perceiver-io
A PyTorch implementation of Perceiver, Perceiver IO and Perceiver AR with PyTorch Lightning scripts for distributed training
File Explorer
Download Latest Version (.zip)- ci_install-pkg.yml
- ci_testing.yml
- docker.yml
- library-design-small.jpg
- library-design.jpg
- optical-flow.gif
- perceiver-ar.png
- perceiver-io.png
- perceiver.png
- small-perceiver-ar.png
- small-perceiver-io.png
- small-perceiver.png
- building-blocks.md
- dataset-preproc.md
- docker-image.md
- library-design.md
- model-construction.md
- pretrained-models.md
- training-examples.md
- approach_1.csv
- approach_2.csv
- run-logs-0.8.0_scaling-1_version_0-tag-val_loss.csv
- run-logs-0.8.0_scaling-1_version_1-tag-val_loss.csv
- run-logs-0.8.0_scaling-1_version_2-tag-val_loss.csv
- run-logs-0.8.0_scaling-2a_version_0-tag-val_loss.csv
- run-logs-0.8.0_scaling-2a_version_1-tag-val_loss.csv
- run-logs-0.8.0_scaling-2a_version_2-tag-val_loss.csv
- run-logs-0.8.0_scaling-2a_version_3-tag-val_loss.csv
- run-logs-0.8.0_scaling-2a_version_4-tag-val_loss.csv
- run-logs-0.8.0_scaling-2b_version_0-tag-val_loss.csv
- run-logs-0.8.0_scaling-2b_version_1-tag-val_loss.csv
- perceiver-ar.png
- __init__.py
- flops.py
- laws.py
- __init__.py
- article.ipynb
- article.py
- train.md
- train.py
- update.md
- __init__.py
- prep.sh
- train.py
- train.sh
- train_fsdp.sh
- train.py
- train.sh
- valid.sh
- prep.sh
- train.py
- train.sh
- prep.sh
- train.py
- train.sh
- prep.sh
- train.py
- train.sh
- prep.sh
- train_all.py
- train_all.sh
- train_dec.py
- train_dec.sh
- valid_all.sh
- valid_dec.sh
- __init__.py
- download_checkpoints.sh
- __init__.py
- convert.py
- inference.ipynb
- __init__.py
- giantmidi_piano.py
- maestro_v3.py
- midi_processor.py
- symbolic.py
- utils.py
- __init__.py
- bookcorpus.py
- bookcorpusopen.py
- c4.py
- collator.py
- common.py
- enwik8.py
- imdb.py
- utils.py
- wikipedia.py
- wikitext.py
- __init__.py
- common.py
- imagenet.py
- mnist.py
- optical_flow.py
- video_utils.py
- __init__.py
- __init__.py
- backend.py
- huggingface.py
- lightning.py
- __init__.py
- __init__.py
- adapter.py
- classifier.py
- config.py
- huggingface.py
- lightning.py
- modules.py
- position.py
- utils.py
- __init__.py
- backend.py
- huggingface.py
- lightning.py
- __init__.py
- backend.py
- huggingface.py
- lightning.py
- __init__.py
- backend.py
- huggingface.py
- __init__.py
- backend.py
- huggingface.py
- lightning.py
- utils.py
- __init__.py
- __init__.py
- backend.py
- huggingface.py
- lightning.py
- __init__.py
- backend.py
- huggingface.py
- __init__.py
- __init__.py
- __init__.py
- preproc.py
- symbolic.py
- __init__.py
- classifier.py
- clm.py
- clm_fsdp.py
- mlm.py
- preproc.py
- __init__.py
- image_classifier.py
- __init__.py
- cli.py
- lrs.py
- trainer.yaml
- __init__.py
- audio_prompt_1.mid
- causal_language_model_generate_test.py
- causal_language_model_pipeline_test.py
- conftest.py
- image_classifier_convert_test.py
- image_classifier_mnist_test.py
- kv_cache_test.py
- mask_filler_test.py
- masked_language_model_convert_test.py
- masked_language_model_filler_test.py
- optical_flow_pipeline_test.py
- optical_flow_test.py
- symbolic_audio_model_generate_test.py
- symbolic_audio_model_pipeline_test.py
- text_classifier_imdb_test.py
- text_classifier_test.py
- text_data_module_test.py
- utils.py
- .dockerignore
- .gitattributes
- .gitignore
- .pre-commit-config.yaml
- CITATION.cff
- Dockerfile
- environment.yml
- LICENSE
- poetry.lock
- pyproject.toml
- README.md
- tasks.py
๐ Installation Guide
1. Get the code
git clone https://github.com/krasserm/perceiver-io
Downloads the entire project code from GitHub to your computer.
cd perceiver-io
Moves into the project folder you just downloaded.
2. Official Install Script
Easy RecommendedPrerequisites
- Python 3 Python is required to use pip.
pip install perceiver-io[text,vision,audio]
Installs the package published on PyPI directly โ no need to clone the source.
After installing, open a new terminal and run the program's version command (e.g. --version) to confirm it worked.
Pulled directly from this repo's README.
3. Docker
EasyPrerequisites
- Git Needed to download the project code from GitHub.
- Docker Desktop Needed to build and run containers. Install it and keep it running in the background.
docker pull ghcr.io/krasserm/perceiver-io:latest
Type this command into your terminal and run it.
Run docker compose ps to check the containers are Up. If the README mentions a port, open http://localhost:PORT in your browser.
Pulled directly from this repo's README.
4. Python
EasyPrerequisites
pip install perceiver-io[text,vision,audio]
Installs the package published on PyPI directly โ no need to clone the source.
poetry install --all-extras
Installs the libraries listed in pyproject.toml.
poetry install --all-extras --with examples
Installs the libraries listed in pyproject.toml.
python -m perceiver.scripts.vision.image_classifier fit \
Runs the Python script (or module).
If it runs without errors and prints output in the terminal, it worked.
Pulled directly from this repo's README.
// repository documentation
Was this content helpful?
(0 ratings)