Machine Learning - PerceptiLabs

Benchmarking DL Models: How to do it Right

Check this out to discuss the importance of benchmarking models to derive an optimal version and show how PerceptiLabs makes it easy for you.

Low-code

Using Evaluations and Comparisons to Make Awesome DL Models

Think your DL model is awesome? Think again. In this blog, we'll review some metrics to evaluate your model to uncover whether it's actually solving your problem.

Low-code

Integrations: FastAPI for Fast SaaS Application Deployments

This blog shows why PerceptiLabs added the FastAPI deployment target, and to inspire how you might build on that output.

AI

Integrations: Gradio for Quick and Easy Model Testing

You asked how to deploy, use, and even test a model that has been exported by PerceptiLabs. So, we added a Gradio deployment target. This blog discusses the rationale behind this target and how you can use it to test your model.

Low-code

Low-Code: The Best Approach for TensorFlow

PerceptiLabs adopted a low-code approach right from the start to increase your DL workflow productivity, provide tools for blazingly-fast modeling, and get you to a working model faster. This blog explores why low-code is so effective for working with TensorFlow code.

Machine Learning

Apply Datasets and Pre-Trained Models Faster and Easier with PerceptiLabs

Our mission is evolving to help you apply your DL models to real-world problems, easier, and quicker. How could we possibly do that? Read on to find out!

Machine Learning

Overcoming Deep Double Descent Using PerceptiLabs

If you've ever heard the phrase: "this is gonna get worse before it gets better", then you're already halfway there to understanding Deep Double Descent. In this blog we review the Deep Double Descent phenomenon and how you can use PerceptiLabs to overcome it.

Machine Learning

A Guide to Using U-Nets for Image Segmentation

Image segmentation is a powerful technique to locate objects or boundaries in images on a per-pixel level. In this blog we take a quick look at how U-Nets work, explore how they build on standard CNNs, and show how easy it is to build and iterate on U-Nets in PerceptiLabs.

Machine Learning

Working With Unbalanced Datasets

ML practitioners wrangle data to create the best training datasets possible for their models. This generally means that the data is unbiased, well-structured, and will help to train a model that gives good predictive performance in the real world. Whether you collect data yourself from the real world, use an

healthcare

Use Case: Retinal OCT

The human eye is a complex sensing organ that reacts to light to provide us with a field of view into the world around us. Taking care of our eyes is an important aspect of our overall healthcare. However, our eyes are susceptible to a number of diseases, some of

IIoT

Use Case: Classifying Wood Veneers Into Dry and Wet

Industrial IoT is playing a significant role in Industry 4.0, especially for automating manufacturing processes, including manufacturing wood veneers. We built an image recognition model that could identify veneers as either dry or wet.

Optometry

Use Case: Ocular Disease Recognition

We decided to build an ML model in PerceptiLabs that applies image recognition techniques on fundus images to detect possible cataracts in patients. Using a model like this could help doctors, optometrists, and researchers to more easily classify and detect such conditions.

Machine Learning

Case Study: Genetic Screening Using CNNs in PerceptiLabs

One group of scientists leveraging ML is the Buchser lab at Washington University in Saint Louis School of Medicine in the Department of Genetics. There, Dr. Buchser and his team are working in functional genomics, High-Throughput Screening, and Next-Generation Sequencing.

Manufacturing

Use Case: Defect Detection in Metal Surfaces

With the rise of Industrial IoT (IIoT) and Industry 4.0, ML is playing an ever-increasing role in enabling automation. With this in mind, we built an ML model that can classify defects on metal materials using image recognition.

Machine Learning

Use Case: Skin Cancer Classification

To accurately diagnose cancer cases, doctors need every possible tool at their disposal. In the case of skin cancer, the affected areas can often be visually examined without the need for invasive diagnostics typically involved with cancers inside the body.

Use Case

Use Case: Classifying Chest X-Rays to Detect Pneumonia

To accurately diagnose ailments, doctors need every possible tool at their disposal. This Use Case shows how to classify chest x-rays to detect Pneumonia.

image classification

Use Case: Automated Weather Analysis Using Image Recognition

Many industries have the need to identify current and past weather conditions. The data helps them plan, organize, and/or optimize their operations. This Use Case shows automated weather analysis using Image Recognition.

Machine Learning

Essential Lingo for the Aspiring ML Practitioner

Every industry has its lingo and machine learning (ML) is no different. Here's a brief checklist of key terminology that all aspiring ML developers should be familiar with when building models.

Machine Learning

Tips to Steer Your Developer Career Towards Machine Learning

Excited by the hype around machine learning (ML)? Perhaps you’d like to transition into the field but you’re not sure where to start. Learn more here.

Machine Learning

Inference for the Real World: What to do with your ML Model Once You've Trained it

The end result of the ML modeling process is a file containing the model's trained weights ready for inference. But once you have this file, how do you actually use it for inference?

Machine Learning

The Power of Abstraction: How ML Frameworks and PerceptiLabs Provide a more Human-Readable Language for Describing ML Models

Everyone wants to enhance their understanding of machine learning or ML these days. But what does it really mean to do ML?

Loss Function

Choosing and Customizing Loss Functions for Image Processing

A loss function is the algorithm used during training to determine how well your model can make predictions using a given set of weights and biases. In this blog we review how a loss function works, its role and importance in optimization, how to choose one, and how to work with it in Perceptilabs.

visual modeling tool

Comparing PerceptiLabs to Pure Code in Transfer Learning

Pure-code ML frameworks like TensorFlow are powerful tools, but are often limited when it comes to the insight and explainability of models. In this blog we explore these issues, and introduce PerceptiLabs to solve these challenges.

Machine Learning

Transfer Learning Part II: When to use them in Image Processing

Transfer learning allows you to leverage an existing model by modifying and retraining it to fulfill a new use case. In this blog, we explore some key scenarios for why and when you should choose transfer learning over building a new machine learning model from scratch for image processing.

Machine Learning

Five GANs for Better Image Processing

Machine learning practitioners are increasingly turning to the power of generative adversarial networks (GANs) for image processing. Applications that really benefit from using GANs include: generating art and photos from text-based descriptions, upscaling images, transferring images across domains (e.g., changing day time scenes to night time), and many others.