jacketsj

Results 5 comments of jacketsj

feat: cuda/cuvs acceleration for PQ training/assignment

switched branch, new pr at https://github.com/lancedb/lance/pull/2946

lance.torch.data.LanceDataset + torch.utils.data.DataLoader is slower than to_batches + frombuffer

Thanks for the idea, although I'm currently convinced that it's from [this function](https://github.com/lancedb/lance/blob/main/python/python/lance/torch/data.py#L34), which copies data into numpy as an intermediary (at least in 0.16.1). That's similar to what I...

AttributeError: 'NoneType' object has no attribute 'get' on `dataset.filter`

Try `split == 'test'`

AttributeError: 'NoneType' object has no attribute 'get' on `dataset.filter`

I'm not sure if it does what you're expecting from `.filter`, but `ds.to_table(filter="split == 'test'")` should work if `split` is a column.

AttributeError: 'NoneType' object has no attribute 'get' on `dataset.filter`

I get the same error. Putting a simple repro here: ``` import lance import pyarrow as pa table = pa.Table.from_pylist([{"name": "Alice", "age": 20}, {"name": "Bob", "age": 30}]) lance.write_dataset(table, "./alice_and_bob.lance", mode="overwrite")...