File size: 8,049 Bytes
638b325 195c5d9 638b325 195c5d9 638b325 195c5d9 638b325 195c5d9 638b325 195c5d9 638b325 195c5d9 638b325 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 |
---
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- ccdv/arxiv-summarization
metrics:
- rouge
model-index:
- name: arxiv-summarization-fb-bart-base-2022-09-21
results:
- task:
name: Summarization
type: summarization
dataset:
name: ccdv/arxiv-summarization
type: ccdv/arxiv-summarization
config: section
split: train
args: section
metrics:
- name: Rouge1
type: rouge
value: 42.9082
---
<!-- 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. -->
# arxiv-summarization-fb-bart-base-2022-09-21
This model is a fine-tuned version of [facebook/bart-base](https://huggingface.co/facebook/bart-base) on the ccdv/arxiv-summarization dataset.
It achieves the following results on the evaluation set:
- Loss: 2.1597
- Rouge1: 42.9082
- Rouge2: 15.7763
- Rougel: 25.9239
- Rougelsum: 37.7957
- Gen Len: 110.5816
## 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: 1
- eval_batch_size: 1
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3.0
### Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|:-------------:|:-----:|:------:|:---------------:|:-------:|:------:|:-------:|:---------:|:-------:|
| 2.9142 | 0.05 | 10000 | 2.7522 | 17.073 | 6.7502 | 13.6779 | 15.6668 | 20.0 |
| 2.7876 | 0.1 | 20000 | 2.6888 | 16.7954 | 6.7038 | 13.4939 | 15.3426 | 19.9992 |
| 2.715 | 0.15 | 30000 | 2.6308 | 17.3324 | 6.8771 | 13.7918 | 15.7839 | 20.0 |
| 2.6431 | 0.2 | 40000 | 2.5858 | 16.7055 | 6.8108 | 13.4796 | 15.2959 | 20.0 |
| 2.6381 | 0.25 | 50000 | 2.5393 | 17.4643 | 7.0687 | 13.9507 | 16.012 | 20.0 |
| 2.6269 | 0.3 | 60000 | 2.5159 | 17.5934 | 7.0022 | 13.9394 | 16.0203 | 20.0 |
| 2.5482 | 0.34 | 70000 | 2.4894 | 17.5428 | 7.1822 | 13.9788 | 16.0355 | 20.0 |
| 2.4962 | 0.39 | 80000 | 2.4476 | 17.3587 | 7.1501 | 13.9215 | 15.8637 | 20.0 |
| 2.513 | 0.44 | 90000 | 2.4309 | 18.0806 | 7.5429 | 14.4201 | 16.561 | 20.0 |
| 2.4464 | 0.49 | 100000 | 2.4128 | 17.9813 | 7.5454 | 14.3403 | 16.52 | 19.9989 |
| 2.4969 | 0.54 | 110000 | 2.4114 | 17.3353 | 7.1382 | 13.9109 | 15.873 | 20.0 |
| 2.4417 | 0.59 | 120000 | 2.3866 | 18.0241 | 7.553 | 14.3892 | 16.5077 | 19.9980 |
| 2.4333 | 0.64 | 130000 | 2.3903 | 18.0578 | 7.4999 | 14.3901 | 16.5134 | 20.0 |
| 2.4296 | 0.69 | 140000 | 2.3793 | 17.7742 | 7.5182 | 14.2794 | 16.2879 | 20.0 |
| 2.4277 | 0.74 | 150000 | 2.3571 | 17.8015 | 7.4677 | 14.226 | 16.3288 | 20.0 |
| 2.4258 | 0.79 | 160000 | 2.3539 | 17.5335 | 7.399 | 14.09 | 16.0936 | 20.0 |
| 2.4006 | 0.84 | 170000 | 2.3469 | 17.5983 | 7.4285 | 14.1315 | 16.1385 | 20.0 |
| 2.367 | 0.89 | 180000 | 2.3344 | 17.297 | 7.2361 | 13.9286 | 15.8352 | 20.0 |
| 2.373 | 0.94 | 190000 | 2.3377 | 17.7189 | 7.4993 | 14.2603 | 16.2546 | 19.9980 |
| 2.3762 | 0.99 | 200000 | 2.3106 | 17.7883 | 7.4766 | 14.2675 | 16.3115 | 20.0 |
| 2.2538 | 1.03 | 210000 | 2.3197 | 17.4487 | 7.4171 | 14.0473 | 15.9771 | 20.0 |
| 2.268 | 1.08 | 220000 | 2.3044 | 17.9603 | 7.5806 | 14.3755 | 16.4328 | 20.0 |
| 2.2986 | 1.13 | 230000 | 2.3002 | 17.9268 | 7.5321 | 14.3503 | 16.4191 | 20.0 |
| 2.241 | 1.18 | 240000 | 2.3059 | 17.4542 | 7.3224 | 14.0578 | 16.0157 | 20.0 |
| 2.2534 | 1.23 | 250000 | 2.2927 | 17.8039 | 7.6232 | 14.2916 | 16.3442 | 20.0 |
| 2.26 | 1.28 | 260000 | 2.2910 | 17.8607 | 7.5645 | 14.318 | 16.3336 | 19.9983 |
| 2.3 | 1.33 | 270000 | 2.2818 | 17.8203 | 7.4815 | 14.3171 | 16.3309 | 20.0 |
| 2.2964 | 1.38 | 280000 | 2.2721 | 17.983 | 7.6867 | 14.3971 | 16.493 | 20.0 |
| 2.2564 | 1.43 | 290000 | 2.2701 | 18.059 | 7.7273 | 14.4806 | 16.5792 | 19.9988 |
| 2.2576 | 1.48 | 300000 | 2.2663 | 17.5706 | 7.4424 | 14.1424 | 16.1297 | 20.0 |
| 2.2605 | 1.53 | 310000 | 2.2607 | 17.8057 | 7.5219 | 14.3226 | 16.3355 | 19.9988 |
| 2.2587 | 1.58 | 320000 | 2.2552 | 18.0396 | 7.7064 | 14.5005 | 16.5823 | 20.0 |
| 2.2423 | 1.63 | 330000 | 2.2523 | 18.2229 | 7.8398 | 14.5868 | 16.7408 | 20.0 |
| 2.2793 | 1.68 | 340000 | 2.2431 | 17.6785 | 7.5437 | 14.1971 | 16.1724 | 19.9988 |
| 2.2005 | 1.72 | 350000 | 2.2343 | 17.7552 | 7.6026 | 14.2152 | 16.2797 | 19.9988 |
| 2.2454 | 1.77 | 360000 | 2.2339 | 17.9292 | 7.699 | 14.4099 | 16.4682 | 20.0 |
| 2.2175 | 1.82 | 370000 | 2.2345 | 17.7413 | 7.4892 | 14.2223 | 16.2442 | 20.0 |
| 2.238 | 1.87 | 380000 | 2.2259 | 17.6679 | 7.4976 | 14.24 | 16.243 | 19.9988 |
| 2.2108 | 1.92 | 390000 | 2.2210 | 17.8474 | 7.6054 | 14.3494 | 16.3635 | 19.9988 |
| 2.2124 | 1.97 | 400000 | 2.2170 | 17.8019 | 7.5182 | 14.264 | 16.3003 | 20.0 |
| 2.0976 | 2.02 | 410000 | 2.2248 | 17.8063 | 7.5383 | 14.2782 | 16.275 | 20.0 |
| 2.0932 | 2.07 | 420000 | 2.2196 | 17.9171 | 7.6187 | 14.3508 | 16.4333 | 20.0 |
| 2.0956 | 2.12 | 430000 | 2.2135 | 18.0616 | 7.7655 | 14.4837 | 16.5627 | 19.9988 |
| 2.0515 | 2.17 | 440000 | 2.2091 | 18.0281 | 7.7301 | 14.4696 | 16.5196 | 19.9981 |
| 2.1216 | 2.22 | 450000 | 2.2015 | 18.0609 | 7.7541 | 14.4633 | 16.5705 | 19.9988 |
| 2.1222 | 2.27 | 460000 | 2.1983 | 18.0717 | 7.7473 | 14.4725 | 16.5399 | 19.9988 |
| 2.0903 | 2.32 | 470000 | 2.2007 | 18.0751 | 7.7486 | 14.4583 | 16.546 | 20.0 |
| 2.1124 | 2.37 | 480000 | 2.1934 | 17.888 | 7.7124 | 14.3899 | 16.3901 | 20.0 |
| 2.1094 | 2.41 | 490000 | 2.1901 | 18.0254 | 7.7682 | 14.4427 | 16.5181 | 20.0 |
| 2.1085 | 2.46 | 500000 | 2.1924 | 17.9077 | 7.7004 | 14.3843 | 16.4221 | 19.9988 |
| 2.0781 | 2.51 | 510000 | 2.1781 | 18.1591 | 7.8456 | 14.565 | 16.6435 | 19.9988 |
| 2.0875 | 2.56 | 520000 | 2.1801 | 18.0389 | 7.7342 | 14.4259 | 16.5378 | 20.0 |
| 2.0945 | 2.61 | 530000 | 2.1758 | 18.0999 | 7.8217 | 14.5163 | 16.5784 | 19.9988 |
| 2.0723 | 2.66 | 540000 | 2.1756 | 17.9684 | 7.7369 | 14.4279 | 16.4815 | 19.9988 |
| 2.0918 | 2.71 | 550000 | 2.1738 | 18.1183 | 7.8414 | 14.5298 | 16.6119 | 19.9988 |
| 2.0835 | 2.76 | 560000 | 2.1671 | 17.8837 | 7.7379 | 14.3727 | 16.4068 | 19.9988 |
| 2.0936 | 2.81 | 570000 | 2.1670 | 17.9631 | 7.7708 | 14.4566 | 16.4823 | 19.9988 |
| 2.0518 | 2.86 | 580000 | 2.1631 | 18.0601 | 7.8112 | 14.5158 | 16.5816 | 19.9988 |
| 2.065 | 2.91 | 590000 | 2.1611 | 18.0548 | 7.8147 | 14.5271 | 16.5606 | 19.9988 |
| 2.0427 | 2.96 | 600000 | 2.1611 | 18.0642 | 7.8284 | 14.5293 | 16.5736 | 19.9988 |
### Framework versions
- Transformers 4.23.0.dev0
- Pytorch 1.12.0
- Datasets 2.5.1
- Tokenizers 0.13.0
|