mt5-summarize-te
This model is a fine-tuned version of google/mt5-small on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.1257
- Rouge1: 0.5211
- Rouge2: 0.4338
- Rougel: 0.4813
- Rougelsum: 0.4819
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: 0.0005
- train_batch_size: 2
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 90
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
---|---|---|---|---|---|---|---|
3.8667 | 0.2 | 100 | 2.5990 | 0.4450 | 0.3695 | 0.4151 | 0.4154 |
3.2326 | 0.39 | 200 | 2.5107 | 0.5023 | 0.4156 | 0.4567 | 0.4571 |
3.1169 | 0.59 | 300 | 2.4503 | 0.5092 | 0.4204 | 0.4775 | 0.4762 |
2.9083 | 0.79 | 400 | 2.4005 | 0.5053 | 0.4179 | 0.4699 | 0.4709 |
2.9652 | 0.98 | 500 | 2.3218 | 0.4833 | 0.4003 | 0.4528 | 0.4555 |
2.8848 | 1.18 | 600 | 2.3262 | 0.5309 | 0.4415 | 0.4868 | 0.4879 |
2.6585 | 1.37 | 700 | 2.3118 | 0.5168 | 0.4273 | 0.4780 | 0.4773 |
2.6662 | 1.57 | 800 | 2.2823 | 0.5112 | 0.4233 | 0.4713 | 0.4727 |
2.7628 | 1.77 | 900 | 2.2381 | 0.5158 | 0.4269 | 0.4798 | 0.4798 |
2.7156 | 1.96 | 1000 | 2.2466 | 0.5280 | 0.4452 | 0.4836 | 0.4844 |
2.5683 | 2.16 | 1100 | 2.2495 | 0.5184 | 0.4300 | 0.4779 | 0.4773 |
2.5248 | 2.36 | 1200 | 2.2498 | 0.5179 | 0.4282 | 0.4790 | 0.4803 |
2.5809 | 2.55 | 1300 | 2.2336 | 0.5233 | 0.4385 | 0.4895 | 0.4920 |
2.7113 | 2.75 | 1400 | 2.2368 | 0.5079 | 0.4207 | 0.4707 | 0.4716 |
2.6151 | 2.95 | 1500 | 2.1993 | 0.5108 | 0.4236 | 0.4681 | 0.4679 |
2.5172 | 3.14 | 1600 | 2.2197 | 0.5138 | 0.4257 | 0.4778 | 0.4781 |
2.5873 | 3.34 | 1700 | 2.1900 | 0.5185 | 0.4312 | 0.4823 | 0.4821 |
2.4245 | 3.53 | 1800 | 2.1982 | 0.5222 | 0.4332 | 0.4837 | 0.4853 |
2.4983 | 3.73 | 1900 | 2.1756 | 0.5125 | 0.4247 | 0.4809 | 0.4810 |
2.3963 | 3.93 | 2000 | 2.1900 | 0.5259 | 0.4400 | 0.4870 | 0.4884 |
2.3465 | 4.12 | 2100 | 2.1963 | 0.5300 | 0.4412 | 0.4900 | 0.4915 |
2.4625 | 4.32 | 2200 | 2.1818 | 0.5277 | 0.4384 | 0.4868 | 0.4882 |
2.4257 | 4.52 | 2300 | 2.1504 | 0.5212 | 0.4342 | 0.4833 | 0.4842 |
2.368 | 4.71 | 2400 | 2.1463 | 0.5252 | 0.4418 | 0.4856 | 0.4869 |
2.427 | 4.91 | 2500 | 2.1581 | 0.5161 | 0.4267 | 0.4766 | 0.4771 |
2.3443 | 5.11 | 2600 | 2.1551 | 0.5167 | 0.4281 | 0.4794 | 0.4794 |
2.2923 | 5.3 | 2700 | 2.1596 | 0.5183 | 0.4255 | 0.4668 | 0.4686 |
2.2956 | 5.5 | 2800 | 2.1438 | 0.5125 | 0.4268 | 0.4747 | 0.4754 |
2.2973 | 5.69 | 2900 | 2.1523 | 0.5139 | 0.4259 | 0.4712 | 0.4722 |
2.3013 | 5.89 | 3000 | 2.1514 | 0.5138 | 0.4236 | 0.4741 | 0.4742 |
2.2222 | 6.09 | 3100 | 2.1558 | 0.5172 | 0.4300 | 0.4773 | 0.4784 |
2.3957 | 6.28 | 3200 | 2.1451 | 0.5203 | 0.4326 | 0.4815 | 0.4817 |
2.1995 | 6.48 | 3300 | 2.1476 | 0.5146 | 0.4264 | 0.4747 | 0.4752 |
2.2931 | 6.68 | 3400 | 2.1252 | 0.5120 | 0.4252 | 0.4683 | 0.4683 |
2.3062 | 6.87 | 3500 | 2.1313 | 0.5197 | 0.4339 | 0.4803 | 0.4807 |
2.2844 | 7.07 | 3600 | 2.1281 | 0.5197 | 0.4339 | 0.4868 | 0.4876 |
2.1158 | 7.27 | 3700 | 2.1438 | 0.5208 | 0.4333 | 0.4818 | 0.4823 |
2.2523 | 7.46 | 3800 | 2.1221 | 0.5197 | 0.4324 | 0.4783 | 0.4788 |
2.2389 | 7.66 | 3900 | 2.1336 | 0.5144 | 0.4262 | 0.4769 | 0.4771 |
2.2209 | 7.85 | 4000 | 2.1317 | 0.5211 | 0.4338 | 0.4813 | 0.4819 |
2.1828 | 8.05 | 4100 | 2.1366 | 0.5208 | 0.4336 | 0.4814 | 0.4816 |
2.2746 | 8.25 | 4200 | 2.1325 | 0.5219 | 0.4342 | 0.4819 | 0.4823 |
2.229 | 8.44 | 4300 | 2.1334 | 0.5214 | 0.4329 | 0.4809 | 0.4812 |
2.2762 | 8.64 | 4400 | 2.1223 | 0.5161 | 0.4288 | 0.4761 | 0.4769 |
2.2005 | 8.84 | 4500 | 2.1322 | 0.5197 | 0.4320 | 0.4793 | 0.4799 |
2.1975 | 9.03 | 4600 | 2.1294 | 0.5211 | 0.4338 | 0.4813 | 0.4819 |
2.3219 | 9.23 | 4700 | 2.1251 | 0.5148 | 0.4260 | 0.4768 | 0.4772 |
2.252 | 9.43 | 4800 | 2.1261 | 0.5211 | 0.4338 | 0.4813 | 0.4819 |
2.2594 | 9.62 | 4900 | 2.1236 | 0.5200 | 0.4331 | 0.4808 | 0.4814 |
2.1675 | 9.82 | 5000 | 2.1257 | 0.5211 | 0.4338 | 0.4813 | 0.4819 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1
- Downloads last month
- 2
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
๐
Ask for provider support
Model tree for aashish-249/mt5-summarize-te
Base model
google/mt5-small