File size: 3,466 Bytes
ebcfe17
 
 
171b36b
 
 
 
 
fb4759f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
171b36b
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
---
license: apache-2.0
pipeline_tag: tabular-classification
tags:
- Physics
- Stellar Classification
datasets:
- Allanatrix/Astro
---
# NexaAstro: Stellar Classification with SDSS Data

**NexaAstro - Stellar Classification** is a hybrid machine learning model for classifying stars using data from the Sloan Digital Sky Survey (SDSS). It leverages a two-stage architecture:

- **CatBoost**: Gradient boosting for robust feature modeling.
- **Feedforward Neural Network (NN)**: Refines predictions for enhanced accuracy.

This model is part of the [Nexa Scientific Model Suite](https://huggingface.co/spaces/Allanatrix/NexaHub), dedicated to advancing scientific discovery through machine learning.

---

## Model Overview

- **Task**: Multi-class stellar classification (e.g., Main Sequence, White Dwarf, Giant).
- **Input**: SDSS stellar attributes (u, g, r, i, z magnitudes, spectral lines, etc.).
- **Output**: Predicted stellar class label.
- **Architecture**: CatBoost for feature extraction, followed by a Feedforward Neural Network for classification.

---

## Applications

- **Stellar Population Studies**: Analyzing distributions and characteristics of stellar types.
- **Galaxy Classification Support**: Providing stellar data for broader galactic studies.
- **Astrophysics Education**: Enabling interactive learning and research tools.
- **Feature Engineering**: Supporting advanced astronomical machine learning workflows.

---

## Getting Started

### Example Usage
```python
import joblib
import torch
import numpy as np
from my_nn_model import StellarNN  # Replace with actual neural network module

# Load CatBoost model
catboost_model = joblib.load("Allanatrix/catboost_model.pkl")

# Load PyTorch neural network model
nn_model = StellarNN()
nn_model.load_state_dict(torch.load("Allanatrix/stellar_nn.pt"))
nn_model.eval()

# Example prediction with SDSS features
features = np.array([...])  # SDSS input features (e.g., magnitudes, spectral data)
catboost_out = catboost_model.predict(features)
refined_pred = nn_model(torch.tensor(catboost_out).float())
```

Refer to the model documentation for detailed preprocessing and input requirements.

---

## Dataset

- **Source**: [Sloan Digital Sky Survey (SDSS)](https://www.sdss.org/).
- **Preprocessing**: Data cleaned, normalized, and filtered by magnitude thresholds.
- **Labels**: Discrete stellar class labels derived from expert annotations.

---

## Citation and License

If you use NexaAstro in your research, please cite this repository and acknowledge the SDSS dataset.  
The model and associated code are licensed under the **Boost Software License 1.1 (BSL-1.1)**.

---

## Part of the Nexa Scientific Ecosystem

Explore related tools and models in the Nexa ecosystem:
- [Nexa Data Studio](https://huggingface.co/spaces/Allanatrix/NexaDataStudio): Tools for data processing and visualization.
- [Nexa R&D](https://huggingface.co/spaces/Allanatrix/NexaR&D): Research-focused model development environment.
- [Nexa Infrastructure](https://huggingface.co/spaces/Allanatrix/NexaInfrastructure): Scalable ML deployment solutions.
- [Nexa Hub](https://huggingface.co/spaces/Allanatrix/NexaHub): Central portal for Nexa resources.

*Coming Soon:*
- Galaxy Morphology Classifier
- Exoplanet Transit Detection Model

---

*Developed and maintained by [Allan](https://huggingface.co/Allanatrix), an independent machine learning researcher specializing in astrophysical and scientific AI systems.*