Hugging Face's logo

maliced
/
mdd-transformer-tiny

Transformers
TensorBoard
Safetensors
English
mdd_transformer
Generated from Trainer
Model card Files
xet
Metrics Community
maliced commited on
Commit
0c8de93
·
verified ·
1 Parent(s): d7ff41c

End of training

Browse files
Files changed (1) hide show
  1. README.md +223 -55
README.md CHANGED
@@ -1,55 +1,223 @@
1
- ---
2
- library_name: transformers
3
- language:
4
- - en
5
- tags:
6
- - generated_from_trainer
7
- datasets:
8
- - maliced/l2-arctic
9
- model-index:
10
- - name: MDD Transformer Tiny
11
- results: []
12
- ---
13
-
14
- <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
- should probably proofread and complete it, then remove this comment. -->
16
-
17
- # MDD Transformer Tiny
18
-
19
- This model is a fine-tuned version of [](https://huggingface.co/) on the L2 Arctic dataset.
20
-
21
- ## Model description
22
-
23
- More information needed
24
-
25
- ## Intended uses & limitations
26
-
27
- More information needed
28
-
29
- ## Training and evaluation data
30
-
31
- More information needed
32
-
33
- ## Training procedure
34
-
35
- ### Training hyperparameters
36
-
37
- The following hyperparameters were used during training:
38
- - learning_rate: 1e-05
39
- - train_batch_size: 1
40
- - eval_batch_size: 8
41
- - seed: 42
42
- - gradient_accumulation_steps: 16
43
- - total_train_batch_size: 16
44
- - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
45
- - lr_scheduler_type: linear
46
- - lr_scheduler_warmup_steps: 500
47
- - training_steps: 4000
48
- - mixed_precision_training: Native AMP
49
-
50
- ### Framework versions
51
-
52
- - Transformers 4.52.1
53
- - Pytorch 2.7.0+cpu
54
- - Datasets 3.6.0
55
- - Tokenizers 0.21.1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ language:
4
+ - en
5
+ tags:
6
+ - generated_from_trainer
7
+ datasets:
8
+ - maliced/l2-arctic
9
+ model-index:
10
+ - name: MDD Transformer Tiny
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # MDD Transformer Tiny
18
+
19
+ This model is a fine-tuned version of [](https://huggingface.co/) on the L2 Arctic dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 2.7396
22
+
23
+ ## Model description
24
+
25
+ More information needed
26
+
27
+ ## Intended uses & limitations
28
+
29
+ More information needed
30
+
31
+ ## Training and evaluation data
32
+
33
+ More information needed
34
+
35
+ ## Training procedure
36
+
37
+ ### Training hyperparameters
38
+
39
+ The following hyperparameters were used during training:
40
+ - learning_rate: 1e-05
41
+ - train_batch_size: 4
42
+ - eval_batch_size: 4
43
+ - seed: 42
44
+ - gradient_accumulation_steps: 4
45
+ - total_train_batch_size: 16
46
+ - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
47
+ - lr_scheduler_type: linear
48
+ - lr_scheduler_warmup_steps: 500
49
+ - training_steps: 4000
50
+ - mixed_precision_training: Native AMP
51
+
52
+ ### Training results
53
+
54
+ | Training Loss | Epoch | Step | Validation Loss |
55
+ |:-------------:|:------:|:----:|:---------------:|
56
+ | 3.9162 | 0.0199 | 25 | 3.9053 |
57
+ | 3.9087 | 0.0398 | 50 | 3.8977 |
58
+ | 3.8963 | 0.0598 | 75 | 3.8853 |
59
+ | 3.8862 | 0.0797 | 100 | 3.8683 |
60
+ | 3.867 | 0.0996 | 125 | 3.8470 |
61
+ | 3.8441 | 0.1195 | 150 | 3.8220 |
62
+ | 3.8197 | 0.1394 | 175 | 3.7936 |
63
+ | 3.7952 | 0.1594 | 200 | 3.7623 |
64
+ | 3.755 | 0.1793 | 225 | 3.7280 |
65
+ | 3.7265 | 0.1992 | 250 | 3.6916 |
66
+ | 3.695 | 0.2191 | 275 | 3.6536 |
67
+ | 3.6635 | 0.2390 | 300 | 3.6135 |
68
+ | 3.62 | 0.2590 | 325 | 3.5717 |
69
+ | 3.5774 | 0.2789 | 350 | 3.5281 |
70
+ | 3.5498 | 0.2988 | 375 | 3.4830 |
71
+ | 3.4899 | 0.3187 | 400 | 3.4364 |
72
+ | 3.4769 | 0.3386 | 425 | 3.3878 |
73
+ | 3.4147 | 0.3586 | 450 | 3.3388 |
74
+ | 3.377 | 0.3785 | 475 | 3.2900 |
75
+ | 3.3185 | 0.3984 | 500 | 3.2455 |
76
+ | 3.3012 | 0.4183 | 525 | 3.2052 |
77
+ | 3.264 | 0.4382 | 550 | 3.1712 |
78
+ | 3.203 | 0.4582 | 575 | 3.1432 |
79
+ | 3.188 | 0.4781 | 600 | 3.1190 |
80
+ | 3.1786 | 0.4980 | 625 | 3.0983 |
81
+ | 3.1699 | 0.5179 | 650 | 3.0802 |
82
+ | 3.1318 | 0.5378 | 675 | 3.0647 |
83
+ | 3.1235 | 0.5578 | 700 | 3.0516 |
84
+ | 3.101 | 0.5777 | 725 | 3.0393 |
85
+ | 3.0758 | 0.5976 | 750 | 3.0277 |
86
+ | 3.1111 | 0.6175 | 775 | 3.0175 |
87
+ | 3.101 | 0.6375 | 800 | 3.0075 |
88
+ | 3.0834 | 0.6574 | 825 | 2.9986 |
89
+ | 3.0237 | 0.6773 | 850 | 2.9908 |
90
+ | 3.0341 | 0.6972 | 875 | 2.9833 |
91
+ | 3.0629 | 0.7171 | 900 | 2.9757 |
92
+ | 3.0283 | 0.7371 | 925 | 2.9693 |
93
+ | 2.9794 | 0.7570 | 950 | 2.9626 |
94
+ | 3.0241 | 0.7769 | 975 | 2.9566 |
95
+ | 3.0006 | 0.7968 | 1000 | 2.9509 |
96
+ | 2.9883 | 0.8167 | 1025 | 2.9452 |
97
+ | 2.9556 | 0.8367 | 1050 | 2.9405 |
98
+ | 2.9768 | 0.8566 | 1075 | 2.9350 |
99
+ | 2.9641 | 0.8765 | 1100 | 2.9300 |
100
+ | 2.9836 | 0.8964 | 1125 | 2.9251 |
101
+ | 2.9594 | 0.9163 | 1150 | 2.9206 |
102
+ | 2.9328 | 0.9363 | 1175 | 2.9160 |
103
+ | 2.967 | 0.9562 | 1200 | 2.9121 |
104
+ | 2.9729 | 0.9761 | 1225 | 2.9075 |
105
+ | 2.9495 | 0.9960 | 1250 | 2.9039 |
106
+ | 2.9256 | 1.0159 | 1275 | 2.8999 |
107
+ | 2.9329 | 1.0359 | 1300 | 2.8959 |
108
+ | 2.9471 | 1.0558 | 1325 | 2.8924 |
109
+ | 2.9211 | 1.0757 | 1350 | 2.8885 |
110
+ | 2.9457 | 1.0956 | 1375 | 2.8850 |
111
+ | 2.8959 | 1.1155 | 1400 | 2.8816 |
112
+ | 2.917 | 1.1355 | 1425 | 2.8777 |
113
+ | 2.9581 | 1.1554 | 1450 | 2.8745 |
114
+ | 2.892 | 1.1753 | 1475 | 2.8711 |
115
+ | 2.9026 | 1.1952 | 1500 | 2.8676 |
116
+ | 2.8666 | 1.2151 | 1525 | 2.8647 |
117
+ | 2.8683 | 1.2351 | 1550 | 2.8614 |
118
+ | 2.8808 | 1.2550 | 1575 | 2.8582 |
119
+ | 2.8811 | 1.2749 | 1600 | 2.8552 |
120
+ | 2.9028 | 1.2948 | 1625 | 2.8520 |
121
+ | 2.9027 | 1.3147 | 1650 | 2.8489 |
122
+ | 2.8698 | 1.3347 | 1675 | 2.8459 |
123
+ | 2.8509 | 1.3546 | 1700 | 2.8429 |
124
+ | 2.8765 | 1.3745 | 1725 | 2.8398 |
125
+ | 2.9026 | 1.3944 | 1750 | 2.8371 |
126
+ | 2.8643 | 1.4143 | 1775 | 2.8344 |
127
+ | 2.8731 | 1.4343 | 1800 | 2.8315 |
128
+ | 2.8199 | 1.4542 | 1825 | 2.8288 |
129
+ | 2.8492 | 1.4741 | 1850 | 2.8262 |
130
+ | 2.8163 | 1.4940 | 1875 | 2.8236 |
131
+ | 2.9043 | 1.5139 | 1900 | 2.8210 |
132
+ | 2.8232 | 1.5339 | 1925 | 2.8186 |
133
+ | 2.8258 | 1.5538 | 1950 | 2.8164 |
134
+ | 2.8085 | 1.5737 | 1975 | 2.8141 |
135
+ | 2.8648 | 1.5936 | 2000 | 2.8120 |
136
+ | 2.8648 | 1.6135 | 2025 | 2.8097 |
137
+ | 2.8369 | 1.6335 | 2050 | 2.8075 |
138
+ | 2.8402 | 1.6534 | 2075 | 2.8057 |
139
+ | 2.8694 | 1.6733 | 2100 | 2.8037 |
140
+ | 2.8073 | 1.6932 | 2125 | 2.8019 |
141
+ | 2.8293 | 1.7131 | 2150 | 2.8001 |
142
+ | 2.803 | 1.7331 | 2175 | 2.7980 |
143
+ | 2.8113 | 1.7530 | 2200 | 2.7961 |
144
+ | 2.8499 | 1.7729 | 2225 | 2.7944 |
145
+ | 2.8048 | 1.7928 | 2250 | 2.7927 |
146
+ | 2.832 | 1.8127 | 2275 | 2.7912 |
147
+ | 2.809 | 1.8327 | 2300 | 2.7894 |
148
+ | 2.8517 | 1.8526 | 2325 | 2.7879 |
149
+ | 2.8059 | 1.8725 | 2350 | 2.7863 |
150
+ | 2.7564 | 1.8924 | 2375 | 2.7849 |
151
+ | 2.8427 | 1.9124 | 2400 | 2.7834 |
152
+ | 2.7905 | 1.9323 | 2425 | 2.7820 |
153
+ | 2.7755 | 1.9522 | 2450 | 2.7805 |
154
+ | 2.7935 | 1.9721 | 2475 | 2.7790 |
155
+ | 2.8234 | 1.9920 | 2500 | 2.7777 |
156
+ | 2.8659 | 2.0120 | 2525 | 2.7764 |
157
+ | 2.7873 | 2.0319 | 2550 | 2.7751 |
158
+ | 2.8081 | 2.0518 | 2575 | 2.7739 |
159
+ | 2.82 | 2.0717 | 2600 | 2.7726 |
160
+ | 2.8274 | 2.0916 | 2625 | 2.7715 |
161
+ | 2.7901 | 2.1116 | 2650 | 2.7705 |
162
+ | 2.7661 | 2.1315 | 2675 | 2.7692 |
163
+ | 2.777 | 2.1514 | 2700 | 2.7679 |
164
+ | 2.8142 | 2.1713 | 2725 | 2.7667 |
165
+ | 2.8137 | 2.1912 | 2750 | 2.7658 |
166
+ | 2.7921 | 2.2112 | 2775 | 2.7648 |
167
+ | 2.7432 | 2.2311 | 2800 | 2.7637 |
168
+ | 2.7788 | 2.2510 | 2825 | 2.7626 |
169
+ | 2.7904 | 2.2709 | 2850 | 2.7616 |
170
+ | 2.7674 | 2.2908 | 2875 | 2.7606 |
171
+ | 2.7698 | 2.3108 | 2900 | 2.7596 |
172
+ | 2.7987 | 2.3307 | 2925 | 2.7588 |
173
+ | 2.7801 | 2.3506 | 2950 | 2.7579 |
174
+ | 2.7838 | 2.3705 | 2975 | 2.7570 |
175
+ | 2.7811 | 2.3904 | 3000 | 2.7561 |
176
+ | 2.7864 | 2.4104 | 3025 | 2.7554 |
177
+ | 2.8022 | 2.4303 | 3050 | 2.7545 |
178
+ | 2.7476 | 2.4502 | 3075 | 2.7535 |
179
+ | 2.7964 | 2.4701 | 3100 | 2.7527 |
180
+ | 2.7696 | 2.4900 | 3125 | 2.7522 |
181
+ | 2.8205 | 2.5100 | 3150 | 2.7515 |
182
+ | 2.7584 | 2.5299 | 3175 | 2.7508 |
183
+ | 2.7627 | 2.5498 | 3200 | 2.7502 |
184
+ | 2.7485 | 2.5697 | 3225 | 2.7495 |
185
+ | 2.749 | 2.5896 | 3250 | 2.7490 |
186
+ | 2.7788 | 2.6096 | 3275 | 2.7484 |
187
+ | 2.7639 | 2.6295 | 3300 | 2.7478 |
188
+ | 2.7427 | 2.6494 | 3325 | 2.7472 |
189
+ | 2.7432 | 2.6693 | 3350 | 2.7467 |
190
+ | 2.8058 | 2.6892 | 3375 | 2.7462 |
191
+ | 2.7592 | 2.7092 | 3400 | 2.7457 |
192
+ | 2.7299 | 2.7291 | 3425 | 2.7452 |
193
+ | 2.7342 | 2.7490 | 3450 | 2.7447 |
194
+ | 2.7837 | 2.7689 | 3475 | 2.7442 |
195
+ | 2.767 | 2.7888 | 3500 | 2.7438 |
196
+ | 2.7443 | 2.8088 | 3525 | 2.7434 |
197
+ | 2.7535 | 2.8287 | 3550 | 2.7429 |
198
+ | 2.7426 | 2.8486 | 3575 | 2.7425 |
199
+ | 2.7515 | 2.8685 | 3600 | 2.7422 |
200
+ | 2.7643 | 2.8884 | 3625 | 2.7419 |
201
+ | 2.7729 | 2.9084 | 3650 | 2.7416 |
202
+ | 2.7254 | 2.9283 | 3675 | 2.7413 |
203
+ | 2.7428 | 2.9482 | 3700 | 2.7411 |
204
+ | 2.7475 | 2.9681 | 3725 | 2.7408 |
205
+ | 2.7293 | 2.9880 | 3750 | 2.7406 |
206
+ | 2.8002 | 3.0080 | 3775 | 2.7404 |
207
+ | 2.7434 | 3.0279 | 3800 | 2.7403 |
208
+ | 2.7217 | 3.0478 | 3825 | 2.7401 |
209
+ | 2.7748 | 3.0677 | 3850 | 2.7400 |
210
+ | 2.7278 | 3.0876 | 3875 | 2.7399 |
211
+ | 2.7534 | 3.1076 | 3900 | 2.7398 |
212
+ | 2.7662 | 3.1275 | 3925 | 2.7397 |
213
+ | 2.7514 | 3.1474 | 3950 | 2.7397 |
214
+ | 2.7905 | 3.1673 | 3975 | 2.7396 |
215
+ | 2.7737 | 3.1873 | 4000 | 2.7396 |
216
+
217
+
218
+ ### Framework versions
219
+
220
+ - Transformers 4.52.3
221
+ - Pytorch 2.7.0+cpu
222
+ - Datasets 3.6.0
223
+ - Tokenizers 0.21.1