data
data copied to clipboard
A PyTorch repo for data loading and utilities to be shared by the PyTorch domain libraries.
Summary: If no iterator is created in the middle of load_state_dict and state_dict calls, we should be able to directly return the original state dict without triggering reading service because...
### 🐛 Describe the bug Hello, I found that a standard DataLoader takes unreasonably long to construct itself and to load the first batch if there is a filed in...
### 🐛 Describe the bug Prefetcher will hang indefinitely on shutdown(), the faulthandler stack traces indicates that main thread is blocked on https://github.com/pytorch/data/blob/main/torchdata/datapipes/iter/util/prefetcher.py#L113 while child thread is blocked on https://github.com/pytorch/data/blob/main/torchdata/datapipes/iter/util/prefetcher.py#L81,...
### Changes - change `portalocker>=2/0.0` to `portalocker>=2.0.0`
### 🐛 Describe the bug We are noticing a strange error specifically when using torch2.1.1+cu118 and torch2.1.2+cu118 , that is not an issue with torch2.1.0+cu118. The error looks like this:...
### 🐛 Describe the bug iterating trough a data pipe, generated to a random split iters correctly trough all the data it is supposed to , but unfortunately it does...
### 📚 The doc issue The default for S3 buffer size is 128 MB - or 128 * (1024**2) https://github.com/pytorch/data/blob/a5b4720dece60565788ac4c9a85e01719188b28e/torchdata/csrc/pybind/S3Handler/S3Handler.cpp#L15 The example for S3FileLoaderIterDataPipe uses a buffer_size of 256. https://github.com/pytorch/data/blob/a5b4720dece60565788ac4c9a85e01719188b28e/torchdata/datapipes/iter/load/s3io.py#L154...
since release of mypy 0.981 recursive types are supported; i have just removed the `#` as per suggestion in the TODO comment and have changed the mypy version in the...
This updates README.md with latest pytorch version 2.1.0 and torchdata version 0.7.0