addons

addons copied to clipboard

**System information** - OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Windows 11 - TensorFlow version and how it was installed (source or binary): 2.10.1 (binary) - TensorFlow-Addons version and...

digital-thinking

**Describe the feature and the current behavior/state.** Addition of interpolators beyond bilinear in resampler_ops. As detailed in #2535, gradients of the bilinear interpolation fail when applied to an integer-pixel warp....

andrevitorelli

Fixes #2774 TQDMProgressBar(leave_overall_progress=False) now works with EarlyStopping Callback

1

# Description Brief Description of the PR: Fixes #2774 ## Type of change - [x] Bug fix - [ ] New Tutorial - [ ] Updated or additional documentation -...

PhorstenkampFuzzy

**System information** - Widnwos10: - TensorFlow version and how it was installed (source or binary): 2.9.1 binary - TensorFlow-Addons version and how it was installed (source or binary): 0.17.1 -...

PhorstenkampFuzzy

Using tfa.seq2seq.BeamSearchDecoder with transformer architecture.

3

**Describe the feature and the current behavior/state.** I am new to deep learning and tensorflow and recently I created a model similar to the [transformer model](https://www.tensorflow.org/tutorials/text/transformer) tutorial. Although the tutorial...

pritam-dey3

Restore single class MatthewsCorrelationCoefficient test cases

1

# Description This PR restores single class tests for `MatthewsCorrelationCoefficient`. These tests [have been removed](https://github.com/tensorflow/addons/pull/2406/files#diff-e829e6b4a6d8a09df73c9f183a92d8ba1e81f81c2e40b56ab6410a663267d885L39-R47) in #2406 which unfortunately broke `MatthewsCorrelationCoefficient` with `num_classes=1`. This was a breaking change which doesn't...

lgeiger

Errors when using tf.keras.Input & tfa.seq2seq in eager mode

8

**System information** - OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Windows 10 - TensorFlow version and how it was installed (source or binary): 2.2.0, binary - TensorFlow-Addons version and...

hccho2

BasicDecoderOutput with None lengths in networks_seq2seq_nmt example

1

# URL(s) with the issue: https://www.tensorflow.org/addons/tutorials/networks_seq2seq_nmt # Description of issue: It is not clear what the output of the BasicDecoder is. The output variable is is from type BasicDecoderOutput and...

soerenetler

If we run the evaluation as a different script, then loading from checkpoint doesn't set all the weights to all layers. Only AttentionMechanism memory_layer is being set. After ` checkpoint.restore(tf.train.latest_checkpoint(checkpoint_dir))`...

thepurpleowl

Metrics don't work with categorical / one-hot outputs

5

Apologies, this might actually be my egregious lack of experience in this domain. I've set up a little network to play with: ```python # xs_training.shape is (255722, 6, 7) #...

lnicola