keras-io
keras-io copied to clipboard
Keras documentation, hosted live at keras.io
when you run examples/rl/deep_q_network_breakout.py, you will find that the memory leak. even when the buff reach its max lenght (max_memory_length), memory still will increase.
After training prior with PixelCNN, during inference probabilities of codebook indices are calculated. There's a distribution sampler used to pick/sample a index instead of choosing the highest probability index- ```...
### Issue Type Bug ### Source source ### Keras Version Keras 2.13.1 ### Custom Code Yes ### OS Platform and Distribution Linux Ubuntu 20.04.5 LTS ### Python version 3.8.18 ###...
Following this tutorial, [Abstractive Summarization with Hugging Face Transformers](https://github.com/keras-team/keras-io/blob/master/examples/nlp/t5_hf_summarization.py) I created a text summarization ml model by fine-tuning t5-small with a custom dataset setting `MAX_INPUT_LENGTH = 1024`. But if I...
Add a new example using DeepLabV3+ pretrained model for semantic segmentation in healthcare Google colab : https://colab.research.google.com/drive/17UQr7-UM_Et0A7g7638Gq7Vb-e1171kC?usp=sharing
https://github.com/keras-team/keras-io/blob/master/examples/audio/transformer_asr.py In the code at the above link, I found that `source_maxlen` is defaulted to `100` in the transformer. The problem, though, is that the inputs are actually padded to...
Hi, I have added my code example for multiclass image segmentation using an Ensemble approach of modified U-nets. I am eagerly waiting for the response.
changes in base.html keros_core_master py files
# resolves https://github.com/keras-team/keras-io/issues/1518 and https://github.com/keras-team/keras-io/issues/1526 # Issue: https://github.com/keras-team/keras-io/issues/1518 (AttributeError: 'DataFrame' object has no attribute 'append') # Description This pull request addresses issues https://github.com/keras-team/keras-io/issues/1518 and https://github.com/keras-team/keras-io/issues/1526 by replacing the append function...
Fixes https://github.com/keras-team/keras-io/issues/1579. This PR adds read-only permissions to continuous_integration.yml. It also adds some future-proofing to the other workflows by moving write permissions to job-level instead of top-level, keeping only read-only...