Fardeen Ashraf

The entire dataset contains 30,906 Bangla raw and corresponding romanized sentences, labeled with three types of sentiments: Positive (1), Negative (0), Neutral (2). The dataset is split into training, validation, and testing sets, with the ratio of 80:10:10, resulting in each set having 24724, 3091, and 3091 samples, respectively.

This automated freshness and defect detection project intends to create an automated system to identify raw and stale elements and separate them from the fresh fruits.

An academic project made using html css and javascript on frontend and firebase for backend .