| CortexLab |
| Copyright 2026 CortexLab Contributors |
|
|
| This project is a derivative work built upon TRIBE v2 by Meta Platforms, Inc. |
|
|
| ================================================================================ |
|
|
| TRIBE v2 |
| Copyright (c) Meta Platforms, Inc. and affiliates. |
| Licensed under Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) |
| https://github.com/facebookresearch/tribev2 |
|
|
| The following source files in this repository are derived from TRIBE v2: |
| - src/cortexlab/core/model.py (modified: added return_attn, compile_backbone) |
| - src/cortexlab/data/loader.py (original: tribev2/utils.py) |
| - src/cortexlab/data/transforms.py (original: tribev2/eventstransforms.py) |
| - src/cortexlab/data/fmri.py (original: tribev2/utils_fmri.py) |
| - src/cortexlab/data/studies/* (original: tribev2/studies/*) |
| - src/cortexlab/training/experiment.py (original: tribev2/main.py) |
| - src/cortexlab/training/pl_module.py (original: tribev2/pl_module.py) |
| - src/cortexlab/inference/predictor.py (original: tribev2/demo_utils.py) |
| - src/cortexlab/viz/* (original: tribev2/plotting/*) |
|
|
| The pretrained model weights are hosted by Meta at: |
| https://huggingface.co/facebook/tribev2 |
|
|
| ================================================================================ |
|
|
| neuralset (v0.0.2) |
| Copyright (c) Meta Platforms, Inc. and affiliates. |
| Licensed under CC BY-NC 4.0 |
|
|
| neuraltrain (v0.0.2) |
| Copyright (c) Meta Platforms, Inc. and affiliates. |
| Licensed under CC BY-NC 4.0 |
|
|
| ================================================================================ |
|
|
| Third-party dependencies and their licenses: |
|
|
| PyTorch (torch) - BSD 3-Clause |
| Transformers - Apache 2.0 |
| HuggingFace Hub - Apache 2.0 |
| x_transformers - MIT |
| einops - MIT |
| NumPy - BSD 3-Clause |
| pandas - BSD 3-Clause |
| PyYAML - MIT |
| MoviePy - MIT |
| gTTS - MIT |
| langdetect - Apache 2.0 |
| spaCy - MIT |
| SoundFile - BSD 3-Clause |
| julius - MIT |
| pydantic - MIT |
| requests - Apache 2.0 |
| tqdm - MIT / MPL 2.0 |
| SciPy - BSD 3-Clause |
| nilearn - BSD 3-Clause |
| MNE-Python - BSD 3-Clause |
| PyVista - MIT |
| matplotlib - PSF / BSD |
| PyTorch Lightning - Apache 2.0 |
| Weights & Biases - MIT |
|
|
| ================================================================================ |
|
|
| LLaMA 3.2-3B (used as text feature extractor by the pretrained TRIBE v2 model) |
| Copyright (c) Meta Platforms, Inc. |
| Licensed under the Meta Llama 3.2 Community License Agreement |
| https://llama.meta.com/llama3_2/license/ |
|
|
| Users must independently accept Meta's LLaMA license to use the pretrained |
| TRIBE v2 model for text feature extraction. CortexLab does not redistribute |
| LLaMA weights. |
| |
