 | 4.6653 | - |
154
+ | 0.0018 | 1 | 0.0059 | - |
155
+ | 0.0880 | 50 | 3.4564 | - |
156
+ | 0.1761 | 100 | 0.5523 | - |
157
+ | 0.2641 | 150 | 0.2372 | - |
158
+ | 0.3521 | 200 | 4.288 | - |
159
+ | 0.4401 | 250 | 0.0027 | - |
160
+ | 0.5282 | 300 | 0.0002 | - |
161
+ | 0.6162 | 350 | 0.0002 | - |
162
+ | 0.7042 | 400 | 0.0001 | - |
163
+ | 0.7923 | 450 | 0.0015 | - |
164
+ | 0.8803 | 500 | 3.5596 | - |
165
+ | 0.9683 | 550 | 0.0 | - |
166
+ | 1.0 | 568 | - | 10.2261 |
167
+ | 1.0563 | 600 | 0.0 | - |
168
+ | 1.1444 | 650 | 0.0011 | - |
169
+ | 1.2324 | 700 | 0.0013 | - |
170
+ | 1.3204 | 750 | 0.0037 | - |
171
+ | 1.4085 | 800 | 0.0013 | - |
172
+ | 1.4965 | 850 | 0.0002 | - |
173
+ | 1.5845 | 900 | 0.0 | - |
174
+ | 1.6725 | 950 | 0.0 | - |
175
+ | 1.7606 | 1000 | 0.0001 | - |
176
+ | 1.8486 | 1050 | 0.0001 | - |
177
+ | 1.9366 | 1100 | 0.0001 | - |
178
+ | 2.0 | 1136 | - | 8.4908 |
179
+ | 2.0246 | 1150 | 0.0001 | - |
180
+ | 2.1127 | 1200 | 0.0 | - |
181
+ | 2.2007 | 1250 | 0.0005 | - |
182
+ | 2.2887 | 1300 | 0.0004 | - |
183
+ | 2.3768 | 1350 | 0.0 | - |
184
+ | 2.4648 | 1400 | 0.0009 | - |
185
+ | 2.5528 | 1450 | 0.0 | - |
186
+ | 2.6408 | 1500 | 0.0 | - |
187
+ | 2.7289 | 1550 | 0.0 | - |
188
+ | 2.8169 | 1600 | 0.0 | - |
189
+ | 2.9049 | 1650 | 0.0001 | - |
190
+ | 2.9930 | 1700 | 0.0003 | - |
191
+ | 3.0 | 1704 | - | 8.5594 |
192
+ | 3.0810 | 1750 | 0.0001 | - |
193
+ | 3.1690 | 1800 | 0.0 | - |
194
+ | 3.2570 | 1850 | 0.0002 | - |
195
+ | 3.3451 | 1900 | 0.0001 | - |
196
+ | 3.4331 | 1950 | 0.0 | - |
197
+ | 3.5211 | 2000 | 0.0 | - |
198
+ | 3.6092 | 2050 | 0.0 | - |
199
+ | 3.6972 | 2100 | 0.0 | - |
200
+ | 3.7852 | 2150 | 0.0 | - |
201
+ | 3.8732 | 2200 | 0.0002 | - |
202
+ | 3.9613 | 2250 | 0.0001 | - |
203
+ | **4.0** | **2272** | **-** | **8.4573** |
204
+ | 4.0493 | 2300 | 0.0 | - |
205
+ | 4.1373 | 2350 | 0.0 | - |
206
+ | 4.2254 | 2400 | 0.0002 | - |
207
+ | 4.3134 | 2450 | 0.0 | - |
208
+ | 4.4014 | 2500 | 0.0003 | - |
209
+ | 4.4894 | 2550 | 0.0001 | - |
210
+ | 4.5775 | 2600 | 0.0001 | - |
211
+ | 4.6655 | 2650 | 0.0001 | - |
212
+ | 4.7535 | 2700 | 0.0001 | - |
213
+ | 4.8415 | 2750 | 0.0001 | - |
214
+ | 4.9296 | 2800 | 0.0012 | - |
215
+ | 5.0 | 2840 | - | 8.6305 |
216
+ | 5.0176 | 2850 | 0.0009 | - |
217
+ | 5.1056 | 2900 | 0.0 | - |
218
+ | 5.1937 | 2950 | 0.0001 | - |
219
+ | 5.2817 | 3000 | 0.0 | - |
220
+ | 5.3697 | 3050 | 0.0 | - |
221
+ | 5.4577 | 3100 | 0.0001 | - |
222
+ | 5.5458 | 3150 | 0.0007 | - |
223
+ | 5.6338 | 3200 | 0.0002 | - |
224
+ | 5.7218 | 3250 | 0.0 | - |
225
+ | 5.8099 | 3300 | 0.0001 | - |
226
+ | 5.8979 | 3350 | 0.0002 | - |
227
+ | 5.9859 | 3400 | 0.0 | - |
228
+ | 6.0 | 3408 | - | 8.9528 |
229
 
230
  * The bold row denotes the saved checkpoint.
231
  ### Framework Versions
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "checkpoints/step_247",
3
  "architectures": [
4
  "BertModel"
5
  ],
 
1
  {
2
+ "_name_or_path": "checkpoints/step_2272",
3
  "architectures": [
4
  "BertModel"
5
  ],
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:aa0062895387b7b9c938b60ce80ca18d46d33838add4813a584fb49de42bb0a6
3
  size 437951328
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed0b25593019a6905927414543d4ff10496193c8bf59624e161ea5273861afa9
3
  size 437951328
model_head.pkl CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1f2ae84e5670d630985fc9139ffe7d05254c490b0c28f4ce5c230d0eae56703e
3
  size 13858
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4b9e8c7cf74976a54667babd7f6b61832fc6eecaf7d53a3736c39a16061a3fe6
3
  size 13858