Image_Captioner / README.md
ChayanM's picture
Model save
451f6fd verified
|
raw
history blame
6.57 kB
metadata
tags:
  - generated_from_trainer
metrics:
  - rouge
model-index:
  - name: Image_Captioner
    results: []

Image_Captioner

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

  • Loss: 0.0975
  • Rouge1: 24.6871
  • Rouge2: 9.5762
  • Rougel: 20.8694
  • Rougelsum: 23.5961
  • Gen Len: 18.9192

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: 12
  • eval_batch_size: 12
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 50

Training results

Training Loss Epoch Step Validation Loss Rouge1 Rouge2 Rougel Rougelsum Gen Len
0.2412 1.0 558 0.1433 25.8229 12.3048 23.7573 25.4769 19.0
0.1504 2.0 1116 0.1213 25.6896 12.1913 23.6153 25.3363 19.0
0.132 3.0 1674 0.1099 26.1044 12.1738 23.7995 25.4389 19.0
0.1165 4.0 2232 0.1018 24.4958 11.6079 21.9255 24.1601 19.0
0.105 5.0 2790 0.0942 26.0341 12.2669 23.7171 25.6348 19.0
0.0942 6.0 3348 0.0874 25.3352 10.6409 22.3911 24.6749 19.0
0.0847 7.0 3906 0.0825 24.3455 11.1631 22.0193 23.8951 19.0
0.0764 8.0 4464 0.0782 25.4345 11.7359 23.0169 24.8774 19.0
0.0615 9.0 5022 0.0742 26.5655 12.5315 24.0201 26.0177 19.0
0.0546 10.0 5580 0.0714 26.984 10.977 23.3161 25.9544 19.0
0.0483 11.0 6138 0.0689 26.2815 11.5641 23.1829 25.4578 19.0
0.044 12.0 6696 0.0663 25.2328 11.3217 22.4545 24.653 19.0
0.0383 13.0 7254 0.0648 25.9672 10.9082 22.8064 25.1251 19.0
0.0351 14.0 7812 0.0660 26.0833 11.0382 22.6573 25.2428 19.0
0.0313 15.0 8370 0.0658 26.7009 10.9455 22.845 25.6707 19.0
0.0276 16.0 8928 0.0659 26.1769 10.6049 22.4903 25.321 19.0
0.0246 17.0 9486 0.0661 26.1478 10.3981 22.4809 25.1318 19.0
0.0197 18.0 10044 0.0682 25.3438 10.4852 22.1182 24.5048 19.0
0.0178 19.0 10602 0.0689 25.2217 9.6912 21.4433 24.1131 19.0
0.0159 20.0 11160 0.0707 24.6521 9.4214 21.0511 23.6559 18.9771
0.0148 21.0 11718 0.0725 24.6477 9.8393 21.2375 23.8722 18.9933
0.0133 22.0 12276 0.0723 25.2483 9.9924 21.6189 24.2124 18.9933
0.0121 23.0 12834 0.0741 24.3834 9.8801 21.0599 23.5916 18.9664
0.0114 24.0 13392 0.0757 25.0727 9.7857 21.3046 24.0167 18.9758
0.0103 25.0 13950 0.0774 25.1959 10.1108 21.5608 24.1292 18.9098
0.0089 26.0 14508 0.0783 25.5931 9.9812 21.6953 24.3561 18.9219
0.0083 27.0 15066 0.0793 24.8603 10.0231 21.2615 23.9145 18.9879
0.0076 28.0 15624 0.0802 24.741 9.6977 21.112 23.8097 18.9367
0.0074 29.0 16182 0.0812 24.0656 9.4335 20.6021 23.0172 18.8748
0.0067 30.0 16740 0.0838 24.9923 9.9583 21.2749 23.9427 18.9556
0.0063 31.0 17298 0.0844 24.8869 9.6309 21.0218 23.7523 18.8789
0.0058 32.0 17856 0.0870 24.8009 9.9887 21.0596 23.71 18.9139
0.0054 33.0 18414 0.0879 24.9076 9.663 21.0755 23.8641 18.9287
0.0052 34.0 18972 0.0902 25.0668 9.5739 21.2282 23.9928 18.9044
0.0044 35.0 19530 0.0908 25.0616 10.0034 21.6482 23.9978 18.9098
0.0041 36.0 20088 0.0912 25.0681 10.099 21.4527 24.0219 18.8573
0.0039 37.0 20646 0.0916 24.6263 9.7547 20.8695 23.5722 18.9367
0.0037 38.0 21204 0.0922 24.6973 9.6421 21.1171 23.733 18.9435
0.0034 39.0 21762 0.0929 25.3821 9.8435 21.4803 24.3296 18.8439
0.0032 40.0 22320 0.0944 25.2386 9.9245 21.4207 24.1773 18.9287
0.003 41.0 22878 0.0947 25.2413 10.0581 21.5136 24.146 18.9623
0.0028 42.0 23436 0.0958 25.1041 9.8452 21.2494 24.0197 18.9166
0.0027 43.0 23994 0.0960 24.8 9.932 21.1541 23.7546 18.9529
0.0024 44.0 24552 0.0965 25.1426 10.0351 21.3824 24.0439 18.9341
0.0023 45.0 25110 0.0964 24.837 9.7853 21.0592 23.7822 18.9421
0.0022 46.0 25668 0.0968 24.8325 9.8007 20.976 23.7022 18.9596
0.0021 47.0 26226 0.0972 24.7028 9.6921 21.0038 23.6658 18.9394
0.0021 48.0 26784 0.0974 24.7233 9.77 21.0449 23.6333 18.9260
0.002 49.0 27342 0.0977 24.7481 9.6823 20.9024 23.6522 18.9300
0.0019 50.0 27900 0.0975 24.6871 9.5762 20.8694 23.5961 18.9192

Framework versions

  • Transformers 4.37.1
  • Pytorch 1.13.1+cu117
  • Datasets 2.15.0
  • Tokenizers 0.15.1