Downloads · 30 days
5
21% of all-time downloads
Kyungjin-Kim/embc25_finetuned_30000_en_it-ipa
embc25_finetuned_30000_en_it-ipa is a text classification model from Kyungjin-Kim. Use it when you need a label for a piece of text. It is set up for transformers.
should probably proofread and complete it, then remove this comment. --
Downloads · 30 days
5
21% of all-time downloads
All-time downloads
24
Public
Parameters
62.7M
2.3 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors251 MB · 99%
From the Hugging Face model README
This model is a fine-tuned version of Kyungjin-Kim/mmc_roberta_500000_en_it-ipa on the None dataset. It achieves the following results on the evaluation set:
More information needed
More information needed
More information needed
The following hyperparameters were used during training:
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.5157 | 0.5926 | 500 | 0.5070 | 0.7398 | 0.7113 | 0.8073 | 0.7563 |
| 0.4144 | 1.1849 | 1000 | 0.4530 | 0.7873 | 0.7847 | 0.792 | 0.7883 |
| 0.3657 | 1.7775 | 1500 | 0.4287 | 0.7998 | 0.8529 | 0.7247 | 0.7836 |
| 0.3001 | 2.3698 | 2000 | 0.4143 | 0.8135 | 0.8594 | 0.7497 | 0.8008 |
| 0.3098 | 2.9624 | 2500 | 0.4148 | 0.8188 | 0.7983 | 0.8533 | 0.8249 |
| 0.2545 | 3.5547 | 3000 | 0.4482 | 0.8197 | 0.8307 | 0.803 | 0.8166 |
| 0.1709 | 4.1470 | 3500 | 0.4748 | 0.8245 | 0.8197 | 0.832 | 0.8258 |
| 0.1818 | 4.7396 | 4000 | 0.5012 | 0.825 | 0.8450 | 0.796 | 0.8198 |
| 0.1321 | 5.3319 | 4500 | 0.5836 | 0.8272 | 0.8393 | 0.8093 | 0.8240 |
| 0.125 | 5.9244 | 5000 | 0.6145 | 0.8225 | 0.8080 | 0.846 | 0.8266 |
| 0.1117 | 6.5167 | 5500 | 0.6661 | 0.8225 | 0.8312 | 0.8093 | 0.8201 |
| 0.0822 | 7.1090 | 6000 | 0.7212 | 0.8222 | 0.8119 | 0.8387 | 0.8251 |
| 0.0873 | 7.7016 | 6500 | 0.7630 | 0.8202 | 0.8070 | 0.8417 | 0.8240 |
| 0.066 | 8.2939 | 7000 | 0.8090 | 0.8257 | 0.8334 | 0.814 | 0.8236 |
| 0.0688 | 8.8865 | 7500 | 0.8409 | 0.8257 | 0.8314 | 0.817 | 0.8241 |
| 0.0616 | 9.4788 | 8000 | 0.8897 | 0.8225 | 0.8250 | 0.8187 | 0.8218 |