issues
search
pytorch
/
tensordict
TensorDict is a pytorch dedicated tensor container.
MIT License
832
stars
74
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
[Refactor] Remove non-ascii characters from code base
#928
vmoens
closed
3 months ago
1
[BUG] Non-unicode code causes torch.compile to fail
#927
gmmyung
closed
3 months ago
2
[Test] Test non-tensor propagation in arithmetic ops
#926
vmoens
closed
3 months ago
2
[BugFix] Fix vmap compatibility with torch<2.2
#925
vmoens
closed
3 months ago
2
[BugFix] Fix compile + vmap
#924
vmoens
closed
3 months ago
2
[Feature] Symmetric errors in arithmetic ops
#923
vmoens
closed
3 months ago
2
[Doc] Add docstrings to elementwise arithmetic ops
#922
vmoens
closed
3 months ago
2
[Feature] default for add, sub, pow, clamp_min, clamp_max, div and mul
#921
vmoens
closed
3 months ago
2
[Feature] `reduce` arg in reductions
#920
vmoens
closed
3 months ago
2
[Performance] Faster multithreaded pin_memory
#919
vmoens
closed
3 months ago
2
[Doc] Fix symbolic trace reference in doc
#918
vmoens
closed
3 months ago
2
[BugFix] Fix torch version assertion
#917
vmoens
closed
3 months ago
2
[Refactor] Refactor is_dynamo_compile imports
#916
vmoens
closed
3 months ago
2
[Quality] non_blocking_pin instead of pin_memory
#915
vmoens
closed
3 months ago
2
[Performance] Faster to-module
#914
vmoens
closed
3 months ago
0
[BugFix] use as_subclass in Buffer
#913
vmoens
closed
3 months ago
2
[Performance] Faster getattr in TC
#912
vmoens
closed
3 months ago
2
[Feature] from_modules expand_identical kwarg
#911
vmoens
closed
3 months ago
2
[Feature] add_custom_mapping and NPE refactors
#910
vmoens
closed
3 months ago
2
[Feature] ``split_keys``
#909
vmoens
closed
3 months ago
2
[Feature] flatten_keys and unflatten_keys as context managers
#908
vmoens
closed
3 months ago
2
[BugFix] Fix (keys, values) in sub
#907
vmoens
closed
3 months ago
2
[BugFix] Allow any tensorclass to have a data field
#906
vmoens
closed
3 months ago
2
[Feature] construct tds with kwargs
#905
vmoens
closed
3 months ago
2
[Feature] grad and data for tensorclasses
#904
vmoens
closed
3 months ago
2
[BugFix] fix construction of lazy stacks from tds
#903
vmoens
closed
3 months ago
2
[BugFix] fix _expand_to_match_shape for single bool tensor
#902
vmoens
closed
3 months ago
2
[Feature] `zero_grad` and `requires_grad_`
#901
vmoens
closed
3 months ago
0
[Benchmark] Fix recursion and cache errors in benchmarks
#900
vmoens
closed
3 months ago
1
[BugFix] Fix bool indexing of lazy stacks when sub-tds are involved
#899
vmoens
closed
3 months ago
0
[Quality] Better `__repr__` for tensorclasses
#898
vmoens
closed
3 months ago
0
[Quality] Better repr of keys
#897
vmoens
closed
3 months ago
0
[Feature] layout for get_nestedtensor
#896
vmoens
closed
3 months ago
0
[Feature] TD.new_tensor
#895
vmoens
closed
3 months ago
0
[Feature] TD.new_full
#894
vmoens
closed
3 months ago
0
[Feature] TD.new_empty
#893
vmoens
closed
3 months ago
0
[Feature] TD.new_ones
#892
vmoens
closed
3 months ago
0
[Feature] TD.new_zeros
#891
vmoens
closed
3 months ago
0
[Formatting] Lint revamp
#890
vmoens
closed
3 months ago
2
[Feature] lazy-stack with different TD shapes
#889
vmoens
closed
3 months ago
0
[BugFix] Fix lazy stack indexing and split with empty slices
#888
vmoens
closed
3 months ago
0
[BugFix] Fix lazy stack indexing and split with empty slices
#887
vmoens
closed
3 months ago
0
[Quality] zip-strict when possible
#886
vmoens
closed
3 months ago
2
Add `_check_dim_name` to `tensorclass`
#885
kurtamohler
closed
3 months ago
1
[BugFix] Fix `pad_sequence` behavior for non-tensor attributes of tensorclass
#884
kurtamohler
closed
3 months ago
0
[Feature] Compile - benchmarks
#883
vmoens
closed
3 months ago
1
[Feature] Compile - tensorclass compatibility
#882
vmoens
closed
3 months ago
2
[Feature] Compile - nn compatibility
#881
vmoens
closed
3 months ago
2
[Performance] Faster tensorclass set
#880
vmoens
closed
3 months ago
2
[Feature] Compile - benchmarks
#879
vmoens
closed
3 months ago
0
Previous
Next