torchextractor
mapextrackt
Our great sponsors
| torchextractor | mapextrackt | |
|---|---|---|
| 1 | 1 | |
| 99 | 46 | |
| - | - | |
| 4.2 | 0.0 | |
| about 3 years ago | 14 days ago | |
| Python | Python | |
| Apache License 2.0 | MIT License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
torchextractor
-
[P] Pytorch: Intermediate Feature Extraction
Recently I worked on torchextrator, a standalone python package that makes it simple to extract features in PyTorch. You no longer need to duplicate code and rewrite the forward function. Also the extractor supports nested modules, custom caching operations and is ONNX compatible!
mapextrackt
-
[D] Visualize network activation in non-classification models
You can try visualizing the convolution filters. I dunno about something good for keras but this is a nice pytorch library I’ve used before that works well. https://github.com/lewis-morris/mapextrackt
What are some alternatives?
stringlifier - Stringlifier is on Opensource ML Library for detecting random strings in raw text. It can be used in sanitising logs, detecting accidentally exposed credentials and as a pre-processing step in unsupervised ML-based analysis of application text data.
PyTorch-Model-Compare - Compare neural networks by their feature similarity
muzero-general - MuZero
nni - An open source AutoML toolkit for automate machine learning lifecycle, including feature engineering, neural architecture search, model compression and hyper-parameter tuning.
merged_depth - Monocular Depth Estimation - Weighted-average prediction from multiple pre-trained depth estimation models
carbon - :black_heart: Create and share beautiful images of your source code
Unredactor - In this project we are tryinbg to create unredactor. Unredactor will take a redacted document and the redacted flag as input, inreturn it will give the most likely candidates to fill in redacted location. In this project we are only considered about unredacting names only. The data that we are considering is imdb data set with many review files. These files are used to buils corpora for finding tfidf score. Few files are used to train and in these files names are redacted and written into redacted folder. These redacted files are used for testing and different classification models are built to predict the probabilies of each class. Top 5 classes i.e names similar to the test features are written at the end of text in unreddacted foleder.