<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
indonesian-roberta-base-ler
This model is a fine-tuned version of flax-community/indonesian-roberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0291
- Overall Precision: 0.9294
- Overall Recall: 0.9191
- Overall F1: 0.9242
- Overall Accuracy: 0.9968
- Jenis amar F1: 0.9379
- Jenis dakwaan F1: 0.8644
- Jenis perkara F1: 0.9096
- Melanggar uu (dakwaan) F1: 0.8062
- Melanggar uu (pertimbangan hukum) F1: 0.6441
- Melanggar uu (tuntutan) F1: 0.9248
- Nama hakim anggota F1: 0.9640
- Nama hakim ketua F1: 0.9741
- Nama jaksa F1: 0.9614
- Nama panitera F1: 0.9756
- Nama pengacara F1: 0.9000
- Nama pengadilan F1: 0.9982
- Nama saksi F1: 0.9386
- Nama terdakwa F1: 0.9786
- Nomor putusan F1: 0.9963
- Putusan hukuman F1: 0.9433
- Tanggal kejadian F1: 0.3988
- Tanggal putusan F1: 0.9680
- Tingkat kasus F1: 0.9853
- Tuntutan hukuman F1: 0.8867
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 8
- eval_batch_size: 8
- max_sequence_length: 128
- stride: 0%
- decay_rate: 0.01
- num_epochs: 16
Training results
Training Loss | Epoch | Step | Validation Loss | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy | Jenis amar F1 | Jenis dakwaan F1 | Jenis perkara F1 | Melanggar uu (dakwaan) F1 | Melanggar uu (pertimbangan hukum) F1 | Melanggar uu (tuntutan) F1 | Nama hakim anggota F1 | Nama hakim ketua F1 | Nama jaksa F1 | Nama panitera F1 | Nama pengacara F1 | Nama pengadilan F1 | Nama saksi F1 | Nama terdakwa F1 | Nomor putusan F1 | Putusan hukuman F1 | Tanggal kejadian F1 | Tanggal putusan F1 | Tingkat kasus F1 | Tuntutan hukuman F1 |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
0.0204 | 1.0 | 5641 | 0.0163 | 0.8647 | 0.8564 | 0.8605 | 0.9960 | 0.8723 | 0.5028 | 0.7307 | 0.6945 | 0.5383 | 0.8472 | 0.9192 | 0.9389 | 0.9086 | 0.9449 | 0.7881 | 0.9821 | 0.8989 | 0.9423 | 0.9530 | 0.7655 | 0.3135 | 0.9630 | 0.9575 | 0.7803 |
0.0133 | 2.0 | 11282 | 0.0193 | 0.8305 | 0.8274 | 0.8289 | 0.9945 | 0.8316 | 0.6958 | 0.6978 | 0.6186 | 0.3940 | 0.8116 | 0.8620 | 0.8495 | 0.8338 | 0.8849 | 0.5220 | 0.9690 | 0.9036 | 0.9532 | 0.9927 | 0.1196 | 0.3154 | 0.9290 | 0.8864 | 0.6835 |
0.0099 | 3.0 | 16923 | 0.0163 | 0.8455 | 0.8801 | 0.8624 | 0.9960 | 0.9 | 0.7671 | 0.7539 | 0.5686 | 0.4050 | 0.4949 | 0.9267 | 0.9168 | 0.9281 | 0.9353 | 0.7831 | 0.9910 | 0.8946 | 0.9722 | 0.9895 | 0.8827 | 0.3423 | 0.9474 | 0.9610 | 0.8459 |
0.0079 | 4.0 | 22564 | 0.0164 | 0.8627 | 0.9019 | 0.8819 | 0.9958 | 0.9022 | 0.7602 | 0.7336 | 0.7157 | 0.5674 | 0.8599 | 0.9550 | 0.9515 | 0.9631 | 0.9695 | 0.8184 | 0.9679 | 0.9131 | 0.9780 | 0.9963 | 0.8650 | 0.3234 | 0.9564 | 0.9722 | 0.8262 |
0.0059 | 5.0 | 28205 | 0.0179 | 0.9157 | 0.8947 | 0.9050 | 0.9968 | 0.9017 | 0.7932 | 0.8425 | 0.7648 | 0.5989 | 0.8992 | 0.9531 | 0.9373 | 0.9560 | 0.9660 | 0.8232 | 0.9784 | 0.9136 | 0.9642 | 0.9898 | 0.9051 | 0.3933 | 0.9645 | 0.9630 | 0.8470 |
0.0052 | 6.0 | 33846 | 0.0183 | 0.8523 | 0.8960 | 0.8736 | 0.9960 | 0.8923 | 0.8015 | 0.8443 | 0.7440 | 0.5949 | 0.8528 | 0.9339 | 0.8898 | 0.9348 | 0.9620 | 0.8814 | 1.0 | 0.9156 | 0.9613 | 0.9936 | 0.8604 | 0.2037 | 0.8600 | 0.9646 | 0.8483 |
0.005 | 7.0 | 39487 | 0.0183 | 0.8901 | 0.9073 | 0.8986 | 0.9965 | 0.9150 | 0.7942 | 0.8355 | 0.7872 | 0.6258 | 0.8641 | 0.9514 | 0.9573 | 0.9665 | 0.9676 | 0.8746 | 0.9964 | 0.9223 | 0.9680 | 0.9945 | 0.8970 | 0.3249 | 0.9354 | 0.9759 | 0.8407 |
0.0039 | 8.0 | 45128 | 0.0197 | 0.8915 | 0.9016 | 0.8965 | 0.9962 | 0.9125 | 0.7638 | 0.7435 | 0.7406 | 0.5828 | 0.8394 | 0.9562 | 0.9683 | 0.9456 | 0.9702 | 0.7469 | 1.0 | 0.8969 | 0.9595 | 0.9969 | 0.9067 | 0.3916 | 0.9404 | 0.9722 | 0.8621 |
0.0031 | 9.0 | 50769 | 0.0225 | 0.8661 | 0.9179 | 0.8913 | 0.9959 | 0.9306 | 0.7714 | 0.7939 | 0.7900 | 0.6084 | 0.9049 | 0.9591 | 0.9643 | 0.9457 | 0.9527 | 0.8127 | 0.9964 | 0.9080 | 0.9716 | 0.9970 | 0.9064 | 0.3388 | 0.8412 | 0.9593 | 0.8727 |
0.0022 | 10.0 | 56410 | 0.0232 | 0.9254 | 0.9111 | 0.9182 | 0.9967 | 0.9212 | 0.8411 | 0.9080 | 0.8044 | 0.6126 | 0.9243 | 0.9560 | 0.9741 | 0.9591 | 0.9642 | 0.9102 | 0.9874 | 0.9240 | 0.9734 | 0.9941 | 0.9351 | 0.4186 | 0.9626 | 0.9779 | 0.8687 |
0.0023 | 11.0 | 62051 | 0.0209 | 0.9289 | 0.9114 | 0.9201 | 0.9969 | 0.9297 | 0.8423 | 0.8843 | 0.7986 | 0.6318 | 0.8808 | 0.9645 | 0.9624 | 0.9585 | 0.9674 | 0.8963 | 0.9946 | 0.9309 | 0.9752 | 0.9966 | 0.9320 | 0.4092 | 0.9697 | 0.9871 | 0.8790 |
0.001 | 12.0 | 67692 | 0.0230 | 0.9279 | 0.9075 | 0.9176 | 0.9968 | 0.9377 | 0.8665 | 0.8771 | 0.7951 | 0.6213 | 0.9079 | 0.9611 | 0.9768 | 0.9576 | 0.9638 | 0.9174 | 0.9964 | 0.9353 | 0.9621 | 0.9967 | 0.9391 | 0.3735 | 0.9665 | 0.9703 | 0.8666 |
0.0007 | 13.0 | 73333 | 0.0244 | 0.9095 | 0.9190 | 0.9142 | 0.9965 | 0.9400 | 0.8610 | 0.8974 | 0.8030 | 0.6337 | 0.9338 | 0.9660 | 0.9712 | 0.9565 | 0.9668 | 0.9181 | 0.9964 | 0.9273 | 0.9640 | 0.9961 | 0.9233 | 0.3664 | 0.9697 | 0.9668 | 0.8845 |
0.0006 | 14.0 | 78974 | 0.0258 | 0.9213 | 0.9186 | 0.9200 | 0.9967 | 0.9315 | 0.8533 | 0.9119 | 0.7934 | 0.6453 | 0.9311 | 0.9617 | 0.9749 | 0.9614 | 0.9702 | 0.8718 | 0.9910 | 0.9320 | 0.9726 | 0.9966 | 0.9249 | 0.3936 | 0.9680 | 0.9871 | 0.8728 |
0.0003 | 15.0 | 84615 | 0.0281 | 0.9260 | 0.9208 | 0.9234 | 0.9969 | 0.9313 | 0.8463 | 0.9150 | 0.7996 | 0.6601 | 0.9176 | 0.9677 | 0.9712 | 0.9599 | 0.9749 | 0.8928 | 0.9946 | 0.9351 | 0.9793 | 0.9963 | 0.9347 | 0.3956 | 0.9680 | 0.9852 | 0.8854 |
0.0001 | 16.0 | 90256 | 0.0291 | 0.9294 | 0.9191 | 0.9242 | 0.9968 | 0.9379 | 0.8644 | 0.9096 | 0.8062 | 0.6441 | 0.9248 | 0.9640 | 0.9741 | 0.9614 | 0.9756 | 0.9000 | 0.9982 | 0.9386 | 0.9786 | 0.9963 | 0.9433 | 0.3988 | 0.9680 | 0.9853 | 0.8867 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.1
- Datasets 2.12.0
- Tokenizers 0.13.3