lixg_chong_model002

This model is a fine-tuned version of Falconsai/nsfw_image_detection on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 1.3551
• Accuracy: 0.75

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: 8
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	4	1.9344	0.1667
No log	2.0	8	1.8649	0.3333
1.6376	3.0	12	1.7836	0.5
1.6376	4.0	16	1.7310	0.4167
1.3504	5.0	20	1.6152	0.5833
1.3504	6.0	24	1.5817	0.5833
1.3504	7.0	28	1.5656	0.5833
1.3337	8.0	32	1.5015	0.6667
1.3337	9.0	36	1.5062	0.6667
1.1244	10.0	40	1.4488	0.5833
1.1244	11.0	44	1.4459	0.75
1.1244	12.0	48	1.4110	0.6667
1.172	13.0	52	1.4307	0.8333
1.172	14.0	56	1.4241	0.6667
1.0562	15.0	60	1.3551	0.75

Framework versions

• Transformers 4.49.0
• Pytorch 2.6.0+cpu
• Datasets 3.5.0
• Tokenizers 0.21.0