A package for processing and storing image vectors in PostgreSQL
Project description
SimpleAI_Image
A package for processing and storing image vectors in PostgreSQL.
Installation
pip install SimpleAI_Image
from SimpleAI_Image import DatabaseHandler, DataProcessor
from sklearn.datasets import fetch_openml
from tensorflow.keras.applications.vgg16 import preprocess_input as vgg_preprocess_input
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from sqlalchemy import text
# Define the database URL and table name
db_url = 'postgresql+psycopg2://Username:Password@localhost:5432/ThisISATEST'
db_handler = DatabaseHandler(db_url, 'vector_data', 512)
# Instantiate DataProcessor with VGG16 model
data_processor = DataProcessor(db_handler, model_name='VGG16', preprocess_func=vgg_preprocess_input, image_size=(32, 32))
# Load the example dataset (MNIST)
mnist = fetch_openml('mnist_784', version=1)
X = mnist.data[:500] # Limit to 500 instances for testing
y = mnist.target[:500].astype(int) # Ensure targets are integers
# Process data and store in database
X_embedded, y = data_processor.process_data(X, y)
# Fetch and preprocess data for visualization
query = text("SELECT * FROM vector_data")
X_embedded, y = data_processor.fetch_and_preprocess_data(query)
# Visualize data
data_processor.visualize_data(X_embedded, y)
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
simpleai_image-0.1.8.tar.gz
(3.9 kB
view hashes)
Built Distribution
Close
Hashes for SimpleAI_Image-0.1.8-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | ea995dd08eb027da134792f52c29b191a9de1f1e6f66b4510987f07c8d1acbc2 |
|
MD5 | 86996471efa718fe0091a84e79049a85 |
|
BLAKE2b-256 | 7f6979297f53db5d0118dfd81a36b976290b950b44d09dc2f9c23718f71ed8ec |