gpad-v22-taskA-entropy-only-smpl

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3891
• Accuracy: 0.86
• F1 Macro: 0.8727
• F1 Weighted: 0.8592
• Precision Macro: 0.8276
• Recall Macro: 0.9231

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 24
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Macro	F1 Weighted	Precision Macro	Recall Macro
No log	0.9231	6	0.6606	0.5	0.0741	0.3541	1.0	0.0385
No log	2.0	13	0.5480	0.82	0.8235	0.8201	0.84	0.8077
No log	2.9231	19	0.4500	0.84	0.8519	0.8395	0.8214	0.8846
No log	4.0	26	0.3971	0.86	0.8727	0.8592	0.8276	0.9231
No log	4.6154	30	0.3891	0.86	0.8727	0.8592	0.8276	0.9231

Framework versions

• Transformers 4.41.0
• Pytorch 2.8.0+cu126
• Datasets 4.3.0
• Tokenizers 0.19.1