Clothes Classification with the DeepFashion Dataset and Fastai

By Medium - 2021-02-02

Description

How to outperform the benchmark in clothes recognition with fastai and DeepFashion Dataset. How to use fastai models in PyTorch. Code, explanation, evaluation on the user data.

Summary

A story of how I started cleaning my wardrobe and ended up training a neural network The first of January is a perfect day to sort out your winter clothes.
In this article, I am writing about the approach that I took for creating my dataset, specifically how to recognize the clothes type of a random item.
Early Evaluation Early evaluation allows us to check the learning progress on the earlier stages and catch mistakes in our approaches before we spent a lot of time training the model.
Learning Rate Finder Now let’s go through the data in the DataLoader and gradually increase the learning rate with each mini-batch to observe how the value of loss changes with the change of learning rate.

Topics

Machine_Learning (0.33)
Backend (0.17)
NLP (0.11)

Similar Articles

K-fold Cross Validation with PyTorch

By MachineCurve - 2021-02-02

Explanations and code examples showing you how to use K-fold Cross Validation for Machine Learning model evaluation/testing with PyTorch.

How to Accelerate Learning of Deep Neural Networks With Batch Normalization

By Machine Learning Mastery - 2019-01-17

Batch normalization is a technique designed to automatically standardize the inputs to a layer in a deep learning neural network. Once implemented, batch normalization has the effect of dramatically a ...

Something Every Data Scientist Should Know But Probably Doesn’t: The Bias-Variance Trade-off…

By Medium - 2021-01-04

A groundbreaking and relatively new discovery upends classical statistics with relevant implications for data science practitioners and…

Fine-Tune Wav2Vec2 for English ASR in Hugging Face with 🤗 Transformers

By huggingface - 2021-03-12

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

trekhleb/links-detector

By GitHub - 2020-12-07

📖 👆🏻 Links Detector makes printed links clickable via your smartphone camera. No need to type a link in, just scan and click on it. - trekhleb/links-detector

Integrate Trained Machine Learning Model into API in Just 15 mins

By Medium - 2020-12-14

How to deploy a trained sentiment analysis machine learning model to a REST API using Microsoft ML.NET and ASP.NET Core, in just 15 mins.