furina_seed42_eng_esp_hau

This model is a fine-tuned version of yihongLiu/furina on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0133
• Spearman Corr: 0.8711

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: 32
• eval_batch_size: 128
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Spearman Corr
No log	0.54	200	0.0318	0.6526
No log	1.08	400	0.0280	0.7226
No log	1.62	600	0.0253	0.7469
0.0472	2.15	800	0.0230	0.7583
0.0472	2.69	1000	0.0219	0.7738
0.0472	3.23	1200	0.0222	0.7832
0.0472	3.77	1400	0.0206	0.7891
0.0218	4.31	1600	0.0207	0.7936
0.0218	4.85	1800	0.0198	0.8013
0.0218	5.38	2000	0.0187	0.8112
0.0218	5.92	2200	0.0192	0.8131
0.0153	6.46	2400	0.0180	0.8233
0.0153	7.0	2600	0.0193	0.8265
0.0153	7.54	2800	0.0183	0.8305
0.0111	8.08	3000	0.0171	0.8339
0.0111	8.61	3200	0.0175	0.8337
0.0111	9.15	3400	0.0166	0.8350
0.0111	9.69	3600	0.0165	0.8362
0.0085	10.23	3800	0.0161	0.8436
0.0085	10.77	4000	0.0169	0.8442
0.0085	11.31	4200	0.0164	0.8414
0.0085	11.84	4400	0.0152	0.8492
0.0068	12.38	4600	0.0152	0.8477
0.0068	12.92	4800	0.0158	0.8512
0.0068	13.46	5000	0.0150	0.8542
0.0057	14.0	5200	0.0149	0.8564
0.0057	14.54	5400	0.0159	0.8555
0.0057	15.07	5600	0.0146	0.8581
0.0057	15.61	5800	0.0143	0.8576
0.0049	16.15	6000	0.0143	0.8587
0.0049	16.69	6200	0.0146	0.8593
0.0049	17.23	6400	0.0158	0.8597
0.0049	17.77	6600	0.0140	0.8611
0.0043	18.3	6800	0.0147	0.8610
0.0043	18.84	7000	0.0141	0.8634
0.0043	19.38	7200	0.0143	0.8635
0.0043	19.92	7400	0.0143	0.8647
0.0038	20.46	7600	0.0138	0.8650
0.0038	21.0	7800	0.0142	0.8657
0.0038	21.53	8000	0.0138	0.8655
0.0035	22.07	8200	0.0142	0.8663
0.0035	22.61	8400	0.0137	0.8664
0.0035	23.15	8600	0.0138	0.8669
0.0035	23.69	8800	0.0140	0.8694
0.0033	24.23	9000	0.0134	0.8690
0.0033	24.76	9200	0.0146	0.8683
0.0033	25.3	9400	0.0138	0.8678
0.0033	25.84	9600	0.0134	0.8701
0.003	26.38	9800	0.0136	0.8702
0.003	26.92	10000	0.0135	0.8713
0.003	27.46	10200	0.0133	0.8704
0.0029	27.99	10400	0.0135	0.8714
0.0029	28.53	10600	0.0133	0.8711

Framework versions

• Transformers 4.37.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.1