Skip to content

Latest commit

 

History

History
64 lines (49 loc) · 2.57 KB

README.md

File metadata and controls

64 lines (49 loc) · 2.57 KB

AN IMPORTANT NOTICE

Due to a major update in Huggingface, the ChatBot option isn't functioning. To understand the entire thing and how it works, here is the tutorial that will help you to get idea about this amazing project.

https://youtu.be/A23aHxtiApw?si=e3qVegqD30mOro4w

The Good thing is another version of this ML project is still online, you can try here:

https://www.startech.com.bd/tool/finder

laptop-guru

This ML Model will help you to choose the desired laptop based on your given information.

Problem Statement

In student life or professional life, almost 95% of people struggle to buy the best laptop according to their demands & purposes. Some of them ended up buying underpowered laptops based on their work & activities. Some of them ended up buying overpowered laptops. It's the kind of thing that, we must confuse to buy the best laptop among the thousands of laptops in the market. To solve this issue, I introduced to you the Laptop Guru ML Recommendation System. Just put your desired specifications & working purposes & it will suggest the best laptop for you.

A-Z Process of Building Laptop Guru

Choosing the Appropriate Model

In this case, **distilroberta-base ** model has been chosen for this ML project. Even though 2000 Laptops sound a lot but in ML world, it's very tiny dataset. To efffectively the tiny dataset, **distilroberta-base ** is the best for this. To understand more check out this link:

https://huggingface.co/distilroberta-base

Sites that I have scrapped to collect data

I have choosen this site that contain 4k laptops dataset for this ML projects:

https://www.startech.com.bd/laptop-notebook/laptop

###process of scraping & data saving

  1. Go to Scraped_data folder
  2. Open details_scraper.py
  3. run this command:
    python -u details_scraper.py
    '''

After couples of hours you will get desired data.

Model Training

To understand the entire process, there is a file, please run it at your machine:

python -u text_Data_trainer_with_blurr.ipynb

I am using transformer blurr package to train this model.

Training accuracy

Stage_0_Training_Image

Stage_0_Training_Image

Understand How to use

https://youtu.be/A23aHxtiApw

Huggingface Deployment

https://huggingface.co/spaces/tdnathmlenthusiast/laptop_guru