Quickstart

This guide will help you get up and running with Nenya quickly.

Basic Usage

1. Loading a pre-trained model

from nenya import io as nenya_io
from nenya import analyze_image

# Load model options and model file path
opt, nenya_model_file = nenya_io.load_opt('v5')  # Available models: 'v4', 'v5', 'viirs_v1'

2. Analyzing a single image

import numpy as np

# Create a sample image or load from file
image = np.random.rand(64, 64)  # Or load your actual satellite image (64x64)

# Extract latent representation and embed in UMAP space
embedding, pp_img, table_file, DT, latents = analyze_image.umap_image('v5', image)

print(f'UMAP coordinates: U0={embedding[0,0]:.3f}, U1={embedding[0,1]:.3f}')
print(f'DT value: {DT:.2f}')

3. Working with UMAP

from nenya import nenya_umap

# Load UMAP model
umap_model, table_file = nenya_umap.load('v5')

# Load table containing UMAP coordinates and metadata
import pandas as pd
umap_tbl = pd.read_parquet(table_file)

# Plot UMAP (with matplotlib)
import matplotlib.pyplot as plt
plt.figure(figsize=(10, 8))
plt.scatter(umap_tbl.US0, umap_tbl.US1, s=1, alpha=0.5)
plt.xlabel('U0')
plt.ylabel('U1')
plt.title('UMAP Embedding of Satellite Images')
plt.show()

4. Interactive visualization portal

import os
from nenya.portal import OSSinglePortal
from bokeh.server.server import Server

# Path to the UMAP table
table_file = os.path.join(os.getenv('OS_SST'), 'MODIS_L2', 'Nenya', 'Tables', 'MODIS_Nenya_96clear_v5_DT15.parquet')

# Function to create a session
def get_session(doc):
    sess = OSSinglePortal(table_file)
    return sess(doc)

# Start Bokeh server
server = Server({'/': get_session}, num_procs=1)
server.start()
print('Opening Bokeh application on http://localhost:5006/')

server.io_loop.add_callback(server.show, "/")
server.io_loop.start()

Training a New Model

from nenya.train import main as train_main

# Path to options file
opt_path = "path/to/opts_nenya_model.json"

# Train model
train_main(opt_path, debug=False)

Extracting Latents

from nenya.latents_extraction import model_latents_extract
from nenya import io as nenya_io

# Load options
opt, model_path = nenya_io.load_opt('v5')

# Data file path
data_file = "path/to/preprocessed_data.h5"

# Extract latents
latent_dict = model_latents_extract(opt, data_file, model_path)

# Access latents for valid and train sets
valid_latents = latent_dict['valid']
train_latents = latent_dict['train']

What’s Next

Check out the Examples section for more detailed usage examples
Learn about the Core Concepts behind Nenya
Explore the api-reference for a comprehensive list of functions and classes