Manage AI Modules
Deep Learning Studio provides some Artificial intelligence Modules.
The Artificial part here means that humans are not the ones (directly) using intelligence, but a machine, or software or algorithm. These algorithms are not programmed as usual, where you tell them exactly what to do. They are learning through data.
Get your data, upload to DLS and train project with Deep Learning AI module in two ways :
- 1.Classification AI App Module
- 2.Segmentation AI App Module
- Click on “Plugins” tab from the left sidebar.
- Here you will see the two tabs :
1) AI Modules
2) Development Environments
- Select AI Modules and you will get Available AI App Modules list of DLS.
- If you want to use any module, click on it, you will get a pop-up slide which is asking for INSTALL.
- Click on INSTALL
- Wait for a few seconds, it will collect all the required libraries.
- Then you will see the selected module in the Installed AI Modules section.
Classification is a process of categorizing a given set of data into classes. It can be performed on both structured or unstructured data. The process starts with predicting the class of given data points. The classes are often referred to as target, label, or categories.
The classification predictive modeling is the task of approximating the mapping function from input variables to discrete output variables. The main goal is to identify which class/category the new data will fall into. Some classifiers are binary, resulting in a yes/no decision. Others are multi-class, able to categorize an item into one of several categories.
A neural network is one of several machine learning algorithms that can help solve classification problems. Its unique strength is its ability to dynamically create complex prediction functions, and emulate human thinking, in a way that no other algorithm can.
This AI module provides various state of art Deep Learning Image classification algorithms , it provides full flexibility to select and configure Neural Networks based on your need. You can train on your custom dataset, No coding required. Trained model will be in GluonCV/mxnet format.
This AI module currently supports following architecture: -
- SENet:
- MobileNets:
- Versions supported :
- mobilenet1.0, mobilenet0.75, mobilenet0.5, mobilenet0.25, mobilenetv2_1.0, mobilenetv2_0.75, mobilenetv2_0.5, mobilenetv2_0.25, mobilenetv3_large, mobilenetv3_small
- Xception:
- DenseNet:
- Versions supported :
- densenet121, densenet161, densenet169, densenet201
- ResNeXt:
- Versions supported :
- se_resnext50_32x4d, se_resnext101_32x4d , se_resnext101_64x4d
- ResNets:
- Versions supported :
- resnet18_v1, resnet34_v1, resnet50_v1, resnet101_v1, resnet152_v1, resnet18_v2, resnet34_v2, resnet50_v2, resnet101_v2, resnet152_v2, resnet18_v1b, resnet34_v1b, resnet50_v1b, resnet50_v1b_gn, resnet101_v1b_gn, resnet101_v1b, resnet152_v1b, resnet50_v1c, resnet101_v1c, resnet152_v1c, resnet50_v1d, resnet101_v1d, resnet152_v1d, resnet50_v1e, resnet101_v1e, resnet152_v1e, resnet50_v1s, resnet101_v1s, resnet152_v1s
Many computer vision tasks require intelligent segmentation of an image, to understand what is in the image and enable easier analysis of each part. Deep learning can learn patterns in visual inputs in order to predict object classes that make up an image.
The main deep learning architecture used for image processing is a Convolutional Neural Network (CNN), or specific CNN frameworks like AlexNet, VGG, Inception, and ResNet. Models of deep learning for computer vision are typically trained and executed on specialized graphics processing units (GPUs) to reduce computation time.
This AI module solves instance segmentation problems using Mask RCNN Neural Network. This module allows you to choose one of various available CNN backbones (ResNet, MobileNet, etc.) for your Mask RCNN model. It also supports the ability to enable Feature Pyramid Network (FPN). You can train on your custom dataset, No coding required. Trained model can be exported in GluonCV/mxnet format.
- Mask RCNN: