[MOD-8200] [MOD-8202] INT8 index #566

meiravgri · 2024-12-10T16:29:45Z

This PR introduces support for vectors with int8 elements.
When using the VecSimMetric_Cosine distance metric, vectors will include the norm of the vector at the end of the blob after preprocessing.

Index factory

Added an additional member to AbstractIndexInitParams: dataSize

This represents the size of the vector to be stored or queried after preprocessing.
Typically, it equals dim * sizeof(datatype), except for int8 with the VecSimMetric_Cosine metric, where it will be dim * sizeof(datatype) + sizeof(float).

Spaces normalization

Introduced a compute norm API:

float IntegralType_ComputeNorm<integral type>(vec, dim): Accumulates the norm into an int variable and returns the square root of the sum as a float.
GetNormalizeFunc<int8_t>: Returns a function that stores the norm at the end of the vector. It expects a blob large enough to store the norm at the end.

API utils

VecSimParams_GetDataSize Returns the data size according to the vector type and metric.
Removed normalize_func from VecSimIndexAbstract class members as it was redundant.

unit tests

Introduced unit tests for int8.
Added CommonTypeMetricTests suite with tests that run with all possible {VecSimType, VecSimMetric} combinations on all index types.
- Currently includes test_datasize and test_initial_size_estimation.

add cosine to spaces fix typos in calculator

add spaces unit tests for int8 L2 add compilation flags introduce tests/utils for general utils

change INITIALIZE_BENCHMARKS_SET to INITIALIZE_BENCHMARKS_SET_L2_IP introduce INITIALIZE_BENCHMARKS_SET_COSINE fix typos in Choose_INT8_L2_implementation_AVX512F_BW_VL_VNNI name

change create_int8_vec to populate_int8_vec add compute norm

minimal dim = 32

codecov · 2024-12-12T12:26:59Z

Codecov Report

Attention: Patch coverage is 96.05263% with 3 lines in your changes missing coverage. Please review.

Project coverage is 97.05%. Comparing base (e38fc0b) to head (21520ad).
Report is 1 commits behind head on meiravg_feature_int_uint_8.

Files with missing lines	Patch %	Lines
src/VecSim/index_factories/brute_force_factory.cpp	92.85%	1 Missing ⚠️
src/VecSim/index_factories/hnsw_factory.cpp	94.44%	1 Missing ⚠️
src/VecSim/index_factories/tiered_factory.cpp	83.33%	1 Missing ⚠️

Additional details and impacted files

@@                      Coverage Diff                       @@
##           meiravg_feature_int_uint_8     #566      +/-   ##
==============================================================
+ Coverage                       96.91%   97.05%   +0.13%     
==============================================================
  Files                             103      104       +1     
  Lines                            5442     5496      +54     
==============================================================
+ Hits                             5274     5334      +60     
+ Misses                            168      162       -6

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

… CreateIndexComponents: pass bool is_normalized get distnce function according to original metric get pp according to is_normalized && metric == VecSimMetric_Cosine, and remove this logic from the indexes factories. add dataSize member to AbstractIndexInitParams add VecSimType_INT8 type introduce VecSimParams_GetDataSize: returns datasize introduce and implement GetNormalizeFunc<int8_t> thtat returns int8_normalizeVector int8_normalizeVector computes the norm and stores it at the emd of argument vector.

remove normalize_func from VecSimIndexAbstract members tests: int8 unit test create int8 indexes unit_test_utils: CalcIndexDataSize: casts VecSimIndex * to VecSimIndexAbstract<dist_t, data_t> * and calls VecSimIndexAbstract<dist_t, data_t>::getDataSize() cast_to_tiered_index<data_t, dist_t>: takes VecSimIndex * ans casts to TieredHNSWIndex<data_t, dist_t> *

2 new function to test_utils:: CreateTieredParams CreateNewTieredHNSWIndex add test_initial_size_estimation to CommonTypeMetricTests use CommonTypeMetricTieredTests for tiered tests

add int8 to * VecSimDebug_GetElementNeighborsInHNSWGraph * VecSim_Normalize *HNSW NewIndex from file

… meiravg_compute_norm

move ASSERT_DEBUG_DEATH of CalcIndexDataSize to a separate test

remove test_normalize.cpp file

avoid evaluating statement in typeid to avoid clang warnig

alonre24

Looking good!!
Few comments, mainly organizational

tests/unit/test_hnsw_tiered.cpp

tests/unit/CMakeLists.txt

tests/unit/unit_test_utils.cpp

tests/unit/test_common.cpp

tests/unit/test_int8.cpp

move components tests from test_common to test_components

src/VecSim/index_factories/components/components_factory.h

meiravgri · 2024-12-22T10:34:52Z

will backport the entire feature to 8.0

* [MOD-8198] Introduce INT8 distance functions (#560) * naive implementation of L2 * update * implment naive disatnce for int8 add cosine to spaces fix typos in calculator * imp choose L2 int8 with 256bit loop add spaces unit tests for int8 L2 add compilation flags introduce tests/utils for general utils * imp space bm for int8 change INITIALIZE_BENCHMARKS_SET to INITIALIZE_BENCHMARKS_SET_L2_IP introduce INITIALIZE_BENCHMARKS_SET_COSINE fix typos in Choose_INT8_L2_implementation_AVX512F_BW_VL_VNNI name * fix INITIALIZE_BENCHMARKS_SET_L2_IP and add include to F_BW_VL_VNNI * rename unit/test_utuils to unit_test_utils * seed create vec * format * implmenet IP + unit test * ip bm * format * implement cosine in ip API change create_int8_vec to populate_int8_vec add compute norm * use mask sub instead of msk load * loop size = 512 minimal dim = 32 * add int8 to bm * reanme to simd64 * convert to int before multiplication * review comments: align to vector size ncluding the norm in cosine dist unit test cover small dim in cosine chooser * use sizeof(float)instead of 4 * remove int conversion in test_utils::compute_norm * REVERT!!! malicious test to see if we get to the code * assert dummt * fix alignemnt test * remove assert * remove cosine alignment * Override missing intrinsincs in gcc <11 (#572) * override _mm256_loadu_epi8 with mm256_maskz_loadu_epi8 if gcc < 11 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95483 * fix * disable flow temp * add comment * [MOD-8200] [MOD-8202] INT8 index (#566) * naive implementation of L2 * update * implment naive disatnce for int8 add cosine to spaces fix typos in calculator * imp choose L2 int8 with 256bit loop add spaces unit tests for int8 L2 add compilation flags introduce tests/utils for general utils * imp space bm for int8 change INITIALIZE_BENCHMARKS_SET to INITIALIZE_BENCHMARKS_SET_L2_IP introduce INITIALIZE_BENCHMARKS_SET_COSINE fix typos in Choose_INT8_L2_implementation_AVX512F_BW_VL_VNNI name * fix INITIALIZE_BENCHMARKS_SET_L2_IP and add include to F_BW_VL_VNNI * rename unit/test_utuils to unit_test_utils * seed create vec * format * implmenet IP + unit test * ip bm * format * implement cosine in ip API change create_int8_vec to populate_int8_vec add compute norm * use mask sub instead of msk load * loop size = 512 minimal dim = 32 * add int8 to bm * reanme to simd64 * convert to int before multiplication * introduce IntegralType_ComputeNorm * move preprocessor logic to choose if cosine preprocessor is needed to CreateIndexComponents: pass bool is_normalized get distnce function according to original metric get pp according to is_normalized && metric == VecSimMetric_Cosine, and remove this logic from the indexes factories. add dataSize member to AbstractIndexInitParams add VecSimType_INT8 type introduce VecSimParams_GetDataSize: returns datasize introduce and implement GetNormalizeFunc<int8_t> thtat returns int8_normalizeVector int8_normalizeVector computes the norm and stores it at the emd of argument vector. * add int8 tests * fix include unint_test_utils * add int 8 to index factories remove normalize_func from VecSimIndexAbstract members tests: int8 unit test create int8 indexes unit_test_utils: CalcIndexDataSize: casts VecSimIndex * to VecSimIndexAbstract<dist_t, data_t> * and calls VecSimIndexAbstract<dist_t, data_t>::getDataSize() cast_to_tiered_index<data_t, dist_t>: takes VecSimIndex * ans casts to TieredHNSWIndex<data_t, dist_t> * * add EstimateInitialSize for int8 to indexes factories 2 new function to test_utils:: CreateTieredParams CreateNewTieredHNSWIndex add test_initial_size_estimation to CommonTypeMetricTests use CommonTypeMetricTieredTests for tiered tests * add int8 unit tests add int8 to * VecSimDebug_GetElementNeighborsInHNSWGraph * VecSim_Normalize *HNSW NewIndex from file * remove duplicated GetDistFunc<int8_t, float> move ASSERT_DEBUG_DEATH of CalcIndexDataSize to a separate test * remove assert test, the statement is excuted and causes crash * imporve normalize test * rename test_utils::compute_norm -> test_utils::integral_compute_norm remove test_normalize.cpp file * use stack allocation instead of heap allocation in tests * fix float comparison in test_serialization avoid evaluating statement in typeid to avoid clang warnig * renae CalcIndexDataSize -> CalcVectorDataSize move components tests from test_common to test_components * add comment to INSTANTIATE_TEST_SUITE_P * [MOD-8206] INT8 flow tests (#573) * test_hnsw.py intiital * int8 hnsw tests * general tests class * flow_bruteforce.py: introduce GeneralTest call from TestINT8 common.py: introduce create_flat_index create_add_vectors move fp32_expand_and_calc_cosine_dist to common.py * tiered flow tests: * add optional create_data_func to IndexCtx, use for special datatypes *inntroduce test_create_int8 and test_search_insert_int8 create_int8_vectors expectes shape (tuple) * use query.flat * revert using flat (not helping in int8) fix float16 calling query.flat * revert changes in Data class in bf tests revert test_bf_float16_range_query change * fix merge

* [MOD-8198] Introduce INT8 distance functions (#560) * naive implementation of L2 * update * implment naive disatnce for int8 add cosine to spaces fix typos in calculator * imp choose L2 int8 with 256bit loop add spaces unit tests for int8 L2 add compilation flags introduce tests/utils for general utils * imp space bm for int8 change INITIALIZE_BENCHMARKS_SET to INITIALIZE_BENCHMARKS_SET_L2_IP introduce INITIALIZE_BENCHMARKS_SET_COSINE fix typos in Choose_INT8_L2_implementation_AVX512F_BW_VL_VNNI name * fix INITIALIZE_BENCHMARKS_SET_L2_IP and add include to F_BW_VL_VNNI * rename unit/test_utuils to unit_test_utils * seed create vec * format * implmenet IP + unit test * ip bm * format * implement cosine in ip API change create_int8_vec to populate_int8_vec add compute norm * use mask sub instead of msk load * loop size = 512 minimal dim = 32 * add int8 to bm * reanme to simd64 * convert to int before multiplication * review comments: align to vector size ncluding the norm in cosine dist unit test cover small dim in cosine chooser * use sizeof(float)instead of 4 * remove int conversion in test_utils::compute_norm * REVERT!!! malicious test to see if we get to the code * assert dummt * fix alignemnt test * remove assert * remove cosine alignment * Override missing intrinsincs in gcc <11 (#572) * override _mm256_loadu_epi8 with mm256_maskz_loadu_epi8 if gcc < 11 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95483 * fix * disable flow temp * add comment * [MOD-8200] [MOD-8202] INT8 index (#566) * naive implementation of L2 * update * implment naive disatnce for int8 add cosine to spaces fix typos in calculator * imp choose L2 int8 with 256bit loop add spaces unit tests for int8 L2 add compilation flags introduce tests/utils for general utils * imp space bm for int8 change INITIALIZE_BENCHMARKS_SET to INITIALIZE_BENCHMARKS_SET_L2_IP introduce INITIALIZE_BENCHMARKS_SET_COSINE fix typos in Choose_INT8_L2_implementation_AVX512F_BW_VL_VNNI name * fix INITIALIZE_BENCHMARKS_SET_L2_IP and add include to F_BW_VL_VNNI * rename unit/test_utuils to unit_test_utils * seed create vec * format * implmenet IP + unit test * ip bm * format * implement cosine in ip API change create_int8_vec to populate_int8_vec add compute norm * use mask sub instead of msk load * loop size = 512 minimal dim = 32 * add int8 to bm * reanme to simd64 * convert to int before multiplication * introduce IntegralType_ComputeNorm * move preprocessor logic to choose if cosine preprocessor is needed to CreateIndexComponents: pass bool is_normalized get distnce function according to original metric get pp according to is_normalized && metric == VecSimMetric_Cosine, and remove this logic from the indexes factories. add dataSize member to AbstractIndexInitParams add VecSimType_INT8 type introduce VecSimParams_GetDataSize: returns datasize introduce and implement GetNormalizeFunc<int8_t> thtat returns int8_normalizeVector int8_normalizeVector computes the norm and stores it at the emd of argument vector. * add int8 tests * fix include unint_test_utils * add int 8 to index factories remove normalize_func from VecSimIndexAbstract members tests: int8 unit test create int8 indexes unit_test_utils: CalcIndexDataSize: casts VecSimIndex * to VecSimIndexAbstract<dist_t, data_t> * and calls VecSimIndexAbstract<dist_t, data_t>::getDataSize() cast_to_tiered_index<data_t, dist_t>: takes VecSimIndex * ans casts to TieredHNSWIndex<data_t, dist_t> * * add EstimateInitialSize for int8 to indexes factories 2 new function to test_utils:: CreateTieredParams CreateNewTieredHNSWIndex add test_initial_size_estimation to CommonTypeMetricTests use CommonTypeMetricTieredTests for tiered tests * add int8 unit tests add int8 to * VecSimDebug_GetElementNeighborsInHNSWGraph * VecSim_Normalize *HNSW NewIndex from file * remove duplicated GetDistFunc<int8_t, float> move ASSERT_DEBUG_DEATH of CalcIndexDataSize to a separate test * remove assert test, the statement is excuted and causes crash * imporve normalize test * rename test_utils::compute_norm -> test_utils::integral_compute_norm remove test_normalize.cpp file * use stack allocation instead of heap allocation in tests * fix float comparison in test_serialization avoid evaluating statement in typeid to avoid clang warnig * renae CalcIndexDataSize -> CalcVectorDataSize move components tests from test_common to test_components * add comment to INSTANTIATE_TEST_SUITE_P * [MOD-8206] INT8 flow tests (#573) * test_hnsw.py intiital * int8 hnsw tests * general tests class * flow_bruteforce.py: introduce GeneralTest call from TestINT8 common.py: introduce create_flat_index create_add_vectors move fp32_expand_and_calc_cosine_dist to common.py * tiered flow tests: * add optional create_data_func to IndexCtx, use for special datatypes *inntroduce test_create_int8 and test_search_insert_int8 create_int8_vectors expectes shape (tuple) * use query.flat * revert using flat (not helping in int8) fix float16 calling query.flat * revert changes in Data class in bf tests revert test_bf_float16_range_query change * fix merge (cherry picked from commit babfbe0)

[MOD-8198] Introduce INT8 (#560) (#571) * [MOD-8198] Introduce INT8 distance functions (#560) * naive implementation of L2 * update * implment naive disatnce for int8 add cosine to spaces fix typos in calculator * imp choose L2 int8 with 256bit loop add spaces unit tests for int8 L2 add compilation flags introduce tests/utils for general utils * imp space bm for int8 change INITIALIZE_BENCHMARKS_SET to INITIALIZE_BENCHMARKS_SET_L2_IP introduce INITIALIZE_BENCHMARKS_SET_COSINE fix typos in Choose_INT8_L2_implementation_AVX512F_BW_VL_VNNI name * fix INITIALIZE_BENCHMARKS_SET_L2_IP and add include to F_BW_VL_VNNI * rename unit/test_utuils to unit_test_utils * seed create vec * format * implmenet IP + unit test * ip bm * format * implement cosine in ip API change create_int8_vec to populate_int8_vec add compute norm * use mask sub instead of msk load * loop size = 512 minimal dim = 32 * add int8 to bm * reanme to simd64 * convert to int before multiplication * review comments: align to vector size ncluding the norm in cosine dist unit test cover small dim in cosine chooser * use sizeof(float)instead of 4 * remove int conversion in test_utils::compute_norm * REVERT!!! malicious test to see if we get to the code * assert dummt * fix alignemnt test * remove assert * remove cosine alignment * Override missing intrinsincs in gcc <11 (#572) * override _mm256_loadu_epi8 with mm256_maskz_loadu_epi8 if gcc < 11 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95483 * fix * disable flow temp * add comment * [MOD-8200] [MOD-8202] INT8 index (#566) * naive implementation of L2 * update * implment naive disatnce for int8 add cosine to spaces fix typos in calculator * imp choose L2 int8 with 256bit loop add spaces unit tests for int8 L2 add compilation flags introduce tests/utils for general utils * imp space bm for int8 change INITIALIZE_BENCHMARKS_SET to INITIALIZE_BENCHMARKS_SET_L2_IP introduce INITIALIZE_BENCHMARKS_SET_COSINE fix typos in Choose_INT8_L2_implementation_AVX512F_BW_VL_VNNI name * fix INITIALIZE_BENCHMARKS_SET_L2_IP and add include to F_BW_VL_VNNI * rename unit/test_utuils to unit_test_utils * seed create vec * format * implmenet IP + unit test * ip bm * format * implement cosine in ip API change create_int8_vec to populate_int8_vec add compute norm * use mask sub instead of msk load * loop size = 512 minimal dim = 32 * add int8 to bm * reanme to simd64 * convert to int before multiplication * introduce IntegralType_ComputeNorm * move preprocessor logic to choose if cosine preprocessor is needed to CreateIndexComponents: pass bool is_normalized get distnce function according to original metric get pp according to is_normalized && metric == VecSimMetric_Cosine, and remove this logic from the indexes factories. add dataSize member to AbstractIndexInitParams add VecSimType_INT8 type introduce VecSimParams_GetDataSize: returns datasize introduce and implement GetNormalizeFunc<int8_t> thtat returns int8_normalizeVector int8_normalizeVector computes the norm and stores it at the emd of argument vector. * add int8 tests * fix include unint_test_utils * add int 8 to index factories remove normalize_func from VecSimIndexAbstract members tests: int8 unit test create int8 indexes unit_test_utils: CalcIndexDataSize: casts VecSimIndex * to VecSimIndexAbstract<dist_t, data_t> * and calls VecSimIndexAbstract<dist_t, data_t>::getDataSize() cast_to_tiered_index<data_t, dist_t>: takes VecSimIndex * ans casts to TieredHNSWIndex<data_t, dist_t> * * add EstimateInitialSize for int8 to indexes factories 2 new function to test_utils:: CreateTieredParams CreateNewTieredHNSWIndex add test_initial_size_estimation to CommonTypeMetricTests use CommonTypeMetricTieredTests for tiered tests * add int8 unit tests add int8 to * VecSimDebug_GetElementNeighborsInHNSWGraph * VecSim_Normalize *HNSW NewIndex from file * remove duplicated GetDistFunc<int8_t, float> move ASSERT_DEBUG_DEATH of CalcIndexDataSize to a separate test * remove assert test, the statement is excuted and causes crash * imporve normalize test * rename test_utils::compute_norm -> test_utils::integral_compute_norm remove test_normalize.cpp file * use stack allocation instead of heap allocation in tests * fix float comparison in test_serialization avoid evaluating statement in typeid to avoid clang warnig * renae CalcIndexDataSize -> CalcVectorDataSize move components tests from test_common to test_components * add comment to INSTANTIATE_TEST_SUITE_P * [MOD-8206] INT8 flow tests (#573) * test_hnsw.py intiital * int8 hnsw tests * general tests class * flow_bruteforce.py: introduce GeneralTest call from TestINT8 common.py: introduce create_flat_index create_add_vectors move fp32_expand_and_calc_cosine_dist to common.py * tiered flow tests: * add optional create_data_func to IndexCtx, use for special datatypes *inntroduce test_create_int8 and test_search_insert_int8 create_int8_vectors expectes shape (tuple) * use query.flat * revert using flat (not helping in int8) fix float16 calling query.flat * revert changes in Data class in bf tests revert test_bf_float16_range_query change * fix merge (cherry picked from commit babfbe0) Co-authored-by: meiravgri <109056284+meiravgri@users.noreply.github.com>

meiravgri added 18 commits December 10, 2024 16:09

naive implementation of L2

2f34c15

update

c641d23

implment naive disatnce for int8

1c5eb90

add cosine to spaces fix typos in calculator

imp choose L2 int8 with 256bit loop

fa8e9ff

add spaces unit tests for int8 L2 add compilation flags introduce tests/utils for general utils

imp space bm for int8

a7a556f

change INITIALIZE_BENCHMARKS_SET to INITIALIZE_BENCHMARKS_SET_L2_IP introduce INITIALIZE_BENCHMARKS_SET_COSINE fix typos in Choose_INT8_L2_implementation_AVX512F_BW_VL_VNNI name

fix INITIALIZE_BENCHMARKS_SET_L2_IP and add include to F_BW_VL_VNNI

43064e8

rename unit/test_utuils to unit_test_utils

fb9f1cc

seed create vec

602f8e9

format

cde5e2d

implmenet IP + unit test

cdb4d7f

ip bm

5f01890

format

2dce6f0

implement cosine in ip API

3d3b375

change create_int8_vec to populate_int8_vec add compute norm

use mask sub instead of msk load

6f211b3

loop size = 512

6ac65a3

minimal dim = 32

add int8 to bm

0d07c5d

reanme to simd64

3586a76

convert to int before multiplication

adbc4d7

meiravgri added 3 commits December 15, 2024 05:42

introduce IntegralType_ComputeNorm

03be854

add int8 tests

c32e4fb

meiravgri force-pushed the meiravg_compute_norm branch from 97a7d5c to c32e4fb Compare December 15, 2024 05:49

meiravgri changed the base branch from meiravg_feature_int_uint_8 to meiravg_int8_dist_func December 15, 2024 05:54

meiravgri added 4 commits December 15, 2024 05:54

fix include unint_test_utils

de6769d

Merge branch 'meiravg_feature_int_uint_8' into meiravg_compute_norm

3e3c14f

add EstimateInitialSize for int8 to indexes factories

d5b7d27

2 new function to test_utils:: CreateTieredParams CreateNewTieredHNSWIndex add test_initial_size_estimation to CommonTypeMetricTests use CommonTypeMetricTieredTests for tiered tests

Base automatically changed from meiravg_int8_dist_func to meiravg_feature_int_uint_8 December 17, 2024 05:47

add int8 unit tests

ef9beb6

add int8 to * VecSimDebug_GetElementNeighborsInHNSWGraph * VecSim_Normalize *HNSW NewIndex from file

meiravgri added 4 commits December 17, 2024 16:04

Merge remote-tracking branch 'origin/meiravg_feature_int_uint_8' into…

0f2b65d

… meiravg_compute_norm

remove duplicated GetDistFunc<int8_t, float>

b7d6aed

move ASSERT_DEBUG_DEATH of CalcIndexDataSize to a separate test

remove assert test, the statement is excuted and causes crash

939cc47

imporve normalize test

e3ad80c

meiravgri changed the title ~~introduce IntegralType_ComputeNorm~~ [MOD-8200] [MOD-8202] introduce IntegralType_ComputeNorm Dec 18, 2024

meiravgri changed the title ~~[MOD-8200] [MOD-8202] introduce IntegralType_ComputeNorm~~ [MOD-8200] [MOD-8202] INT8 index Dec 18, 2024

rename test_utils::compute_norm -> test_utils::integral_compute_norm

58fa8e2

remove test_normalize.cpp file

meiravgri requested review from alonre24 and GuyAv46 December 18, 2024 10:31

meiravgri added the backport 8.0 label Dec 18, 2024

meiravgri added 2 commits December 18, 2024 12:11

use stack allocation instead of heap allocation in tests

736e30b

fix float comparison in test_serialization

4a9bb69

avoid evaluating statement in typeid to avoid clang warnig

alonre24 reviewed Dec 19, 2024

View reviewed changes

meiravgri added 2 commits December 19, 2024 13:00

renae CalcIndexDataSize -> CalcVectorDataSize

9796d59

move components tests from test_common to test_components

add comment to INSTANTIATE_TEST_SUITE_P

21520ad

meiravgri requested a review from alonre24 December 19, 2024 13:14

alonre24 approved these changes Dec 19, 2024

View reviewed changes

GuyAv46 reviewed Dec 22, 2024

View reviewed changes

src/VecSim/index_factories/components/components_factory.h Show resolved Hide resolved

GuyAv46 approved these changes Dec 22, 2024

View reviewed changes

meiravgri merged commit d17629c into meiravg_feature_int_uint_8 Dec 22, 2024
32 checks passed

meiravgri deleted the meiravg_compute_norm branch December 22, 2024 10:34

RedisAI deleted a comment from github-actions bot Dec 22, 2024

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[MOD-8200] [MOD-8202] INT8 index #566

[MOD-8200] [MOD-8202] INT8 index #566

Uh oh!

meiravgri commented Dec 10, 2024 •

edited

Loading

Uh oh!

codecov bot commented Dec 12, 2024 •

edited

Loading

Uh oh!

alonre24 left a comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

meiravgri commented Dec 22, 2024

Uh oh!

Uh oh!

[MOD-8200] [MOD-8202] INT8 index #566

[MOD-8200] [MOD-8202] INT8 index #566

Uh oh!

Conversation

meiravgri commented Dec 10, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Index factory

Spaces normalization

API utils

unit tests

Uh oh!

codecov bot commented Dec 12, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

alonre24 left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

meiravgri commented Dec 22, 2024

Uh oh!

Uh oh!

meiravgri commented Dec 10, 2024 •

edited

Loading

codecov bot commented Dec 12, 2024 •

edited

Loading