seedflora commited on
Commit
340443e
·
verified ·
1 Parent(s): d8bf623

Update transformer evaluation results

Browse files
artifacts/transformers/results.csv CHANGED
@@ -1,6 +1,10 @@
1
- model_id,slug,test_accuracy,test_precision,test_recall,test_f1,test_roc_auc
2
- indolem/indobertweet-base-uncased,indolem__indobertweet-base-uncased,0.9857142857142858,1.0,0.9696969696969697,0.9846153846153847,1.0
3
- indobenchmark/indobert-base-p1,indobenchmark__indobert-base-p1,0.9714285714285714,1.0,0.9393939393939394,0.96875,1.0
4
- indobenchmark/indobert-base-p2,indobenchmark__indobert-base-p2,0.9714285714285714,1.0,0.9393939393939394,0.96875,1.0
5
- flax-community/indonesian-roberta-base,flax-community__indonesian-roberta-base,0.9571428571428572,0.96875,0.9393939393939394,0.9538461538461539,0.9967239967239968
6
- indolem/indobert-base-uncased,indolem__indobert-base-uncased,0.9142857142857143,0.9354838709677419,0.8787878787878788,0.90625,0.9918099918099919
 
 
 
 
 
1
+ model_id,slug,test_accuracy,test_precision,test_recall,test_f1,test_roc_auc
2
+ indolem/indobert-base-uncased,indolem__indobert-base-uncased,0.9950738916256158,0.9901960784313726,1.0,0.9950738916256158,0.9998058629392352
3
+ naufalihsan/indonesian-sbert-large,naufalihsan__indonesian-sbert-large,0.9901477832512317,0.9805825242718448,1.0,0.9901960784313726,0.999805862939235
4
+ flax-community/indonesian-roberta-base,flax-community__indonesian-roberta-base,0.9901477832512317,0.9805825242718448,1.0,0.9901960784313726,0.9997087944088526
5
+ sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2,sentence-transformers__paraphrase-multilingual-MiniLM-L12-v2,0.9901477832512317,0.9805825242718448,1.0,0.9901960784313726,0.9996117258784702
6
+ indobenchmark/indobert-base-p1,indobenchmark__indobert-base-p1,0.9901477832512317,0.9805825242718448,1.0,0.9901960784313726,0.9989322461657932
7
+ ChristopherA08/IndoELECTRA,ChristopherA08__IndoELECTRA,0.9901477832512317,0.9805825242718448,1.0,0.9901960784313726,0.998932246165793
8
+ cahya/distilbert-base-indonesian,cahya__distilbert-base-indonesian,0.9901477832512317,0.99009900990099,0.99009900990099,0.99009900990099,0.9998058629392352
9
+ indolem/indobertweet-base-uncased,indolem__indobertweet-base-uncased,0.9852216748768472,0.9803921568627452,0.99009900990099,0.9852216748768472,0.9994175888177051
10
+ w11wo/indonesian-roberta-base-sentiment-classifier,w11wo__indonesian-roberta-base-sentiment-classifier,0.9802955665024632,0.961904761904762,1.0,0.9805825242718448,1.0
artifacts/transformers/results.json CHANGED
@@ -1,155 +1,275 @@
1
- {
2
- "results": [
3
- {
4
- "model_id": "indobenchmark/indobert-base-p1",
5
- "slug": "indobenchmark__indobert-base-p1",
6
- "device": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
7
- "train_rows": 557,
8
- "valid_rows": 70,
9
- "test_rows": 70,
10
- "max_length": 160,
11
- "epochs": 5.0,
12
- "batch_size": 8,
13
- "eval_metrics": {
14
- "eval_loss": 0.0019260338740423322,
15
- "eval_accuracy": 1.0,
16
- "eval_precision": 1.0,
17
- "eval_recall": 1.0,
18
- "eval_f1": 1.0,
19
- "eval_roc_auc": 1.0,
20
- "eval_runtime": 0.4107,
21
- "eval_samples_per_second": 170.432,
22
- "eval_steps_per_second": 12.174,
23
- "epoch": 5.0
24
- },
25
- "test_metrics": {
26
- "accuracy": 0.9714285714285714,
27
- "precision": 1.0,
28
- "recall": 0.9393939393939394,
29
- "f1": 0.96875,
30
- "roc_auc": 1.0
31
- }
32
- },
33
- {
34
- "model_id": "indobenchmark/indobert-base-p2",
35
- "slug": "indobenchmark__indobert-base-p2",
36
- "device": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
37
- "train_rows": 557,
38
- "valid_rows": 70,
39
- "test_rows": 70,
40
- "max_length": 160,
41
- "epochs": 5.0,
42
- "batch_size": 8,
43
- "eval_metrics": {
44
- "eval_loss": 0.0026587690226733685,
45
- "eval_accuracy": 1.0,
46
- "eval_precision": 1.0,
47
- "eval_recall": 1.0,
48
- "eval_f1": 1.0,
49
- "eval_roc_auc": 1.0,
50
- "eval_runtime": 0.3476,
51
- "eval_samples_per_second": 201.368,
52
- "eval_steps_per_second": 14.383,
53
- "epoch": 5.0
54
- },
55
- "test_metrics": {
56
- "accuracy": 0.9714285714285714,
57
- "precision": 1.0,
58
- "recall": 0.9393939393939394,
59
- "f1": 0.96875,
60
- "roc_auc": 1.0
61
- }
62
- },
63
- {
64
- "model_id": "indolem/indobert-base-uncased",
65
- "slug": "indolem__indobert-base-uncased",
66
- "device": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
67
- "train_rows": 557,
68
- "valid_rows": 70,
69
- "test_rows": 70,
70
- "max_length": 160,
71
- "epochs": 5.0,
72
- "batch_size": 8,
73
- "eval_metrics": {
74
- "eval_loss": 0.03823912516236305,
75
- "eval_accuracy": 1.0,
76
- "eval_precision": 1.0,
77
- "eval_recall": 1.0,
78
- "eval_f1": 1.0,
79
- "eval_roc_auc": 1.0,
80
- "eval_runtime": 0.3535,
81
- "eval_samples_per_second": 198.042,
82
- "eval_steps_per_second": 14.146,
83
- "epoch": 5.0
84
- },
85
- "test_metrics": {
86
- "accuracy": 0.9142857142857143,
87
- "precision": 0.9354838709677419,
88
- "recall": 0.8787878787878788,
89
- "f1": 0.90625,
90
- "roc_auc": 0.9918099918099919
91
- }
92
- },
93
- {
94
- "model_id": "indolem/indobertweet-base-uncased",
95
- "slug": "indolem__indobertweet-base-uncased",
96
- "device": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
97
- "train_rows": 557,
98
- "valid_rows": 70,
99
- "test_rows": 70,
100
- "max_length": 160,
101
- "epochs": 5.0,
102
- "batch_size": 8,
103
- "eval_metrics": {
104
- "eval_loss": 0.014322714880108833,
105
- "eval_accuracy": 1.0,
106
- "eval_precision": 1.0,
107
- "eval_recall": 1.0,
108
- "eval_f1": 1.0,
109
- "eval_roc_auc": 1.0,
110
- "eval_runtime": 0.4227,
111
- "eval_samples_per_second": 165.614,
112
- "eval_steps_per_second": 11.83,
113
- "epoch": 5.0
114
- },
115
- "test_metrics": {
116
- "accuracy": 0.9857142857142858,
117
- "precision": 1.0,
118
- "recall": 0.9696969696969697,
119
- "f1": 0.9846153846153847,
120
- "roc_auc": 1.0
121
- }
122
- },
123
- {
124
- "model_id": "flax-community/indonesian-roberta-base",
125
- "slug": "flax-community__indonesian-roberta-base",
126
- "device": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
127
- "train_rows": 557,
128
- "valid_rows": 70,
129
- "test_rows": 70,
130
- "max_length": 160,
131
- "epochs": 5.0,
132
- "batch_size": 8,
133
- "eval_metrics": {
134
- "eval_loss": 0.002546555595472455,
135
- "eval_accuracy": 1.0,
136
- "eval_precision": 1.0,
137
- "eval_recall": 1.0,
138
- "eval_f1": 1.0,
139
- "eval_roc_auc": 1.0,
140
- "eval_runtime": 0.4461,
141
- "eval_samples_per_second": 156.916,
142
- "eval_steps_per_second": 11.208,
143
- "epoch": 5.0
144
- },
145
- "test_metrics": {
146
- "accuracy": 0.9571428571428572,
147
- "precision": 0.96875,
148
- "recall": 0.9393939393939394,
149
- "f1": 0.9538461538461539,
150
- "roc_auc": 0.9967239967239968
151
- }
152
- }
153
- ],
154
- "failures": []
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
155
  }
 
1
+ {
2
+ "results": [
3
+ {
4
+ "model_id": "indolem/indobert-base-uncased",
5
+ "slug": "indolem__indobert-base-uncased",
6
+ "device": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
7
+ "train_rows": 1622,
8
+ "valid_rows": 203,
9
+ "test_rows": 203,
10
+ "max_length": 160,
11
+ "epochs": 5.0,
12
+ "batch_size": 4,
13
+ "eval_metrics": {
14
+ "eval_loss": 0.09579809010028839,
15
+ "eval_accuracy": 0.9901477832512315,
16
+ "eval_precision": 1.0,
17
+ "eval_recall": 0.9803921568627451,
18
+ "eval_f1": 0.9900990099009901,
19
+ "eval_roc_auc": 0.9985439720442633,
20
+ "eval_runtime": 0.7668,
21
+ "eval_samples_per_second": 264.738,
22
+ "eval_steps_per_second": 33.907,
23
+ "epoch": 5.0
24
+ },
25
+ "test_metrics": {
26
+ "accuracy": 0.9950738916256158,
27
+ "precision": 0.9901960784313726,
28
+ "recall": 1.0,
29
+ "f1": 0.9950738916256158,
30
+ "roc_auc": 0.9998058629392351
31
+ }
32
+ },
33
+ {
34
+ "model_id": "naufalihsan/indonesian-sbert-large",
35
+ "slug": "naufalihsan__indonesian-sbert-large",
36
+ "device": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
37
+ "train_rows": 1622,
38
+ "valid_rows": 203,
39
+ "test_rows": 203,
40
+ "max_length": 160,
41
+ "epochs": 5.0,
42
+ "batch_size": 4,
43
+ "eval_metrics": {
44
+ "eval_loss": 0.06560218334197998,
45
+ "eval_accuracy": 0.9901477832512315,
46
+ "eval_precision": 0.9901960784313726,
47
+ "eval_recall": 0.9901960784313726,
48
+ "eval_f1": 0.9901960784313726,
49
+ "eval_roc_auc": 0.9992234517569404,
50
+ "eval_runtime": 1.1515,
51
+ "eval_samples_per_second": 176.289,
52
+ "eval_steps_per_second": 22.579,
53
+ "epoch": 5.0
54
+ },
55
+ "test_metrics": {
56
+ "accuracy": 0.9901477832512315,
57
+ "precision": 0.9805825242718447,
58
+ "recall": 1.0,
59
+ "f1": 0.9901960784313726,
60
+ "roc_auc": 0.999805862939235
61
+ }
62
+ },
63
+ {
64
+ "model_id": "flax-community/indonesian-roberta-base",
65
+ "slug": "flax-community__indonesian-roberta-base",
66
+ "device": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
67
+ "train_rows": 1622,
68
+ "valid_rows": 203,
69
+ "test_rows": 203,
70
+ "max_length": 160,
71
+ "epochs": 5.0,
72
+ "batch_size": 4,
73
+ "eval_metrics": {
74
+ "eval_loss": 0.06540972739458084,
75
+ "eval_accuracy": 0.9901477832512315,
76
+ "eval_precision": 0.9901960784313726,
77
+ "eval_recall": 0.9901960784313726,
78
+ "eval_f1": 0.9901960784313726,
79
+ "eval_roc_auc": 0.9994175888177054,
80
+ "eval_runtime": 0.8868,
81
+ "eval_samples_per_second": 228.92,
82
+ "eval_steps_per_second": 29.32,
83
+ "epoch": 5.0
84
+ },
85
+ "test_metrics": {
86
+ "accuracy": 0.9901477832512315,
87
+ "precision": 0.9805825242718447,
88
+ "recall": 1.0,
89
+ "f1": 0.9901960784313726,
90
+ "roc_auc": 0.9997087944088526
91
+ }
92
+ },
93
+ {
94
+ "model_id": "sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2",
95
+ "slug": "sentence-transformers__paraphrase-multilingual-MiniLM-L12-v2",
96
+ "device": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
97
+ "train_rows": 1622,
98
+ "valid_rows": 203,
99
+ "test_rows": 203,
100
+ "max_length": 160,
101
+ "epochs": 5.0,
102
+ "batch_size": 4,
103
+ "eval_metrics": {
104
+ "eval_loss": 0.11527860164642334,
105
+ "eval_accuracy": 0.9802955665024631,
106
+ "eval_precision": 0.99,
107
+ "eval_recall": 0.9705882352941176,
108
+ "eval_f1": 0.9801980198019802,
109
+ "eval_roc_auc": 0.9977674238012036,
110
+ "eval_runtime": 0.7139,
111
+ "eval_samples_per_second": 284.338,
112
+ "eval_steps_per_second": 36.418,
113
+ "epoch": 5.0
114
+ },
115
+ "test_metrics": {
116
+ "accuracy": 0.9901477832512315,
117
+ "precision": 0.9805825242718447,
118
+ "recall": 1.0,
119
+ "f1": 0.9901960784313726,
120
+ "roc_auc": 0.9996117258784702
121
+ }
122
+ },
123
+ {
124
+ "model_id": "indobenchmark/indobert-base-p1",
125
+ "slug": "indobenchmark__indobert-base-p1",
126
+ "device": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
127
+ "train_rows": 1622,
128
+ "valid_rows": 203,
129
+ "test_rows": 203,
130
+ "max_length": 160,
131
+ "epochs": 5.0,
132
+ "batch_size": 4,
133
+ "eval_metrics": {
134
+ "eval_loss": 0.11625413596630096,
135
+ "eval_accuracy": 0.9753694581280788,
136
+ "eval_precision": 0.970873786407767,
137
+ "eval_recall": 0.9803921568627451,
138
+ "eval_f1": 0.975609756097561,
139
+ "eval_roc_auc": 0.9988351776354106,
140
+ "eval_runtime": 0.5196,
141
+ "eval_samples_per_second": 390.688,
142
+ "eval_steps_per_second": 50.039,
143
+ "epoch": 5.0
144
+ },
145
+ "test_metrics": {
146
+ "accuracy": 0.9901477832512315,
147
+ "precision": 0.9805825242718447,
148
+ "recall": 1.0,
149
+ "f1": 0.9901960784313726,
150
+ "roc_auc": 0.9989322461657931
151
+ }
152
+ },
153
+ {
154
+ "model_id": "ChristopherA08/IndoELECTRA",
155
+ "slug": "ChristopherA08__IndoELECTRA",
156
+ "device": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
157
+ "train_rows": 1622,
158
+ "valid_rows": 203,
159
+ "test_rows": 203,
160
+ "max_length": 160,
161
+ "epochs": 5.0,
162
+ "batch_size": 4,
163
+ "eval_metrics": {
164
+ "eval_loss": 0.07120005786418915,
165
+ "eval_accuracy": 0.9753694581280788,
166
+ "eval_precision": 0.9801980198019802,
167
+ "eval_recall": 0.9705882352941176,
168
+ "eval_f1": 0.9753694581280788,
169
+ "eval_roc_auc": 0.9984469035138808,
170
+ "eval_runtime": 0.4822,
171
+ "eval_samples_per_second": 420.982,
172
+ "eval_steps_per_second": 53.919,
173
+ "epoch": 5.0
174
+ },
175
+ "test_metrics": {
176
+ "accuracy": 0.9901477832512315,
177
+ "precision": 0.9805825242718447,
178
+ "recall": 1.0,
179
+ "f1": 0.9901960784313726,
180
+ "roc_auc": 0.998932246165793
181
+ }
182
+ },
183
+ {
184
+ "model_id": "cahya/distilbert-base-indonesian",
185
+ "slug": "cahya__distilbert-base-indonesian",
186
+ "device": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
187
+ "train_rows": 1622,
188
+ "valid_rows": 203,
189
+ "test_rows": 203,
190
+ "max_length": 160,
191
+ "epochs": 5.0,
192
+ "batch_size": 4,
193
+ "eval_metrics": {
194
+ "eval_loss": 0.08890486508607864,
195
+ "eval_accuracy": 0.9753694581280788,
196
+ "eval_precision": 0.9801980198019802,
197
+ "eval_recall": 0.9705882352941176,
198
+ "eval_f1": 0.9753694581280788,
199
+ "eval_roc_auc": 0.9985439720442633,
200
+ "eval_runtime": 0.326,
201
+ "eval_samples_per_second": 622.647,
202
+ "eval_steps_per_second": 79.748,
203
+ "epoch": 5.0
204
+ },
205
+ "test_metrics": {
206
+ "accuracy": 0.9901477832512315,
207
+ "precision": 0.9900990099009901,
208
+ "recall": 0.9900990099009901,
209
+ "f1": 0.9900990099009901,
210
+ "roc_auc": 0.9998058629392351
211
+ }
212
+ },
213
+ {
214
+ "model_id": "indolem/indobertweet-base-uncased",
215
+ "slug": "indolem__indobertweet-base-uncased",
216
+ "device": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
217
+ "train_rows": 1622,
218
+ "valid_rows": 203,
219
+ "test_rows": 203,
220
+ "max_length": 160,
221
+ "epochs": 5.0,
222
+ "batch_size": 4,
223
+ "eval_metrics": {
224
+ "eval_loss": 0.16416750848293304,
225
+ "eval_accuracy": 0.9753694581280788,
226
+ "eval_precision": 0.98989898989899,
227
+ "eval_recall": 0.9607843137254902,
228
+ "eval_f1": 0.9751243781094527,
229
+ "eval_roc_auc": 0.9959231217239372,
230
+ "eval_runtime": 1.1291,
231
+ "eval_samples_per_second": 179.782,
232
+ "eval_steps_per_second": 23.026,
233
+ "epoch": 5.0
234
+ },
235
+ "test_metrics": {
236
+ "accuracy": 0.9852216748768473,
237
+ "precision": 0.9803921568627451,
238
+ "recall": 0.9900990099009901,
239
+ "f1": 0.9852216748768473,
240
+ "roc_auc": 0.9994175888177053
241
+ }
242
+ },
243
+ {
244
+ "model_id": "w11wo/indonesian-roberta-base-sentiment-classifier",
245
+ "slug": "w11wo__indonesian-roberta-base-sentiment-classifier",
246
+ "device": "NVIDIA GeForce RTX 3070 Ti Laptop GPU",
247
+ "train_rows": 1622,
248
+ "valid_rows": 203,
249
+ "test_rows": 203,
250
+ "max_length": 160,
251
+ "epochs": 5.0,
252
+ "batch_size": 4,
253
+ "eval_metrics": {
254
+ "eval_loss": 0.08981981128454208,
255
+ "eval_accuracy": 0.9901477832512315,
256
+ "eval_precision": 0.9901960784313726,
257
+ "eval_recall": 0.9901960784313726,
258
+ "eval_f1": 0.9901960784313726,
259
+ "eval_roc_auc": 0.9982527664531159,
260
+ "eval_runtime": 0.8957,
261
+ "eval_samples_per_second": 226.647,
262
+ "eval_steps_per_second": 29.029,
263
+ "epoch": 5.0
264
+ },
265
+ "test_metrics": {
266
+ "accuracy": 0.9802955665024631,
267
+ "precision": 0.9619047619047619,
268
+ "recall": 1.0,
269
+ "f1": 0.9805825242718447,
270
+ "roc_auc": 1.0
271
+ }
272
+ }
273
+ ],
274
+ "failures": []
275
  }