Atlas — dynamic reverse-engineering: emulate, live-instrument, and fuzz functions, with a self-improving world model. The runtime companion to spectrIDA.
Project description
🧠 PROJECT ATLAS — World Model AI
An AI that doesn't just predict — it understands. Builds an internal model of reality, imagines futures, and plans actions.
Architecture
Observation → [Encoder] → Latent State → [Neural ODE Dynamics] → Future State
↑ ↓
Real World [Decoder] → Prediction
↑
Surprise Detection
(only learn from surprises)
Quick Start
# Install
cd project_atlas
pip install -e ".[dev]"
# Train on grid world
python scripts/train.py --env grid_world
# Visualize imagined trajectories
python scripts/imagine.py --checkpoint experiments/phase1_v1/best.pt
# Run tests
pytest tests/ -v
Status
- Phase 1: Core Engine (in progress)
- Phase 2: Perception Upgrade
- Phase 3: Causal Reasoning
- Phase 4: Continuous Learning
- Phase 5: Scale + Integration
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
phantomrt-0.1.1.tar.gz
(93.6 kB
view details)
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
phantomrt-0.1.1-py3-none-any.whl
(95.2 kB
view details)
File details
Details for the file phantomrt-0.1.1.tar.gz.
File metadata
- Download URL: phantomrt-0.1.1.tar.gz
- Upload date:
- Size: 93.6 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.12.10
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
83fbbef7bab131cfe23b13b4d2dcba9e562eb16b0da934db2534ac7085f374d3
|
|
| MD5 |
3307a6db0e2f92903ca0a027a9aea9b4
|
|
| BLAKE2b-256 |
5304fe065ecf2d55bfbd6208efded46c34779c02d2f7362c7672c2752fce84a8
|
File details
Details for the file phantomrt-0.1.1-py3-none-any.whl.
File metadata
- Download URL: phantomrt-0.1.1-py3-none-any.whl
- Upload date:
- Size: 95.2 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.12.10
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
845fdb085a91a0b0432726d4b67c3fa9f4c50e4591bad16aef156d44aaea920e
|
|
| MD5 |
2fc080816e52030ce31cb7a9169df339
|
|
| BLAKE2b-256 |
6d8e1a6522ce86eb2266bb218d2c9af142b7dc28fc2ccf308613cef69401d104
|