This project is mirrored from https://review.mlplatform.org/tosa/specification.
Pull mirroring updated .
- Mar 23, 2021
-
-
Eric Kunze authored
Update copyright to include 2021. Change-Id: Iced31530f7f86faa2efa95441bcbacf1c37f126f Signed-off-by:
Eric Kunze <eric.kunze@arm.com>
-
- Mar 18, 2021
-
-
Eric Kunze authored
Integer divide is not commonly used, but would have been difficult to implement using the existing operators. Signed-off-by:
Eric Kunze <eric.kunze@arm.com> Change-Id: I8fb3919cd7a0f1a1fa95074d921d200d23e2f249
-
- Mar 16, 2021
-
-
Eric Kunze authored
It is challenging to work back from the RESCALE->TABLE->RESCALE for a simple 8-bit lookup. This adds support to TOSA for a simple direct 256 entry 8-bit lookup table for the TABLE operator. The 16-bit interpolated table is still required for a conformant implementation. Change-Id: I0ef218444f0b57b880aa8d1c7e96efedae72eb53 Signed-off-by:
Eric Kunze <eric.kunze@arm.com>
-
Eric Kunze authored
This also adds a restriction on CONCAT to have all inputs be the same rank. Update the ARGMAX pseudocode to handle corner cases properly and use axis/rank consistent with other operators. Signed-off-by:
Eric Kunze <eric.kunze@arm.com> Change-Id: I941ac172ee79424ac04d562cedebb75da76659a5
-
- Mar 11, 2021
-
-
Eric Kunze authored
They are all now [source,c++] and use ---- as the delimiter. Also made the pseudocode indentation consistent (4 spaces) in the introduction sections. Signed-off-by:
Eric Kunze <eric.kunze@arm.com> Change-Id: I1287a60cfeae8b7e3c63b9fc8885b1ebf3d3386d
-
Eric Kunze authored
When reading tensor values with zero point, the returned value has one more bit than the original to account for zero point. Update cases of apply_clip to properly represent the types involved. Change-Id: I60c17b1b244c34b4f04f042807936ae0f282ce93
-
- Mar 08, 2021
-
-
Eric Kunze authored
Attempt to get consistent across the pseudocode. Change the data types to all be intN_t instead of some cases of intN. Use float_t as the general floating point data type. Be consistent on use of the term "floating-point" Move general pseudocode helpers to their own section. Change-Id: Ie77666cd3ee438c71f39c62b9c424fe687b0bb51 Signed-off-by:
Eric Kunze <eric.kunze@arm.com>
-
- Mar 03, 2021
-
-
Eric Kunze authored
Originally only a pair of input tensors were allowed. This aligns with what the frameworks do, and simplifies the translation of frameworks into TOSA and allows compilers to view the output as a whole. Change-Id: Id4461abdf0cc763e84e086142222e87d28cd8afc Signed-off-by:
Eric Kunze <eric.kunze@arm.com>
-
- Feb 09, 2021
-
-
Dominic Symes authored
Add batch dimension for consistency with other operators. Change-Id: I9b1734a1a60304f46a14a6cda1bd6be6678f1037 Signed-off-by:
Dominic Symes <dominic.symes@arm.com>
-
- Feb 05, 2021
-
-
Dominic Symes authored
Add pseudo-code for the control flow operations in terms of TOSA sub-graphs. Signed-off-by:
Dominic Symes <dominic.symes@arm.com> Change-Id: I1712e8297a0010a4e68a34df16fcbcf47fc41dd2
-
- Jan 27, 2021
-
-
Eric Kunze authored
Elementwise operators no longer scale their inputs to a common range. The elementwise introductory section reflected the old behavior. Also clear up some language on the unary functions. Signed-off-by:
Eric Kunze <eric.kunze@arm.com> Change-Id: I86bf9da8b51e9a64e4fe6766e01f0c35d43d805a
-
- Jan 26, 2021
-
-
Eric Kunze authored
Change-Id: I5903ccd3edbd5c28bf8282ff39aadd648613f072 Signed-off-by:
Eric Kunze <eric.kunze@arm.com>
-
- Jan 25, 2021
-
-
Eric Kunze authored
The aint8 type represented an asymmetrically quantized range. With the change to move scaling fully into the RESCALE operator, the aint8 and int8 types didn't have significant differences in their use with operators. Unifying to a single data type makes things simpler. Signed-off-by:
Eric Kunze <eric.kunze@arm.com> Change-Id: I3ee89cbdafdc61293ef1f1bec52398d744e716a1
-
- Jan 13, 2021
-
-
Dominic Symes authored
This clarifies that repeated updates to the same output index are not permitted in TOSA SCATTER. Change-Id: Ib188a434d0d4ad4742ee37373491c8a53d501bf0 Signed-off-by:
Dominic Symes <dominic.symes@arm.com>
-
- Dec 16, 2020
-
-
Eric Kunze authored
scale_t is already defined globally for the rescaling. Move pseudocode for count_leading_zeros to introduction, and use it in the implementation of CLZ. Change-Id: I2453bce93b6dd25e870b8a010fc62af4c001cbf9
-
- Nov 30, 2020
-
-
Eric Kunze authored
Profile for EXP operator TANH table formatting Remove dilation from TRANSPOSE_CONV2D Change-Id: I2a265fecfb25ad997fdc992ad8542000014ac7c0
-
- Nov 13, 2020
-
-
Tom Cooksey authored
The new license grants the rights needed to create a completely independent implementation of the specification. Signed-off-by:
Tom Cooksey <tom.cooksey@arm.com> Change-Id: I1ca2cedf83c3f31df49096ecb6faaae060c3dc91
-
- Nov 10, 2020
-
-
Eric Kunze authored
Signed-off-by:
Eric Kunze <eric.kunze@arm.com> Change-Id: Ie2d83c66600cb01f39992c6b0c168fea2917330c
-
Eric Kunze authored
Bring the overview and goals up to date. Incorporate a section on finding the source to the spec and building it. Signed-off-by:
Eric Kunze <eric.kunze@arm.com> Change-Id: I8c862f8e58b01091d5561296702bcae83a8517e9
-
Eric Kunze authored
Try to make the different modes clearer, especially the integer BILINEAR mode. Change-Id: I6a2ab6ee4ed9e6a76d13bec4498e213a7f20b91f Signed-off-by:
Eric Kunze <eric.kunze@arm.com>
-
Dominic Symes authored
Add the scatter operation to the gather/scatter section. Signed-off-by:
Dominic Symes <dominic.symes@arm.com> Change-Id: I9baaef91bf70eae3b13e6e585df6c4821a0c1a93
-
- Nov 06, 2020
-
-
Dominic Symes authored
Add the operation code for GATHER and reduce number of arguments. Signed-off-by:
Dominic Symes <dominic.symes@arm.com> Change-Id: I51b6ede43dcae1e1aec7a11a75a1e4bf6c668673
-
Eric Kunze authored
Add values attribute to CONST operator. Remove stale references to input tensor 0 and 1. Remove out_t types where in and out are guaranteed to be the same type. Signed-off-by:
Eric Kunze <eric.kunze@arm.com> Change-Id: I75a8eb4dee67afbee3652d9e937aa0b82318dbd0
-
Eric Kunze authored
Was out_t, which wasn't in the data types table. Switch to acc_t, the size of the accumulator, which is the output type for the convolutions. Also added some types into the pseudocode to clarify what types variables are. Signed-off-by:
Eric Kunze <eric.kunze@arm.com> Change-Id: I210220199c2d39b16938094022339286df040545
-
- Nov 05, 2020
-
-
Eric Kunze authored
Comparison ops, similar to the other elementwise ops, can operate on tensors of any rank. For those operators with rank limitations, make them consistent and remove "dims". Signed-off-by:
Eric Kunze <eric.kunze@arm.com> Change-Id: I33eb00122cec4c6625f080f9ed640f1d874fae87
-
- Nov 03, 2020
-
-
Eric Kunze authored
Currently only enabled for the accumulator of a convolution and the input to a RESCALE. Signed-off-by:
Eric Kunze <eric.kunze@arm.com> Change-Id: Ic5474124b4a8735f12ef9c20c2be48e55767a3fc
-
Eric Kunze authored
Weights remain as an attribute for inference profile devices. Change-Id: I1cdbbd2ccd95fc1b7afc0f05a1425fb53560b432 Signed-off-by:
Eric Kunze <eric.kunze@arm.com>
-
- Oct 29, 2020
-
-
Dominic Symes authored
Clarify how the TABLE operator can be used in pratice to generate int8 to int8 and int16 to int16 lookup operations. Change-Id: I97154bb8cea46ab1fe48bf1cda615ff164130645 Signed-off-by:
Dominic Symes <dominic.symes@arm.com>
-
- Oct 28, 2020
-
-
Dominic Symes authored
Add floating point as an allowed data type for RESIZE. Signed-off-by:
Dominic Symes <dominic.symes@arm.com> Change-Id: I01766edabe922d395175356387ea325bfc301fd3
-
Dominic Symes authored
The result of 32x32 elementwise multiply exceeds the int32_t result type range. This change adds a right scaling shift argument to shift down the result. Change-Id: I6ae17e6dc3fe342d052304533158ad2d0e7bb7be Signed-off-by:
Dominic Symes <dominic.symes@arm.com>
-
- Oct 27, 2020
-
-
Dominic Symes authored
Add rounding mode to shift right and clarify that shift values cannot be negative. Signed-off-by:
Dominic Symes <dominic.symes@arm.com> Change-Id: I7995c8dfa6f16f9ca4dcbf5241692fcd940e509b
-
- Oct 22, 2020
-
-
Dominic Symes authored
Complete the function naming change from dimensions() to rank(). Fix MATMUL to use apply_add() for additions, consistent with the other operations. Signed-off-by:
Dominic Symes <dominic.symes@arm.com> Change-Id: I6ba84cf8b016505e8477b04dd00f2a2bf3194492
-
- Oct 16, 2020
-
-
Eric Kunze authored
rank() is consistent with the rest of the specification. Also add a bit of whitespace in the pseudocode for easier reading. Fix one typo. Change-Id: Ifa794b2c332cb70cd52a141d9af219e26bd3dca0 Signed-off-by:
Eric Kunze <eric.kunze@arm.com>
-
- Oct 14, 2020
-
-
Dominic Symes authored
Simplify the reduction and reverse reference by apply one axis at a time and default to keep_dims. (Changes to shape can be done separately with RESHAPE) Signed-off-by:
Dominic Symes <dominic.symes@arm.com> Change-Id: I590bdb6bc05b1c673f86e3f45f0a43536d8f362a
-
- Oct 13, 2020
-
-
Eric Kunze authored
Contains build requirements and short instructions. Change-Id: Id2cc62b4a88adab53b0c7314e64221936c6d55dc Signed-off-by:
Eric Kunze <eric.kunze@arm.com>
-
- Oct 12, 2020
-
-
Dominic Symes authored
Fix typing errors in the resize operation and simplify the assert checking. Signed-off-by:
Dominic Symes <dominic.symes@arm.com> Change-Id: If12b9e54a6e9f84f0d6122be7d4a4b567fd81e0f
-
- Oct 08, 2020
-
-
Eric Kunze authored
REDUCE_ANY/ALL PLACEHOLDER/IDENTITY* Signed-off-by:
Eric Kunze <eric.kunze@arm.com> Change-Id: I3436a91b8137fa4fce9b8c2027cd669cfbdcd539
-
- Oct 02, 2020
-
-
Eric Kunze authored
Change-Id: I7c528ebf7e78759ada4bb951b2471c367d3f4525
-
- Oct 01, 2020
-
-
Eric Kunze authored
-