Joppe Geluykens

A dataset with a split column with missing 0 (train) values, yields the following error during Ludwig preprocessing: ```python-traceback File "/home/ray/anaconda3/lib/python3.8/site-packages/ludwig/api.py", line 434, in train preprocessed_data = self.preprocess( File "/home/ray/anaconda3/lib/python3.8/site-packages/ludwig/api.py",...

Character recognition

1

Does anyone know of a good model to train on characters from a digital font (think Google Vision API)? Use case is to recognize characters written on order receipts in...

AutoML: stratify imbalanced datasets

1

Adds stratified splitting to the AutoML generated config if the dataset: - has a single output feature - the output feature is categorical or binary - the output feature's imbalance...

additional benchmark configs

2

Adds configs to the benchmark directory for datasets: - amazon employee access challenge - mushroom edibility - naval - numerai 28.6 - sarcos

XLNet fails with attn_type "uni"

2

### System Info - `transformers` version: 4.26.1 - Platform: Linux-5.10.104-linuxkit-aarch64-with-glibc2.17 - Python version: 3.8.16 - Huggingface_hub version: 0.12.0 - PyTorch version (GPU?): 1.13.0 (False) - Tensorflow version (GPU?): not installed...

Async client to backoff when model overloaded

1

### Feature request Have the (async) client automatically backoff sending requests when the deployment is overloaded. ### Motivation When the async client exceeds the deployment queue capacity / rate limits,...

Allow early stop based on walltime since improvement

1

This PR: - stores the Posix timestamp of the last improvement in ProgressTracker - adds a convenience callback implementing `should_early_stop` that returns True if the given timeout has been reached...