Releases: activeloopai/deeplake
Releases · activeloopai/deeplake
v3.8.3 🌈
v3.8.2 🌈
v3.8.1 🌈
🧭 What's Changed
- Added sample_indices api to virtual tensors. (#2654) @khustup
- Deep memory typo fix (#2653) @adolkhan
- VectorStore docs fix (#2655) @adolkhan
- Saving views of views (#2649) @FayazRahman
- Brought back index regeneration (#2648) @istranic
- Switch lz4 implementation to python-lz4 (#2645) @FayazRahman
⚙️ Who Contributes
v3.8.0 🌈
🧭 What's Changed
- Deep Memory search speed up (#2646) @adolkhan
- Deep Memory virtual tensors support (#2643) @adolkhan
- Better view saving error message (#2639) @FayazRahman
- Bump libdeeplake version. (#2642) @khustup
⚙️ Who Contributes
@FayazRahman, @adolkhan, @davidbuniat, @khustup, @levongh and @nvoxland
v3.7.3 🌈
🧭 What's Changed
- Deep Memory support on aws (#2635) @adolkhan
- [AL-2417] Fix transform memory leak (#2631) @FayazRahman
- [AL-2419] [AL-2416] Fixes for
tensor.path
andtensor.creds_key
(#2620) @FayazRahman - [AL-2422] Tensor update by
__setattr__
(#2632) @FayazRahman - Deep memory support in VectorStore (#2594) @adolkhan
⚙️ Who Contributes
v3.7.2 🌈
v3.7.1 🌈
v3.7.0 🌈
🧭 What's Changed
- Index regeneration on maintenance (#2489) @sounakr
- Ability to release resources associated with a DataLoader (#2611) @levongh
- Removed typing_extensions usage (#2610) @nvoxland
- test_vdb_indexes.py::Test case fix. (#2614) @sounakr
- added verbose argument to ds.dataloader (#2608) @levongh
- [AL-2413] Iterative implementation of merging node maps (#2600) @FayazRahman
- [AL-2414]
ds.extend
for transforms (#2606) @FayazRahman - Cache size propagete to libdeeplake. (#2607) @khustup
- Fixed query chaining. (#2609) @khustup
- enable python3.11 for linux dists (#2603) @levongh
- Fixed large file upload, by passing bytes instead of bytearray. (#2604) @khustup
- Set default embedding htype to float32 (#2595) @FayazRahman
⚙️ Who Contributes
@FayazRahman, @istranic, @khustup, @levongh, @nvoxland and @sounakr