<!-- 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. -->
roberta-with-topic
This model is a fine-tuned version of roberta-large on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.5283
- Ndcg: 0.4453
- Accuracy: 0.2941
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: 16
- eval_batch_size: 16
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- total_train_batch_size: 64
- total_eval_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 2
Training results
| Training Loss | Epoch | Step | Validation Loss | Ndcg | Accuracy |
|---|---|---|---|---|---|
| 1.5951 | 0.07 | 413 | 1.5693 | 0.4220 | 0.2766 |
| 1.5721 | 0.13 | 826 | 1.5537 | 0.4308 | 0.2828 |
| 1.5594 | 0.2 | 1239 | 1.5615 | 0.4236 | 0.2757 |
| 1.5753 | 0.27 | 1652 | 1.5645 | 0.4272 | 0.2778 |
| 1.5778 | 0.33 | 2065 | 1.5859 | 0.3736 | 0.2430 |
| 1.5673 | 0.4 | 2478 | 1.5576 | 0.4262 | 0.2812 |
| 1.5633 | 0.47 | 2891 | 1.5557 | 0.4294 | 0.2815 |
| 1.5606 | 0.53 | 3304 | 1.5459 | 0.4321 | 0.2836 |
| 1.5476 | 0.6 | 3717 | 1.5508 | 0.4269 | 0.2810 |
| 1.552 | 0.67 | 4130 | 1.5479 | 0.4302 | 0.2831 |
| 1.5469 | 0.73 | 4543 | 1.5430 | 0.4345 | 0.2882 |
| 1.5538 | 0.8 | 4956 | 1.5410 | 0.4371 | 0.2877 |
| 1.557 | 0.87 | 5369 | 1.5420 | 0.4368 | 0.2896 |
| 1.5427 | 0.93 | 5782 | 1.5449 | 0.4269 | 0.2814 |
| 1.5427 | 1.0 | 6195 | 1.5381 | 0.4380 | 0.2896 |
| 1.5469 | 1.07 | 6608 | 1.5381 | 0.4362 | 0.2849 |
| 1.5369 | 1.13 | 7021 | 1.5361 | 0.4383 | 0.2895 |
| 1.5465 | 1.2 | 7434 | 1.5361 | 0.4415 | 0.2940 |
| 1.5433 | 1.27 | 7847 | 1.5342 | 0.4399 | 0.2914 |
| 1.5355 | 1.33 | 8260 | 1.5342 | 0.4409 | 0.2937 |
| 1.5363 | 1.4 | 8673 | 1.5342 | 0.4414 | 0.2923 |
| 1.5372 | 1.47 | 9086 | 1.5312 | 0.4440 | 0.2949 |
| 1.5452 | 1.53 | 9499 | 1.5303 | 0.4439 | 0.2937 |
| 1.5386 | 1.6 | 9912 | 1.5293 | 0.4434 | 0.2915 |
| 1.5314 | 1.67 | 10325 | 1.5303 | 0.4443 | 0.2925 |
| 1.5216 | 1.73 | 10738 | 1.5293 | 0.4447 | 0.2930 |
| 1.5341 | 1.8 | 11151 | 1.5293 | 0.4450 | 0.2929 |
| 1.5315 | 1.87 | 11564 | 1.5283 | 0.4456 | 0.2947 |
| 1.5345 | 1.93 | 11977 | 1.5283 | 0.4455 | 0.2950 |
| 1.5238 | 2.0 | 12390 | 1.5283 | 0.4453 | 0.2941 |
Framework versions
- Transformers 4.29.1
- Pytorch 2.0.0-rc1
- Datasets 2.12.0
- Tokenizers 0.13.3
