PSANN-LM
Language-model tasks for PSANN, packaged separately as psannlm. Version 2.0.0 requires psann>=2.0.0 and uses the typed task and architecture configuration API as its authoritative public surface.
Install core first, then this directory with python -m pip install ./psannlm from the repository root. PyTorch, NumPy, SentencePiece, tokenizers, datasets, Hugging Face Hub, and PyYAML are declared runtime dependencies.
Use PSANNLM, PSANNLMDataPrep, LMConfig, and LMArchitectureConfig from psannlm. Select one of four architecture kinds: transformer, residual, wave, or geometric-sparse. Spectral residual is a nested residual policy. Train with model.fit(data, train=TrainConfig(...)), generate with model.generate(...), and persist with save/load.
python -m psannlm --help
python -m psannlm train --help
python -m psannlm generate --help
The LM guide covers a complete small training example, YAML, tokenizer identity, and schema-v1 model/trainer checkpoints. The repository migration guide documents compatibility routes for older applications and artifacts. Core-only installations do not contain this package; LM artifacts contain no core package files.
1 maintainer
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
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
File details
Details for the file psannlm-2.0.0.tar.gz.
File metadata
- Download URL: psannlm-2.0.0.tar.gz
- Upload date:
- Size: 83.9 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via:
twine/7.0.0 CPython/3.11.9
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
45494fdb5852bfade6e90f01e43be935c3d7081f25a2ef9f444108bf001bdc40
|
|
| MD5 |
26cc1d7a7e15b21c4aa1ba3eb4c25d13
|
|
| BLAKE2b-256 |
4ed7a632c78433646ff271e2953efc2f928f336c7e3dfb8788d34e262aeee92a
|
File details
Details for the file psannlm-2.0.0-py3-none-any.whl.
File metadata
- Download URL: psannlm-2.0.0-py3-none-any.whl
- Upload date:
- Size: 105.6 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via:
twine/7.0.0 CPython/3.11.9
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
2f1f04b785635a7633bcf12d044860221f1d0704084fde437c0c914dc5e306cb
|
|
| MD5 |
05bee17f0f98efa14097533f5a654eaf
|
|
| BLAKE2b-256 |
58bcb9db88d38411be0d0effa2ab2d3af1a7520292b94c1f8cf8328d961a2302
|
