diff --git a/docs/index.md b/docs/index.md index 29fd49e..47e00a6 100644 --- a/docs/index.md +++ b/docs/index.md @@ -7,6 +7,22 @@ Ecco runs inside Jupyter notebooks. It is built on top of [pytorch](https://pyto Ecco is not concerned with training or fine-tuning models. Only exploring and understanding existing pre-trained models. +## Installation + +You can install `ecco` either with `pip` or with `conda`. + +**with pip** + +```sh +pip install ecco +``` + +**with conda** + +```sh +conda install -c conda-forge ecco +``` + ## Tutorials - Video: [Take A Look Inside Language Models With Ecco](https://www.youtube.com/watch?v=rHrItfNeuh0). \[Colab Notebook] @@ -51,4 +67,4 @@ The [API reference](api/ecco) and the [architecture](architecture) page explain Having trouble? - The [Discussion](https://github.com/jalammar/ecco/discussions) board might have some relevant information. If not, you can post your questions there. -- Report bugs at Ecco's [issue tracker](https://github.com/jalammar/ecco/issues) \ No newline at end of file +- Report bugs at Ecco's [issue tracker](https://github.com/jalammar/ecco/issues) diff --git a/readme.md b/readme.md index 197d4dc..3f4078a 100644 --- a/readme.md +++ b/readme.md @@ -4,8 +4,20 @@

-[![PyPI Package latest release](https://img.shields.io/pypi/v/ecco.svg)](https://pypi.org/project/ecco) -[![Supported versions](https://img.shields.io/pypi/pyversions/ecco.svg)](https://pypi.org/project/ecco) + +[![GitHub - License](https://img.shields.io/github/license/jalammar/ecco?logo=github&style=flat&color=green)][#github-license] +[![PyPI - Latest Package Version](https://img.shields.io/pypi/v/ecco?logo=pypi&style=flat&color=orange)][#pypi-package] +[![PyPI - Supported Python Versions](https://img.shields.io/pypi/pyversions/ecco?logo=pypi&style=flat&color=blue)][#pypi-package] +[![Conda - Platform](https://img.shields.io/conda/pn/conda-forge/ecco?logo=anaconda&style=flat)][#conda-forge-package] +[![Conda (channel only)](https://img.shields.io/conda/vn/conda-forge/ecco?logo=anaconda&style=flat&color=orange)][#conda-forge-package] +[![Docs - GitHub.io](https://img.shields.io/static/v1?logo=readthedocs&style=flat&color=pink&label=docs&message=ecco)][#docs-package] + + +[#github-license]: https://github.com/jalammar/ecco/blob/main/LICENSE +[#pypi-package]: https://pypi.org/project/ecco/ +[#conda-forge-package]: https://anaconda.org/conda-forge/ecco +[#docs-package]: https://ecco.readthedocs.io/ + Ecco is a python library for exploring and explaining Natural Language Processing models using interactive visualizations. @@ -27,11 +39,27 @@ Documentation: [ecco.readthedocs.io](https://ecco.readthedocs.io/) - **Feature attribution** ([IntegratedGradients](https://arxiv.org/abs/1703.01365), [Saliency](https://arxiv.org/abs/1312.6034), [InputXGradient](https://arxiv.org/abs/1412.6815), [DeepLift](https://arxiv.org/abs/1704.02685), [DeepLiftShap](https://proceedings.neurips.cc/paper/2017/hash/8a20a8621978632d76c43dfd28b67767-Abstract.html), [GuidedBackprop](https://arxiv.org/abs/1412.6806), [GuidedGradCam](https://arxiv.org/abs/1610.02391), [Deconvolution](https://arxiv.org/abs/1311.2901), and [LRP](https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0130140) via [Captum](https://captum.ai/)) - **Capture neuron activations** in the FFNN layer in the Transformer block - Identify and **visualize neuron activation patterns** (via Non-negative Matrix Factorization) -- Examine neuron activations via comparisons of activations spaces using [SVCCA](https://arxiv.org/abs/1706.05806), [PWCCA](https://arxiv.org/abs/1806.05759), and [CKA](https://arxiv.org/abs/1905.00414) +- Examine neuron activations via comparisons of activations spaces using [SVCCA](https://arxiv.org/abs/1706.05806), [PWCCA](https://arxiv.org/abs/1806.05759), and [CKA](https://arxiv.org/abs/1905.00414) (See [this video on inspecting neural networks with CCA](https://www.youtube.com/watch?v=u7Dvb_a1D-0)) - Visualizations for: - Evolution of processing a token through the layers of the model ([Logit lens](https://www.lesswrong.com/posts/AcKRB8wDpdaN6v6ru/interpreting-gpt-the-logit-lens)) - Candidate output tokens and their probabilities (at each layer in the model) +## Installation + +You can install `ecco` either with `pip` or with `conda`. + +**with pip** + +```sh +pip install ecco +``` + +**with conda** + +```sh +conda install -c conda-forge ecco +``` + ## Examples: ### What is the sentiment of this film review?