https://github.com/halide/Halide

sort by:
Revision Author Date Message Commit Date
f9e4c78 Add support for dynamic tensors to hannk (#5942) * Initial support for Dynamic Tensors in hannk * Update hannk_delegate.cpp * Fixes * Smarten Tensor::resize() 27 April 2021, 19:14:54 UTC
01101b2 Add support for LSTM to the hannk delegate (#5943) * Add support for LSTM to the hannk delegate * clang-format 27 April 2021, 19:08:06 UTC
66fa093 Various fixes and improvements. 27 April 2021, 18:34:50 UTC
dd8e0bf More H::R::B overhead cleanup. 27 April 2021, 18:16:33 UTC
5f7fd8d Reduce overhead from H::R::B 27 April 2021, 17:59:17 UTC
285b0f2 clang-format 26 April 2021, 22:44:18 UTC
ba4c0bd Merge branch 'master' of github.com:halide/Halide into interpret_nn 26 April 2021, 21:13:15 UTC
5a69e9f Fix flattening of ramps involving 64-bit mins (#5940) * Fix flattening of ramps involving 64-bit mins. * Use make_const instead of cast. 26 April 2021, 21:12:53 UTC
d24dd20 clang-format 26 April 2021, 21:11:55 UTC
630667f Fix comments. 26 April 2021, 20:53:21 UTC
bf57f35 Fix build breakage. 26 April 2021, 20:44:07 UTC
124cd8b Fix LSTM op 26 April 2021, 20:40:04 UTC
91e42f4 Don't use as_const_int() on temporaries (#5939) Sometimes we get lucky and it's still valid, but it's always wrong. 26 April 2021, 20:10:21 UTC
ad105b2 Fix some precision and scheduling issues of LSTM 26 April 2021, 19:04:48 UTC
b1730c3 Merge branch 'interpret_nn' of github.com:halide/Halide into dsharletg/lstm 26 April 2021, 18:34:56 UTC
b6a392d Merge branch 'interpret_nn' of github.com:halide/Halide into interpret_nn 26 April 2021, 18:34:31 UTC
a6919f5 Fix Interpreter::inputs and outputs. 26 April 2021, 18:34:19 UTC
8c02d6f Merge branch 'master' into interpret_nn 26 April 2021, 18:09:44 UTC
e34380d Merge branch 'interpret_nn' into dsharletg/lstm 26 April 2021, 18:01:20 UTC
1b3cbcb [Hexagon] Try vdelta/vrdelta before vlut for some shuffles. (#5935) The patch tries to generate vdelta/vrdelta instructions for non-ramp shuffles. Eg: shuffle(lut_expr, < 0, 1, 2, 3, 4, 5, 6, 7, 9, 10, 11, 12, 13, 14, 15, 16, 18, 19, 20, 21, 22, 23, 24, 25, 27, 28, 29, 30, 31, 32, 33, 34, 36, 37, 38, 39, 40, 41, 42, 43, 45, 46, 47, 48, 49, 50, 51, 52, 54, 55, 56, 57, 58, 59, 60, 61, 63, 64, 65, 66, 67, 68, 69, 70>) can be generated using vrdelta. The patch also fixes a bug where we bitcast vdelta/vrdelta with 16/32 bits elements to wrong type. User would see the below error: llvm-project/llvm/lib/IR/Instructions.cpp:2905: static llvm::CastInst *llvm::CastInst::Create(Instruction::CastOps, llvm::Value *, llvm::Type *, const llvm::Twine &, llvm::Instruction *): Assertion `castIsValid(op, S, Ty) && "Invalid cast!"' failed. Co-authored-by: Ankit Aggarwal <aankit@quicinc.com> 26 April 2021, 17:55:12 UTC
80737bf Merge branch 'master' of github.com:halide/Halide into interpret_nn 26 April 2021, 17:54:13 UTC
4eebc9d WIP LSTM 26 April 2021, 17:09:27 UTC
a6855c4 clang-format 25 April 2021, 21:29:35 UTC
c48f992 Optimize fully connected a lot 25 April 2021, 20:34:30 UTC
b03d45c Report profiler after each test 25 April 2021, 20:33:20 UTC
0ff9c62 Move elementwise generators to the same file. 25 April 2021, 20:33:09 UTC
ba89623 Small Typo fix in lesson 06 (#5936) Signed-off-by: xgupta <shivam98.tkg@rediffmail.com> 23 April 2021, 16:19:40 UTC
361a637 Fix accidental widening of shift by a constant. 23 April 2021, 16:08:18 UTC
23ea6dc Clean up precision of transcendentals 23 April 2021, 07:24:12 UTC
27301a7 Merge branch 'interpret_nn' of github.com:halide/Halide into interpret_nn 23 April 2021, 06:04:34 UTC
13565a0 Enable approx_exp2 for int16 results without overflow. 23 April 2021, 05:26:27 UTC
b40f8de Merge branch 'master' into interpret_nn 22 April 2021, 21:51:54 UTC
fc9b3a9 Save compare_vs_tflite outputs from first run (not post-benchmark) 22 April 2021, 21:51:45 UTC
bb0bd88 clang-format 22 April 2021, 17:17:56 UTC
05e6087 Fix asserts. 22 April 2021, 16:58:40 UTC
25229ca Fix typo. 22 April 2021, 16:49:12 UTC
a407acd Revert "Temporarily disable hanging test (#5925)" (#5933) This reverts commit 62505857694ab8af2a88a22edf291e630c8c0cfd. 22 April 2021, 16:29:01 UTC
345c1f5 Fix common_halide test. 22 April 2021, 02:11:07 UTC
5da61f2 clang-format 21 April 2021, 22:40:46 UTC
cc01208 Merge branch 'interpret_nn' of github.com:halide/Halide into interpret_nn 21 April 2021, 22:38:59 UTC
9e884ad Avoid horrific clang-format suggestion. 21 April 2021, 22:38:37 UTC
66bbe58 Update normalizations_generator.cpp 21 April 2021, 22:34:03 UTC
8eec221 Merge branch 'dsharletg/interpret_nn2' of github.com:halide/Halide into interpret_nn 21 April 2021, 22:28:27 UTC
1ba73cb clang-format 21 April 2021, 22:26:35 UTC
b22fa6d Add Greater, GreaterEqual to delegate 21 April 2021, 22:22:54 UTC
acc3659 Remove unused functions 21 April 2021, 22:22:43 UTC
78151f9 Merge branch 'master' of github.com:halide/Halide into interpret_nn 21 April 2021, 22:14:41 UTC
fb13fb0 Add mul_shift_right intrinsic and related improvements (#5916) * Add multiply_quantized intrinsic * clang-format * Fix build on some compilers. * Fix incorrect saturating_pmulhrs * multiply_quantized -> mul_shift_right * Remove workaround and just cast shift amounts. * Fix error message * Fix declaration of mul_shift_right. 21 April 2021, 22:10:32 UTC
6867005 Suppress Metal unused function warning (#5913) Co-authored-by: Steven Johnson <srj@google.com> 21 April 2021, 19:06:50 UTC
e06a312 Merge branch 'master' of github.com:halide/Halide into dsharletg/interpret_nn2 21 April 2021, 18:07:48 UTC
5435a0b Merge branch 'dsharletg/interpret_nn2' of https://github.com/halide/Halide into dsharletg/interpret_nn2 21 April 2021, 17:23:27 UTC
d08848d Minor cleanup in ReshapeOp::execute() 21 April 2021, 17:23:24 UTC
5dd85ae Let the user pass the Func to use to the reduction helpers (#5929) * Let the user pass the Func to use to the reduction helpers * Pass Funcs by const ref 21 April 2021, 16:50:56 UTC
17d4771 Update test to reflect behavior we expect. (#5928) 21 April 2021, 16:04:27 UTC
087567f Remove old codegen. LLVM rewrites this back to a multiply anyways. (#5930) 21 April 2021, 16:04:09 UTC
5f2e47b Minor cleanups 21 April 2021, 16:02:27 UTC
03d75f4 Some of these aren't elementwise. 21 April 2021, 01:47:34 UTC
fe13d68 Generically handle elementwise operations of any rank. 21 April 2021, 01:39:25 UTC
af25cad Add Shape op. 21 April 2021, 01:05:15 UTC
07100b9 Merge branch 'master' of github.com:halide/Halide into dsharletg/interpret_nn2 21 April 2021, 00:49:47 UTC
f9c4604 Merge branch 'dsharletg/interpret_nn2' of github.com:halide/Halide into dsharletg/interpret_nn2 21 April 2021, 00:40:32 UTC
06a1d85 Fix details of ReshapeOp to match tflite's impl 21 April 2021, 00:40:00 UTC
6250585 Temporarily disable hanging test (#5925) * Temporarily disable hanging test LLVM13 is causing vector_reductions to hang (https://reviews.llvm.org/D100099 appears to be the injection point). Disabling this test to unbreak the buildbots. * Update vector_reductions.cpp 20 April 2021, 21:23:26 UTC
c1de142 [adams2019] Add caching to autoscheduler (#5697) * add feature caching and block caching to adams2019 autoscheduler * added caching verification for feautures * add caching docstrings 20 April 2021, 21:21:33 UTC
7949541 Merge branch 'interpret_nn' into dsharletg/interpret_nn2 20 April 2021, 16:42:43 UTC
a5979ce Merge branch 'master' into interpret_nn 20 April 2021, 16:42:29 UTC
ac23987 Speed up simd_op_check by only compiling one pipeline per op (#5918) * Speed up simd_op_check and compute_with * Dense vector loads can be written many different ways. 20 April 2021, 15:02:14 UTC
74e33f3 Merge branch 'master' of github.com:halide/Halide into dsharletg/interpret_nn2 20 April 2021, 01:20:49 UTC
6963673 Add Target::ARMv81a and improve shift instruction selection (#5917) * Add Target::ARMv81a and improve shift instruction selection. * Remove merge mistake. * Don't use ARM intrinsic on arm32, it seems to be missing sometimes. 20 April 2021, 00:24:06 UTC
1966f7d Merge branch 'dsharletg/interpret_nn2' of github.com:halide/Halide into dsharletg/interpret_nn2 19 April 2021, 19:05:34 UTC
188ae1a Remove unnecessary check for log2(0) 19 April 2021, 19:04:58 UTC
3ffd55c kTfLiteError -> kTfLiteDelegateError 19 April 2021, 16:48:09 UTC
a245de5 Merge branch 'interpret_nn' into dsharletg/interpret_nn2 19 April 2021, 16:27:45 UTC
a859bd2 Merge branch 'dsharletg/interpret_nn2' of https://github.com/halide/Halide into dsharletg/interpret_nn2 19 April 2021, 16:27:04 UTC
f3e69e3 Merge branch 'master' into interpret_nn 19 April 2021, 16:26:48 UTC
b054c7a Merge branch 'master' of github.com:halide/Halide into dsharletg/interpret_nn2 18 April 2021, 02:57:14 UTC
526a75a Remove multiply_quantized helper. 17 April 2021, 22:53:32 UTC
493dbd4 Comment out specialiations for f64x2.convert_low_i32x4_s/u (#5914) LLVM removed the primitives we need (so our code can't be used), but it also doesn't seem to be generating the expected instructions directly (as claimed). Commenting out to un-break tests; issue has been reported to wasm/llvm team. 17 April 2021, 17:46:20 UTC
8a2c738 Rearrange input parameters. 17 April 2021, 06:44:15 UTC
2e4b9b9 Don't benchmark when testing. 17 April 2021, 06:43:39 UTC
3ff489e Fix softmax correctness issues 17 April 2021, 06:27:04 UTC
0d3d044 Move pooling ops to the same generator file. 17 April 2021, 03:34:01 UTC
f95121f Put the common case specialization first. 17 April 2021, 02:12:10 UTC
2eebd21 Remove now-pointless specialization helper. 17 April 2021, 02:10:54 UTC
586240b Allow broadcasting of c of input 2 17 April 2021, 02:05:54 UTC
42a1d64 Update model.h 17 April 2021, 01:17:13 UTC
20b9a5b Merge branch 'dsharletg/interpret_nn2' of github.com:halide/Halide into dsharletg/interpret_nn2 17 April 2021, 00:51:57 UTC
f2928f2 Remove scalpel left in patient 17 April 2021, 00:33:03 UTC
0a7a1a6 Add TODO 17 April 2021, 00:32:26 UTC
a9c74f4 Lazy-init seeds in compare_vs_tflite, in case use_hannk=0 17 April 2021, 00:31:45 UTC
dbddd33 Merge branch 'dsharletg/interpret_nn2' of github.com:halide/Halide into dsharletg/interpret_nn2 16 April 2021, 23:15:54 UTC
bf50dba Merge branch 'interpret_nn' of github.com:halide/Halide into dsharletg/interpret_nn2 16 April 2021, 23:15:37 UTC
c4a2e26 Fix excessive precision in softmax. 16 April 2021, 23:00:11 UTC
1bea8c8 Remove dead file. 16 April 2021, 22:49:05 UTC
916e5ec Add comment. 16 April 2021, 22:33:46 UTC
d972c4e Merge branch 'interpret_nn' into dsharletg/interpret_nn2 16 April 2021, 22:24:04 UTC
c10cf3e Merge branch 'master' into interpret_nn 16 April 2021, 22:23:49 UTC
9cdb4aa Simplify and improve cuda_mat_mul schedule (#5909) * Simplify and improve cuda_mat_mul schedule 16 April 2021, 22:23:30 UTC
f3ba059 Merge branch 'master' into interpret_nn 16 April 2021, 22:22:48 UTC
a41cce7 Basic support of predicated loads/stores in C++ backend (#5908) * Basic support of predicated load/stores in C++ backend * Fix formatting and maybe build * Fix * trigger buildbots Co-authored-by: Steven Johnson <srj@google.com> 16 April 2021, 20:47:16 UTC
back to top