4b0d9a249363f457ffaf67aac2a45515

This model is a fine-tuned version of albert/albert-xxlarge-v1 on the ccdv/patent-classification [abstract] dataset. It achieves the following results on the evaluation set:

• Loss: 1.1659
• Data Size: 1.0
• Epoch Runtime: 146.4085
• Accuracy: 0.6719
• F1 Macro: 0.6386

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: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• total_train_batch_size: 32
• total_eval_batch_size: 32
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Data Size	Epoch Runtime	Accuracy	F1 Macro
No log	0	0	2.6862	0	8.8151	0.1132	0.0627
No log	1	781	1.9710	0.0078	10.0274	0.2442	0.0987
No log	2	1562	1.8575	0.0156	11.1653	0.2710	0.1666
No log	3	2343	1.3425	0.0312	13.4674	0.5104	0.3718
0.0385	4	3124	1.1359	0.0625	17.6736	0.6098	0.5048
1.1559	5	3905	1.0844	0.125	26.2688	0.6112	0.5831
1.0724	6	4686	0.9974	0.25	43.3797	0.6544	0.5952
0.9218	7	5467	0.9227	0.5	77.5112	0.6877	0.6402
0.8342	8.0	6248	0.9700	1.0	146.1619	0.6765	0.6327
0.6908	9.0	7029	0.9583	1.0	145.8183	0.6861	0.6393
0.4756	10.0	7810	1.0832	1.0	145.9675	0.6779	0.6425
0.3715	11.0	8591	1.1659	1.0	146.4085	0.6719	0.6386

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.3.0
• Tokenizers 0.22.1