Multitask (Flamingo)
Overview
A small Flamingo implementation for Multitask, packaged with an explicit configuration and an initialization checkpoint. The tiny variant is a reproducible starting point, not a trained model release.
The whole setup fits in a single module on purpose.
Changelog
- 2026-10-18 — initial upload
- 2026-10-21 — config and docs sync
Roadmap
Next steps, in no particular order: a cleaner entry point, input-shape docs, and a small evaluation script if time permits.
Repository Status
- Single-file implementation; the
__main__block runs a smoke test. config.jsonandtraining_args.jsonhold the generated settings.model.safetensorsis freshly initialized — plumbing, not results.- No benchmark numbers are reported on purpose.
Quick Check
python main.py --help
Inspect the script's __main__ block for its generated smoke-test example. Because this is a custom implementation, generic automatic loading APIs require an explicit adapter before use.
Training Details
| Item | Value |
|---|---|
| Framework | PyTorch |
| Precision | fp32 |
| Init scheme | xavier |
| Hardware | a single consumer GPU is enough |
| Typical run | minutes, not hours |
Limitations
- weights are random-init, so outputs carry no signal
- robustness, fairness and domain transfer are untested
- the API may change without notice while the design settles
Default Experiment Recipe
Defaults: adafactor optimizer, cosine schedule. Treat these as starting points; a real run should sweep learning rate and batch size before drawing conclusions.
Files
| File | Purpose |
|---|---|
main.py |
primary artifact |
README.md |
this documentation |
config.json |
architecture configuration |
training_args.json |
default experiment settings |
model.safetensors |
initialization checkpoint |
examples/sample_usage.py |
usage sketch |
License
Released under cc-by-4.0. Review the source-data terms separately when this repository is used with external datasets.
- Downloads last month
- 29