adrien
91c5c41fc5
Working comparaison for TensorAlloc
2026-05-24 21:23:20 +02:00
adrien
ba671ee486
Working basic eq between TensorAlloc
2026-05-24 18:40:14 +02:00
adrien
09d6ca1ff5
Renamed Tensor to TensorStatic
2026-05-23 10:29:57 +02:00
adrien
bcd888d59e
Splited main.zig into lib and test.zig
2026-05-20 23:02:30 +02:00
adrien
63e9b6b63d
Removed vs native benchmarks
2026-05-20 22:58:36 +02:00
adrien
5f833a5e58
Removed TensorAlloc and made TensorStatic back to just Tensor
...
Realized I can just do alloc.create instead of a all new struct
2026-05-15 00:46:22 +02:00
adrien
00e0f5ab73
Moved isTensor to shared + added isTensorAlloc/Static
2026-05-15 00:32:58 +02:00
adrien
f67e9d709d
Working add TensorAlloc
2026-05-15 00:24:39 +02:00
adrien
e6d0f62929
TensorAlloc add and to compilable but still error for basic add test
2026-05-14 22:25:35 +02:00
adrien
f702c1e09a
slice can now use null value like python [2:]
2026-05-14 10:56:26 +02:00
adrien
b959f5f28a
Added slice to TensorStatic
2026-05-14 01:28:24 +02:00
adrien
6ba1e664c1
Started to add TensorAlloc
2026-05-14 00:53:05 +02:00
adrien
6559ed9f62
Removed comptime to parseSegment
2026-05-12 09:06:28 +02:00
adrien
55906e0ab7
Small fixs
2026-05-11 17:12:37 +02:00
adrien
7d28de2028
Removed comptime for str UnitParser
2026-05-04 23:55:00 +02:00
adrien
5bdc78c065
Simplified pow
2026-05-04 22:57:53 +02:00
adrien
18830c8b45
Fixed benchmark
2026-05-04 22:25:18 +02:00
adrien
4595397e70
Removed the feature where you can use comptime int or float ar rhs for operation
2026-05-04 22:10:55 +02:00
adrien
7844aacfce
Added a UnitParser to get Dimensions and Scales from a str
2026-05-04 19:10:06 +02:00
adrien
4d275dca2d
Renamed Tensor to TensorStatic to later introduce TensorAlloc and TensorGPU
2026-04-29 18:07:13 +02:00
adrien
9635cfb481
Changed self: Self to self: *const SElf in tensor for performance
2026-04-28 16:06:13 +02:00
adrien
f0029449f0
tmp
2026-04-28 14:50:08 +02:00
adrien
8816a65518
Now pass all test with new *const way
...
I am not quite sure about it yet, but it is faster sooo idk.
Let's see long term
2026-04-28 13:51:10 +02:00
adrien
26ff02c50f
Changed TEnsor to use *const
2026-04-28 13:10:14 +02:00
adrien
bb6dd59b9a
Removed more
2026-04-28 01:02:30 +02:00
adrien
acb908a448
Removed some char
2026-04-28 01:01:40 +02:00
adrien
c6f613a787
I guess I can't do better. Scalar still suck, but at least it is builtin SIDM so ok I guess
2026-04-27 22:07:39 +02:00
adrien
a0961e7571
Start to optimize the shit out of it, still a long way to go
...
After that GPPPPPUUUUUU baby!
2026-04-27 21:24:03 +02:00
adrien
168312b78e
Removed lots of usless inline and comptime in Scales and Dimensions
2026-04-27 19:09:55 +02:00
AdrienBouvais
698e968ef8
Added a high dimension benchmark
2026-04-27 16:18:31 +02:00
AdrienBouvais
44aaa8a8b2
Removed all inline for (0..total) for either builtin or for loop without inline
...
This is to prevent giant binary for Tensor with a lot of Scalar
2026-04-27 16:11:46 +02:00
AdrienBouvais
cd954b379b
Added cross to Tensor + fix benchmark
2026-04-27 15:13:15 +02:00
AdrienBouvais
16d25e7e7e
Added shape comptime check for Tensor add/sub/div/mul
2026-04-27 14:45:41 +02:00
AdrienBouvais
f37a196b15
Fixed new Tensor to be everything (Scalar, Vector, Matrix and above)
2026-04-27 09:11:24 +02:00
Adrien Bouvais
934a40fe1a
Basic untested tensor
2026-04-26 22:16:25 +02:00
adrien
4a2d45b384
Fix benchmarks to work with new Scalar and Vector
2026-04-26 01:28:33 +02:00
adrien
d32de3fe82
Converted Base test to new Scalar
2026-04-26 01:18:41 +02:00
adrien
4b01dfe412
Replaced Vector with Quantity
Deploy MkDocs to Garage / build-and-deploy (push) Failing after 25s
2026-04-26 01:07:12 +02:00
adrien
c350ffd1d6
Working vector initiate
Deploy MkDocs to Garage / build-and-deploy (push) Failing after 30s
2026-04-26 00:40:10 +02:00
adrien
a591736b19
Totally replaced Scalar with Quantity
Deploy MkDocs to Garage / build-and-deploy (push) Failing after 49s
2026-04-26 00:29:00 +02:00
adrien
0257ee5d21
Added Quantity that will be both Scalar and Vector, using only @Vector
...
I did some benchmark, and @Vector(1, type) is just as fast as type for
float and around 40% slower for int.
But the traide off is ok for me, because for @Vector(>1, type), we are
between 2 and 32x faster! And I dont think single value int will be use
for high performance computing, so I am ok with how simple it make
everything.
Because now I dont have to one struct that use @Vector and one that use
just type, I can just use one. Making it even easier to integrate GPU
later. I will even be able to have a single place in my code with a + -
* and / basically. A single function for all, so I only need to optimize
that one.
2026-04-25 22:36:48 +02:00
adrien
7fc7193206
FLush
2026-04-25 18:57:38 +02:00
adrien
3a4783cfa1
Benchmark SIMD vs Native
2026-04-25 18:47:13 +02:00
adrien
80cdcd134e
Added some function to benchmark
2026-04-24 23:50:16 +02:00
adrien
f26f6086ca
Added Imperial unit scales
2026-04-23 00:14:57 +02:00
adrien
5e7b7c5302
Added some constants
2026-04-22 23:53:43 +02:00
adrien
719679aabc
Updated README + Renamed mulBy and divBy to mul div
2026-04-22 23:33:58 +02:00
adrien
2c94df7f4a
Added overflaw security + made benchmark work (set to max)
2026-04-22 23:11:42 +02:00
adrien
5efa42c2e1
Made possible to use comptime float, int and T for Scalar operation of Vector
2026-04-22 22:34:46 +02:00
adrien
1129acc542
Same for Scalar
2026-04-22 22:10:57 +02:00
adrien