How to train a classification model on a custom dataset

<p>In this blog post, we will cover the necessary steps to train a custom image classification model and test it on images.</p> <p>The Ikomia API simplifies the development of Computer Vision workflows and provides an easy way to experiment with different parameters to achieve optimal results.</p> <h1>Get started with Ikomia API</h1> <p>You can train a custom classification model with just a few lines of code. To begin, you will need to install the API within a virtual environment.</p> <p>How to install a virtual environment</p> <pre> pip install ikomia</pre> <p><a href="https://ikomia-dev.github.io/python-api-documentation/getting_started.html" rel="noopener ugc nofollow" target="_blank">API documentation</a></p> <p><a href="https://ikomia-dev.github.io/python-api-documentation/getting_started.html" rel="noopener ugc nofollow" target="_blank">API repo</a></p> <p>In this tutorial, we will use the&nbsp;<a href="https://universe.roboflow.com/ds/48QIfZoFjO?key=NTZLzFA0Q2" rel="noopener ugc nofollow" target="_blank">Rock, Paper, Scissor dataset from Roboflow</a>.</p> <p>Ensure that the dataset is organized in the correct format, as shown below:</p> <p><em>(Note: The &ldquo;validation&rdquo; folder should be renamed to &ldquo;val&rdquo;.)</em></p> <p><img alt="" src="https://miro.medium.com/v2/resize:fit:321/0*ytoXxXZtouFjcBTY.jpg" style="height:460px; width:357px" /></p> <h1>&zwj;Run the train ResNet algorithm</h1> <p>You can also charge directly the open-source&nbsp;notebook&nbsp;we have prepared.</p> <p><a href="https://medium.com/@gdemarcq/how-to-train-a-classification-model-on-a-custom-dataset-6e505921b41c">Website</a></p>