Hey we have a simple Python script already that performs a simple action on Amazon
1. Navigate to url
2. Add product to cart
3. Click to checkout
Your job is to:
- Update it to ensure it still works and performs the necessary steps.
- Optimize it to use proxies to avoid getting blocked
- Saving the results (some data on the checkout page) into database.
- Take the saved data (category, BSR and price) and build a VERY simple machine learning model that correlates the numbers
- Build a simple method that we can use to query with Category and BSR and get the expected price.
After scraping several products using the 1st part, you will have this data in a Firestore table (we already have setup with firestore)
Category, BSR, Price
Pets, 100, $10
Auto, 200, $7
Cars, 1301, $19
.... (Using the scraper, we will populate hundreds of these).
Then we want to be able to use a simple method:
estimate_price(category: "Pets", bsr: 345) .... <-- Get a predicted number.
1. Scraper with the requirements mentioned
2. Predictor model and method.
3. Method to retrain the model as needed. (Have it re-read the table of cat/bsr/price to become more accurate)
As mentioned, the current code already has:
- Scraper working with a minor glitch, but performs all actions, saves into Firestore (doesn't use proxy)
- Predictor: Model created a long time ago, but does work.
You may, or not use the existing code, as long as you achieve the desired results.
I am looking for a mix of experience and value