metadata

license: cc-by-nc-sa-4.0
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: layoutlmv3-finetuned-funsd
    results: []

layoutlmv3-finetuned-funsd

This model is a fine-tuned version of microsoft/layoutlmv3-base on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.6016
Precision: 0.8712
Recall: 0.903
F1: 0.8868
Accuracy: 0.8414

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: 1e-05
train_batch_size: 10
eval_batch_size: 10
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
training_steps: 400

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.67	25	1.1945	0.4801	0.5295	0.5036	0.6659
No log	3.33	50	0.7770	0.7472	0.807	0.7760	0.7687
No log	5.0	75	0.5990	0.7965	0.8515	0.8231	0.8119
No log	6.67	100	0.6814	0.8061	0.879	0.8409	0.7702
No log	8.33	125	0.5392	0.8172	0.8785	0.8467	0.8316
No log	10.0	150	0.5438	0.8352	0.887	0.8603	0.8114
No log	11.67	175	0.5430	0.8586	0.8925	0.8752	0.8382
No log	13.33	200	0.5897	0.8521	0.887	0.8692	0.8258
No log	15.0	225	0.5427	0.86	0.903	0.8810	0.8430
No log	16.67	250	0.5542	0.8713	0.8935	0.8823	0.8411
No log	18.33	275	0.6014	0.8632	0.902	0.8822	0.8321
No log	20.0	300	0.5971	0.8691	0.9	0.8843	0.8366
No log	21.67	325	0.5983	0.8800	0.8985	0.8892	0.8313
No log	23.33	350	0.5867	0.8842	0.909	0.8964	0.8414
No log	25.0	375	0.5948	0.8797	0.9065	0.8929	0.8466
No log	26.67	400	0.6016	0.8712	0.903	0.8868	0.8414

Framework versions

Transformers 4.12.5
Pytorch 1.10.0+cu111
Datasets 2.13.2
Tokenizers 0.10.1