addons

addons copied to clipboard

What's the equivalence of tf.contrib.legacy_seq2seq.attention_decoder in tensorflow2?

5

**Describe the feature and the current behavior/state.** I have to get a model that was built with tf1 to run with tf2 but didn't find the equivalence of the function...

JiashuGuo

**Describe the feature and the current behavior/state.** Convolution layer in keras API only support "same", "valid", "causal" and "full" padding. However, these padding methods is different in shape-wise but same...

Zhaopudark

Fix Sklearn and Skimage Testing Dependencies

2

The build for TF2.8 and specifically python 3.10 requires us to increment the version of sklearn and skimage. Unfortunately this causes conflicts with the lowest version of NumPy we need...

seanpmorgan

y_true not cast to one_hot if needed in f1_score metric

2

Hi! As per [this line](https://github.com/tensorflow/addons/blob/8cec33fcaaf1cf90aec7bdd55a0fcdbb251ce5c2/tensorflow_addons/metrics/f_scores.py#L167), if `y_true` is not under the `one_hot` form (i.e. an integer), the call will fail, so it must be converted to `one_hot` encoding if needed...

BlueskyFR

Pin TF2.8 Build Container to r0.16 Branch

1

After https://github.com/tensorflow/build/issues/79 is solved.

seanpmorgan

Make reduce LR on plateau weight-decay compatible for decoupled optimizers

5

**Describe the feature and the current behavior/state.** It has been shown that decoupling weight decay from the learning rate can simply hyper parameter search and leads to better performance. https://arxiv.org/abs/1711.05101...

ben-arnao

Adding sparse solvers for linear systems

10

**Describe the feature and the current behavior/state.** Tensorflow doesn't have a direct way to solve sparse linear systems of equations. Specifically: solving Ax=b for x where A is a sparse...

tabakg

Nadam with MovingAverage or SWA throws error "no attribute '_m_cache_read'"

2

**System information** - Linux Ubuntu 18.04 - TensorFlow version 2.7.0 (with pip install) - TensorFlow-Addons version 0.15.0 - Python version 3.6 + Ipython version 7.22 - Is GPU used? No....

YinSun0

Make GN and IN the same dtype behavior as BN or LN in mixed_precision

3

This PR makes GroupNorm and InstanceNorm the same dtype behavior as keras' BatchNorm or LayerNorm in mixed_precision. Fixes #2550 ## Type of change - [x] Bug fix # Checklist: -...

Zhaopudark

Building a wheel for Mac M1 is getting an error

2

Hello all, I tried to build a wheel for Mac M1 for the last Addons. I did it like described in #2503. I get the following error. Does someone know...

StephanElter