xmod-shared-roberta-base-legal-multi-downstream-build_rr

This model is a fine-tuned version of MHGanainy/xmod-shared-roberta-base-legal-multi on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.9259
• Precision-macro: 0.6337
• Recall-macro: 0.5884
• Macro-f1: 0.6021
• Precision-micro: 0.7895
• Recall-micro: 0.7895
• Micro-f1: 0.7895
• Accuracy: 0.7895

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: 3e-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 1
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision-macro	Recall-macro	Macro-f1	Precision-micro	Recall-micro	Micro-f1	Accuracy
No log	1.0	124	0.8663	0.5370	0.4683	0.4780	0.7461	0.7461	0.7461	0.7461
No log	2.0	248	0.9013	0.4938	0.5409	0.5023	0.6982	0.6982	0.6982	0.6982
No log	3.0	372	0.7797	0.5765	0.5451	0.5415	0.7635	0.7635	0.7635	0.7635
No log	4.0	496	0.7203	0.6530	0.5478	0.5409	0.7718	0.7718	0.7718	0.7718
0.9675	5.0	620	0.7465	0.5984	0.5960	0.5866	0.7777	0.7777	0.7777	0.7777
0.9675	6.0	744	0.7503	0.6134	0.5692	0.5699	0.7791	0.7791	0.7791	0.7791
0.9675	7.0	868	0.7665	0.6552	0.5732	0.5877	0.7864	0.7864	0.7864	0.7864
0.9675	8.0	992	0.7651	0.6253	0.5880	0.5937	0.7926	0.7926	0.7926	0.7926
0.5065	9.0	1116	0.8560	0.6075	0.5930	0.5945	0.7767	0.7767	0.7767	0.7767
0.5065	10.0	1240	0.8643	0.6354	0.5842	0.5972	0.7902	0.7902	0.7902	0.7902
0.5065	11.0	1364	0.9259	0.6337	0.5884	0.6021	0.7895	0.7895	0.7895	0.7895

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1