metadata
license: apache-2.0
base_model: facebook/bart-base
tags:
- generated_from_trainer
model-index:
- name: bart-with-pubmed-noise-data-0.1
results: []
bart-with-pubmed-noise-data-0.1
This model is a fine-tuned version of facebook/bart-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1956
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: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 10
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.3905 | 0.04 | 500 | 0.3471 |
| 0.3136 | 0.07 | 1000 | 0.3261 |
| 0.2971 | 0.11 | 1500 | 0.2971 |
| 0.3361 | 0.14 | 2000 | 0.2788 |
| 0.2502 | 0.18 | 2500 | 0.2780 |
| 0.2613 | 0.21 | 3000 | 0.2690 |
| 0.2683 | 0.25 | 3500 | 0.2591 |
| 0.2995 | 0.29 | 4000 | 0.2539 |
| 0.2317 | 0.32 | 4500 | 0.2481 |
| 0.2361 | 0.36 | 5000 | 0.2453 |
| 0.2523 | 0.39 | 5500 | 0.2440 |
| 0.236 | 0.43 | 6000 | 0.2391 |
| 0.2301 | 0.46 | 6500 | 0.2372 |
| 0.2259 | 0.5 | 7000 | 0.2328 |
| 0.2231 | 0.54 | 7500 | 0.2344 |
| 0.2098 | 0.57 | 8000 | 0.2285 |
| 0.2663 | 0.61 | 8500 | 0.2220 |
| 0.2139 | 0.64 | 9000 | 0.2265 |
| 0.2372 | 0.68 | 9500 | 0.2204 |
| 0.1946 | 0.71 | 10000 | 0.2213 |
| 0.1843 | 0.75 | 10500 | 0.2214 |
| 0.1872 | 0.79 | 11000 | 0.2178 |
| 0.2182 | 0.82 | 11500 | 0.2127 |
| 0.2123 | 0.86 | 12000 | 0.2118 |
| 0.1865 | 0.89 | 12500 | 0.2113 |
| 0.1782 | 0.93 | 13000 | 0.2080 |
| 0.1894 | 0.96 | 13500 | 0.2053 |
| 0.1989 | 1.0 | 14000 | 0.2097 |
| 0.1721 | 1.03 | 14500 | 0.2083 |
| 0.1353 | 1.07 | 15000 | 0.2102 |
| 0.164 | 1.11 | 15500 | 0.2140 |
| 0.1541 | 1.14 | 16000 | 0.2086 |
| 0.1421 | 1.18 | 16500 | 0.2112 |
| 0.1752 | 1.21 | 17000 | 0.2085 |
| 0.1452 | 1.25 | 17500 | 0.2105 |
| 0.1836 | 1.28 | 18000 | 0.2066 |
| 0.1444 | 1.32 | 18500 | 0.2083 |
| 0.1473 | 1.36 | 19000 | 0.2090 |
| 0.1723 | 1.39 | 19500 | 0.2084 |
| 0.1328 | 1.43 | 20000 | 0.2042 |
| 0.1842 | 1.46 | 20500 | 0.2032 |
| 0.1934 | 1.5 | 21000 | 0.2031 |
| 0.1412 | 1.53 | 21500 | 0.2008 |
| 0.1302 | 1.57 | 22000 | 0.2003 |
| 0.142 | 1.61 | 22500 | 0.2008 |
| 0.1479 | 1.64 | 23000 | 0.2025 |
| 0.1628 | 1.68 | 23500 | 0.2005 |
| 0.1126 | 1.71 | 24000 | 0.2016 |
| 0.1515 | 1.75 | 24500 | 0.1985 |
| 0.1605 | 1.78 | 25000 | 0.1984 |
| 0.1659 | 1.82 | 25500 | 0.1970 |
| 0.1404 | 1.86 | 26000 | 0.1980 |
| 0.1386 | 1.89 | 26500 | 0.1972 |
| 0.1119 | 1.93 | 27000 | 0.1976 |
| 0.168 | 1.96 | 27500 | 0.1940 |
| 0.1318 | 2.0 | 28000 | 0.1958 |
| 0.1307 | 2.03 | 28500 | 0.1987 |
| 0.1312 | 2.07 | 29000 | 0.2012 |
| 0.1237 | 2.11 | 29500 | 0.2002 |
| 0.1339 | 2.14 | 30000 | 0.2010 |
| 0.1471 | 2.18 | 30500 | 0.1999 |
| 0.1195 | 2.21 | 31000 | 0.1998 |
| 0.1002 | 2.25 | 31500 | 0.2000 |
| 0.1009 | 2.28 | 32000 | 0.2012 |
| 0.1608 | 2.32 | 32500 | 0.1995 |
| 0.1198 | 2.36 | 33000 | 0.2009 |
| 0.1053 | 2.39 | 33500 | 0.1990 |
| 0.1399 | 2.43 | 34000 | 0.2001 |
| 0.1043 | 2.46 | 34500 | 0.1994 |
| 0.1254 | 2.5 | 35000 | 0.1996 |
| 0.0987 | 2.53 | 35500 | 0.1966 |
| 0.119 | 2.57 | 36000 | 0.1974 |
| 0.1167 | 2.61 | 36500 | 0.1983 |
| 0.1119 | 2.64 | 37000 | 0.1974 |
| 0.1391 | 2.68 | 37500 | 0.1973 |
| 0.1036 | 2.71 | 38000 | 0.1971 |
| 0.1203 | 2.75 | 38500 | 0.1976 |
| 0.1498 | 2.78 | 39000 | 0.1976 |
| 0.1037 | 2.82 | 39500 | 0.1975 |
| 0.1141 | 2.85 | 40000 | 0.1961 |
| 0.0935 | 2.89 | 40500 | 0.1960 |
| 0.0985 | 2.93 | 41000 | 0.1963 |
| 0.108 | 2.96 | 41500 | 0.1955 |
| 0.1054 | 3.0 | 42000 | 0.1956 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.2+cu121
- Datasets 2.17.0
- Tokenizers 0.15.1