NourFakih's picture
Model save
cc5d1dd verified
|
raw
history blame
6.53 kB
metadata
license: apache-2.0
base_model: nlpconnect/vit-gpt2-image-captioning
tags:
  - generated_from_trainer
metrics:
  - rouge
model-index:
  - name: image-captioning-Vit-GPT2-Flickr8k
    results: []

image-captioning-Vit-GPT2-Flickr8k

This model is a fine-tuned version of nlpconnect/vit-gpt2-image-captioning on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.4624
  • Rouge1: 38.4598
  • Rouge2: 14.1356
  • Rougel: 35.4001
  • Rougelsum: 35.4044
  • Gen Len: 12.1355

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: 4
  • eval_batch_size: 4
  • 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
0.5495 0.06 500 0.4942 35.0813 11.7169 32.4184 32.4273 11.5738
0.4945 0.12 1000 0.4903 35.4868 12.037 32.835 32.8388 11.8682
0.4984 0.19 1500 0.4862 35.3878 11.996 32.8196 32.8268 12.0544
0.4783 0.25 2000 0.4808 36.1063 12.3478 33.4632 33.4783 11.3468
0.4736 0.31 2500 0.4772 35.9266 12.3362 33.5046 33.5103 11.1066
0.4685 0.37 3000 0.4708 36.9089 13.0915 34.2896 34.2995 11.4739
0.4687 0.43 3500 0.4704 36.1844 12.5731 33.4609 33.4733 11.9201
0.4709 0.49 4000 0.4696 36.1774 12.8262 33.3824 33.3814 12.1733
0.4575 0.56 4500 0.4675 37.4417 13.7581 34.5386 34.5523 12.6302
0.4484 0.62 5000 0.4662 36.6864 13.0727 33.9056 33.9339 12.6007
0.4507 0.68 5500 0.4656 36.5144 12.7924 34.0484 34.0759 11.4316
0.4445 0.74 6000 0.4628 37.0553 13.3404 34.4096 34.4153 12.3211
0.4557 0.8 6500 0.4594 37.3241 13.1468 34.45 34.4658 12.2522
0.4451 0.87 7000 0.4600 37.33 13.5726 34.6534 34.6635 12.0494
0.4381 0.93 7500 0.4588 37.6255 13.8048 34.817 34.8252 12.1347
0.4357 0.99 8000 0.4571 37.2088 13.4177 34.3316 34.3372 12.2670
0.3869 1.05 8500 0.4612 37.7054 13.683 34.9637 34.9821 11.3216
0.377 1.11 9000 0.4616 37.2701 13.2182 34.3249 34.3396 12.3221
0.3736 1.17 9500 0.4607 37.2101 13.1285 34.3812 34.3767 11.8274
0.3801 1.24 10000 0.4617 37.9963 13.7537 35.2402 35.2374 11.6079
0.3816 1.3 10500 0.4599 37.3247 13.619 34.6494 34.6538 12.2101
0.377 1.36 11000 0.4619 37.2827 13.4471 34.3588 34.3861 12.3911
0.3745 1.42 11500 0.4604 37.5469 13.3948 34.5403 34.5613 12.2747
0.3785 1.48 12000 0.4568 38.085 14.0087 35.0549 35.0564 12.3179
0.3675 1.54 12500 0.4587 37.6241 13.8529 34.7614 34.7853 11.8732
0.3731 1.61 13000 0.4554 38.4418 14.1464 35.6658 35.6502 11.4294
0.3731 1.67 13500 0.4548 37.9045 13.7524 34.9001 34.9092 12.1241
0.371 1.73 14000 0.4542 38.412 14.212 35.473 35.4781 12.1014
0.3615 1.79 14500 0.4551 38.0734 14.1066 35.1289 35.1552 12.1135
0.3687 1.85 15000 0.4550 38.1762 14.1402 35.288 35.2936 12.2255
0.3711 1.92 15500 0.4532 37.6439 13.611 34.7558 34.7601 12.1632
0.3685 1.98 16000 0.4515 38.5682 14.5305 35.552 35.5703 11.9162
0.3333 2.04 16500 0.4626 38.4527 14.4649 35.6252 35.6307 11.9506
0.3129 2.1 17000 0.4660 38.203 14.0699 35.1626 35.1595 12.3313
0.3155 2.16 17500 0.4674 37.8903 13.9159 34.9097 34.9101 12.4853
0.3134 2.22 18000 0.4644 38.1489 13.9448 35.0351 35.0351 11.9748
0.3167 2.29 18500 0.4653 37.8449 13.9106 34.7773 34.7854 12.5273
0.322 2.35 19000 0.4673 37.9832 14.0115 34.8505 34.8597 12.4680
0.312 2.41 19500 0.4641 38.4627 14.2528 35.4297 35.4377 11.9315
0.3173 2.47 20000 0.4654 38.1591 13.9126 35.1114 35.1042 12.4845
0.3081 2.53 20500 0.4640 38.6969 14.3244 35.6933 35.692 11.8932
0.3093 2.6 21000 0.4633 38.2944 14.103 35.2407 35.2629 11.8932
0.3154 2.66 21500 0.4637 38.0668 13.7427 35.0547 35.0585 12.1310
0.3096 2.72 22000 0.4630 38.3647 14.0445 35.2568 35.2511 12.2591
0.3101 2.78 22500 0.4627 38.6366 14.3013 35.4955 35.4956 12.2836
0.309 2.84 23000 0.4620 38.3486 14.0403 35.3173 35.3265 12.3281
0.312 2.9 23500 0.4623 38.423 14.0759 35.3766 35.3853 12.2208
0.3135 2.97 24000 0.4624 38.4598 14.1356 35.4001 35.4044 12.1355

Framework versions

  • Transformers 4.39.3
  • Pytorch 2.1.2
  • Datasets 2.18.0
  • Tokenizers 0.15.2