blip_finetuned_fashion Open-source Visual Question Answering Model

Blip Finetuned Fashion

Developed by Ornelas

This model is a visual question answering model fine-tuned from Salesforce/blip-vqa-base, specializing in the fashion domain

Downloads 2,281

Release Time : 4/1/2024

Model Overview

A vision-language question answering model based on BLIP architecture, fine-tuned for answering fashion-related questions

Fashion Domain Optimization

Specially fine-tuned for fashion-related questions

Efficient Training

Achieved loss reduction from 9.43 to 0.09 within 10 epochs

Stable Convergence

Training process shows stable convergence with continuously decreasing validation loss

Image Understanding

Visual Question Answering

Fashion Item Recognition

Image Content Description

E-commerce

Fashion Product Q&A

Answering user questions about product attributes, styles, etc.

Fashion Recommendation

Styling Suggestions

Providing styling suggestions based on image content

Property	Details
Model Type	Fine - tuned version of Salesforce/blip-vqa-base
Training Data	Unknown

Training Loss	Epoch	Step	Validation Loss
9.4382	1.0	28	8.7410
5.7906	2.0	56	4.4909
2.6817	3.0	84	1.9184
0.6839	4.0	112	0.4660
0.1304	5.0	140	0.1403
0.0621	6.0	168	0.1171
0.0322	7.0	196	0.0961
0.0178	8.0	224	0.0963
0.0119	9.0	252	0.0918
0.0088	10.0	280	0.0916

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base