Hugging Face's logo

NiamaLynn
/
lilt-roberta-DocLayNet-base_lines_ml256-v1

Token Classification
Transformers
TensorBoard
Safetensors
lilt
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community

lilt-roberta-DocLayNet-base_lines_ml256-v1

This model is a fine-tuned version of nielsr/lilt-xlm-roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.9004
  • Precision: 0.8622
  • Recall: 0.8622
  • F1: 0.8622
  • Accuracy: 0.8622

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: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 5
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
No log 0.07 300 0.7371 0.6945 0.6945 0.6945 0.6945
0.7701 0.14 600 0.8573 0.7488 0.7488 0.7488 0.7488
0.7701 0.21 900 0.7687 0.7606 0.7606 0.7606 0.7606
0.471 0.27 1200 0.7057 0.7750 0.7750 0.7750 0.7750
0.4183 0.34 1500 0.6305 0.7961 0.7961 0.7961 0.7961
0.4183 0.41 1800 0.7039 0.7769 0.7769 0.7769 0.7769
0.3683 0.48 2100 0.5956 0.7980 0.7980 0.7980 0.7980
0.3683 0.55 2400 0.7312 0.7864 0.7864 0.7864 0.7864
0.3429 0.62 2700 0.5868 0.8049 0.8049 0.8049 0.8049
0.3337 0.69 3000 0.5911 0.8010 0.8010 0.8010 0.8010
0.3337 0.76 3300 0.7278 0.7893 0.7893 0.7893 0.7893
0.3056 0.82 3600 0.8030 0.7908 0.7908 0.7908 0.7908
0.3056 0.89 3900 0.6587 0.7978 0.7978 0.7978 0.7978
0.2772 0.96 4200 0.5334 0.8315 0.8315 0.8315 0.8315
0.2456 1.03 4500 0.6787 0.7992 0.7992 0.7992 0.7992
0.2456 1.1 4800 0.7325 0.8037 0.8037 0.8037 0.8037
0.2183 1.17 5100 0.7280 0.7985 0.7985 0.7985 0.7985
0.2183 1.24 5400 0.9041 0.7787 0.7787 0.7787 0.7787
0.2288 1.31 5700 0.7504 0.8076 0.8076 0.8076 0.8076
0.2228 1.37 6000 0.6672 0.8042 0.8042 0.8042 0.8042
0.2228 1.44 6300 0.5468 0.8511 0.8511 0.8511 0.8511
0.1989 1.51 6600 0.5928 0.8229 0.8229 0.8229 0.8229
0.1989 1.58 6900 0.6731 0.8150 0.8150 0.8150 0.8150
0.2062 1.65 7200 0.7504 0.8030 0.8030 0.8030 0.8030
0.1971 1.72 7500 0.6554 0.8255 0.8255 0.8255 0.8255
0.1971 1.79 7800 0.7095 0.8046 0.8046 0.8046 0.8046
0.1929 1.86 8100 0.6244 0.8397 0.8397 0.8397 0.8397
0.1929 1.92 8400 0.8521 0.8067 0.8067 0.8067 0.8067
0.1788 1.99 8700 0.7261 0.8088 0.8088 0.8088 0.8088
0.1631 2.06 9000 0.6650 0.8272 0.8272 0.8272 0.8272
0.1631 2.13 9300 0.8314 0.8142 0.8142 0.8142 0.8142
0.1284 2.2 9600 0.9010 0.8113 0.8113 0.8113 0.8113
0.1284 2.27 9900 0.9008 0.8087 0.8087 0.8087 0.8087
0.1248 2.34 10200 0.9152 0.8065 0.8065 0.8065 0.8065
0.1365 2.4 10500 0.6791 0.8393 0.8393 0.8393 0.8393
0.1365 2.47 10800 0.7301 0.8185 0.8185 0.8185 0.8185
0.1194 2.54 11100 0.8937 0.8050 0.8050 0.8050 0.8050
0.1194 2.61 11400 0.7559 0.8293 0.8293 0.8293 0.8293
0.1282 2.68 11700 0.7903 0.8163 0.8163 0.8163 0.8163
0.1234 2.75 12000 1.0103 0.8090 0.8090 0.8090 0.8090
0.1234 2.82 12300 0.9975 0.8096 0.8096 0.8096 0.8096
0.1104 2.89 12600 0.8443 0.8171 0.8171 0.8171 0.8171
0.1104 2.95 12900 0.8380 0.8125 0.8125 0.8125 0.8125
0.1254 3.02 13200 0.8283 0.8223 0.8223 0.8223 0.8223
0.0806 3.09 13500 0.9232 0.8323 0.8323 0.8323 0.8323
0.0806 3.16 13800 1.0903 0.8176 0.8176 0.8176 0.8176
0.0875 3.23 14100 1.0781 0.8110 0.8110 0.8110 0.8110
0.0875 3.3 14400 0.8806 0.8277 0.8277 0.8277 0.8277
0.0817 3.37 14700 1.0024 0.8212 0.8212 0.8212 0.8212
0.085 3.44 15000 0.9078 0.8294 0.8294 0.8294 0.8294
0.085 3.5 15300 0.8745 0.8377 0.8377 0.8377 0.8377
0.0784 3.57 15600 0.9590 0.8329 0.8329 0.8329 0.8329
0.0784 3.64 15900 0.8027 0.8500 0.8500 0.8500 0.8500
0.0785 3.71 16200 1.0033 0.8171 0.8171 0.8171 0.8171
0.0756 3.78 16500 0.8017 0.8446 0.8446 0.8446 0.8446
0.0756 3.85 16800 1.0721 0.8162 0.8162 0.8162 0.8162
0.078 3.92 17100 1.1095 0.8172 0.8172 0.8172 0.8172
0.078 3.99 17400 1.0099 0.8200 0.8200 0.8200 0.8200
0.0696 4.05 17700 1.0189 0.8249 0.8249 0.8249 0.8249
0.0456 4.12 18000 1.2109 0.8165 0.8165 0.8165 0.8165
0.0456 4.19 18300 1.0789 0.8273 0.8273 0.8273 0.8273
0.0587 4.26 18600 1.0981 0.8277 0.8277 0.8277 0.8277
0.0587 4.33 18900 1.0236 0.8395 0.8395 0.8395 0.8395
0.0485 4.4 19200 0.9660 0.8381 0.8381 0.8381 0.8381
0.056 4.47 19500 0.9447 0.8453 0.8453 0.8453 0.8453
0.056 4.54 19800 0.9226 0.8564 0.8564 0.8564 0.8564
0.0517 4.6 20100 1.1416 0.8313 0.8313 0.8313 0.8313
0.0517 4.67 20400 0.9004 0.8622 0.8622 0.8622 0.8622
0.0555 4.74 20700 1.0452 0.8416 0.8416 0.8416 0.8416
0.0578 4.81 21000 0.9997 0.8480 0.8480 0.8480 0.8480
0.0578 4.88 21300 1.0441 0.8402 0.8402 0.8402 0.8402
0.0495 4.95 21600 1.0393 0.8421 0.8421 0.8421 0.8421

Framework versions

  • Transformers 4.36.0
  • Pytorch 2.0.0
  • Datasets 2.1.0
  • Tokenizers 0.15.0
Downloads last month
28
Safetensors
Model size
284M params
Tensor type
F32
·
Inference Examples
Token Classification
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for NiamaLynn/lilt-roberta-DocLayNet-base_lines_ml256-v1

Base model

nielsr/lilt-xlm-roberta-base
Finetuned
(29)
this model

Space using NiamaLynn/lilt-roberta-DocLayNet-base_lines_ml256-v1 1

Evaluation results

Metadata error: specify a dataset to view leaderboard