You can deploy this model and start earning money today!

reyhanwiira/
eunoiawiira-vgg-realwaste-classification

Name: eunoiawiira-vgg-realwaste-classification
Availability: InStock

generated_from_trainer

apache-2.0

model-index

Views

No views yet

eunoiawiira-vgg-realwaste-classification

This CNN model is a fine-tuned version of VGG11 and VGG16 on the imagefolder dataset of RealWaste Dataset. It achieves the following results on the evaluation set:

Loss: 0.6688
Accuracy: 0.8365

Model description

This CNN model is a fine-tuned version of VGG16 based on imagefolder dataset from kaggle.com/datasets/joebeachcapital/realwaste https://doi.org/10.3390/info14120633

Intended uses & limitations

This CNN model intended to classify an object especially a waste items across 12 major material types

Cardboard
Food Organics
Glass
Metal
Miscellaneous Trash
Mixed
Paper
Plastic
Porcelain
Rubber
Textille Trash
Vegetation

Dataset source is available at dataset_source.text

Training and evaluation data

More information needed

Training procedure

More information needed

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.0002
train_batch_size: 32
eval_batch_size: 32
optimizer: Adam
lr_scheduler_type: linear
num_epochs: 10

Training results

Training Loss	Epoch	Validation Loss	Accuracy
1.3587	1/10	0.8157	0.7285
0.8003	2/10	0.6563	0.7568
0.6828	3/10	0.5165	0.8218
0.5429	4/10	0.5163	0.8103
0.5080	5/10	0.4336	0.8470
0.4128	6/10	0.5926	0.8124
0.4008	7/10	0.5306	0.8270
0.3897	8/10	0.3438	0.8816
0.3073	9/10	0.4028	0.8753
0.3842	10/10	0.4557	0.8501

Epoch : 10 Train Loss : 0.30419 Validation Accuracy : 0.8501 Validation Loss : 0.45567

Framework versions

Transformers 4.37.2
Pytorch 2.1.0+cu121

reyhanwiira/eunoiawiira-vgg-realwaste-classification