MINI WORKING PROJECT
Our Aim is to Deep Learning techniques such that convolutional neural networks (CNNs) models test the frameworks and validate the dataset through training and testing fruit images to get the ripe or damaged quality of the freshness of the fruit.
->Import all the Dependencies
->Set constant size
->Import fruits data into the TensorFlow dataset object
->Visualize some of the images from our dataset
->Function to Split Dataset
->Cache, Shuffle, and Prefetch the Dataset
->Building the Model
->Creating a Layer for Resizing and Normalization
->Data Augmentation
->Apply Data Augmentation on Training Data
->Model Deployment & Architecture
->Compiling the Model
->Plotting the accuracy and loss curves
->Run prediction on a sample image
->Write a function for inference
->Saving the Model
Since very little was known about fruit detection when this study was initiated.
Throughout the construction process, we acquired knowledge regarding the enhancement capabilities.
We can broaden some of our focus to improve our efficiency.
• Interactive with a better model
• Manage the prediction results
• Add and loaded the dataset
• Making the flexible in grading results
• Cost-effective
. Environment Path
Training and Testing(include validating)
