Introduction
NeuralMarker offers a variety of state-of-the-art Pretrained Deep Learning Models for Image Classification from different available frameworks like TensorFlow, Keras, and Caffe.
The Pretrained Models available at your fingertip in NeuralMarker Neural Network Library can be effectively used to fine-tune or make better predictions on Image classification Datasets.
This list is not static, as the more state-of-the-art model will be added in NeuralMarker then accordingly the below list will be updated too:
Pretrained Models
1. EfficientNet-B0
Network Backbone: ResNet-50
DenseNet-169
No of Categories:1000
Dataset Used: ImageNet
2. EfficientNet-B1
Network Backbone: ResNet-152
DenseNet-264
Inception-v3
Xception
No of Categories:1000
Dataset Used: ImageNet
3. EfficientNet-B2
Network Backbone: Inception-v4
Inception-resnet-v2
No of Categories:1000
Dataset Used: ImageNet
4. EfficientNet-B3
Network Backbone:ResNeXt-101
PolyNet
No of Categories:1000
Dataset Used: ImageNet
5. EfficientNet-B4
Network Backbone: SENet
NASNet-A
AmoebaNet-A
PNASNet
No of Categories:1000
Dataset Used: ImageNet
6. EfficientNet-B5
Network Backbone: AmoebaNet-C
No of Categories:1000
Dataset Used: ImageNet
7. EfficientNet-B6
No of Categories:1000
Dataset Used: ImageNet
8. EfficientNet-B7
Network Backbone: GPipe
No of Categories:1000
Dataset Used: ImageNet
9. OpenCV_MobileNet_Classification
No of Categories:1000
Dataset Used: ILSVRC2012
10. MobileNetv2_ImageNet
No of Categories:1000
Dataset Used: ImageNet
11. Resnet50_ImageNet
No of Categories:1000
Dataset Used: ImageNet
12. Vision-transformer-imagenet-L32
No of Categories:1000
Dataset Used: ImageNet
13. Vision-transformer-imagenet-L16
No of Categories:1000
Dataset Used: ImageNet
14. Vision-transformer-imagenet-B32
No of Categories:1000
Dataset Used: ImageNet
15. Vision-transformer-imagenet-B16
No of Categories:1000
Dataset Used: ImageNet
How to create labels on an Image Classification Dataset using S-O-T-A Pre-Trained Models
Steps to Follow
1. Login to the tool
2. Click on the Add button.
3. The Add Dataset Form will appear.
4. Fill in all the fields such as dataset name, dataset description, category-type, and categories.
4 a. Select category-type as Classification
4 b. Choose Pretraining Models from the available list.
4 c. Tags according to the chosen model will appear. The terminology used for these Model tags is as follows:
"Model-name_dataset-name(on which they are trained)-predictions"
All Classification Models are trained on ImageNet Dataset5. Add data to NeuralMarker using the available options listed below:
- Google Drive Link or S3 link
- CSV File with Image URL's
- Drag and Drop
6. Click on Submit Button
7. A New Dataset will be created.
8. After New Dataset is created:
- The Dataset card will display the status "Pretrain Model running".
- The brain symbol on hover will display "AI labeling in Progress".
9. After Pretrain Model stops running over the entire dataset:
- The Dataset card will display the status "Pretrain Model Done & Ready For Annotation".
- The brain symbol on the card will display "AI Labeling Report".
10. After Pretrain Model stops running over the entire dataset, an AI labeling report will be generated:
- With a confidence score of "class labels" over the images in the dataset.
