2 min video showing the process and outcome of the project - click on the image below.
The current size guide used in Zara Kids is a table not very user-friendly. The objective of this project is create a size recommendator using a picture of the user and a Machine Learning model that validate the picture and recommend the best size according to the main measures.
- Sklearn | Machine Learning Library with regression models
- CV2 | Image treatment for python
- Tkinter | GUI for python
- train_human_dataset_images | Shape 565
- train_not_human_dataset_images | Shape 590
- test_kids_dataset_images | Shape 50
- Updated size recommendation dataset
- Models trained
- Code to recommend size
-
Generate the strategy
-
Cleaning and create all datasets Using webscrapping the picture dataset is generated.
In the cleaning work, the following decision has to be done
- Ensure not Null values
- Avoid not numeric values
- Drop the high correlated features
- Standardize or normalize high-value range features
- Get all images path
- Identify the best model Different models apport different results. Some of them are analysed in the project.
- LinearClassification
- RandomForestClassification
- HistGradientBoostingClassification
- Analyse the result There are metrics from sklearn that help you to choose the best model
- Accuracy score
- MRSE…
- Generate the cv2 image treatment
- Get measures in the picture
- Ensure pixel per metric transformation
- Save this measures
- Recommend using query to the size dataset
- Obtain the size for the measures given
As it can be appreciated, the best recommendation for the given measures is addresed.
Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.
Learning and enjoying every day. Next steps:
- Product matrix and % of recommendation based on delivery and satisfaction for simmilar users
- Try better neural networks to improve the effficiency and velocity of the model
- Work on the automathic recognision pattern
- HTML, CSS, JS