image-description_to_emotion

This model is a fine-tuned version of HuggingFaceTB/SmolLM2-360M on the arrow dataset. It achieves the following results on the evaluation set:

• Loss: 0.1650

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: 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: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 4
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.5979	0.3361	50	0.5039
0.3262	0.6723	100	0.2783
0.2599	1.0084	150	0.2305
0.2211	1.3445	200	0.2071
0.2004	1.6807	250	0.1969
0.2094	2.0168	300	0.1840
0.1788	2.3529	350	0.1797
0.1709	2.6891	400	0.1739
0.1604	3.0252	450	0.1693
0.141	3.3613	500	0.1671
0.1479	3.6975	550	0.1650

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.2.0
• Tokenizers 0.19.1