lamma-weather-classifier-improved

This model is a fine-tuned version of meta-llama/Llama-3.2-11B-Vision-Instruct on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0062

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: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 12

Training results

Training Loss	Epoch	Step	Validation Loss
0.0612	0.9983	147	0.0519
0.007	1.9966	294	0.0066
0.0064	2.9949	441	0.0064
0.0065	4.0	589	0.0063
0.0064	4.9983	736	0.0063
0.0064	5.9966	883	0.0064
0.0063	6.9949	1030	0.0063
0.0064	8.0	1178	0.0062
0.0061	8.9983	1325	0.0062
0.0066	9.9966	1472	0.0063
0.0063	10.9949	1619	0.0062
0.0064	11.9796	1764	0.0062

Framework versions

• PEFT 0.13.0
• Transformers 4.45.1
• Pytorch 2.4.0+cu121
• Datasets 3.0.1
• Tokenizers 0.20.3