Page 333 - AI Ver 3.0 class 10_Flipbook
P. 333

3.  Build your model:  Use  intuitive,  drag-and-drop  interfaces  to  create  models  for  tasks  like  classification,
                    regression, or clustering. The platform will guide you through the process and offer suggestions based on best
                    practices.
                 4.  Train the model: Once the model is built, use the platform’s Auto-ML (Automated Machine Learning) features
                    to optimise the training process. These tools automatically select the best algorithms and parameters for the
                    task.
                 5.  Fine-tune  parameters:  After  initial  training,  adjust  model  parameters  to  improve  performance.  No-code
                    platforms provide automated suggestions or allow manual fine-tuning for further optimisation.
                 6.  Evaluate and refine: Assess the model’s performance using built-in metrics. Interpret the results through
                    visualizations, and make necessary adjustments to enhance accuracy and efficiency.
                 7.  Deploy the model: Finally, deploy your model with ease. No-code tools provide deployment options like APIs,
                    web services, or seamless integration with your existing systems, making the deployment process simple.


                         Introduction to Lobe

                 Lobe is a user-friendly, no-code AI tool designed to make it easy for anyone to create AI models without needing
                 programming skills. It is part of a category called Automated Machine Learning (Auto-ML) tools, which handle the
                 complex processes of building, training, and optimising AI models automatically.






























                 Lobe specialises in image classification, a type of AI task where the goal is to identify and categorise objects
                 within images.

                    • It provides images with labels: You start by uploading a set of images to Lobe. Each image should be labelled
                   to indicate what it contains. For example:
                   An image of a cat is labelled as "cat."

                   An image of a dog is labelled as "dog."
                    • Automated model creation: Once you upload and label your images, Lobe automatically takes care of the
                   technical work:
                      ✶ It analyses the images.
                      ✶ It tries different AI models to find the best one.
                      ✶ It optimises the model for the highest accuracy in classifying your images.

                                                                                    Computer Vision (Practical)  331
   328   329   330   331   332   333   334   335   336   337   338