Usearch

[2.8.16](https://github.com/unum-cloud/usearch/compare/v2.8.15...v2.8.16) (2024-01-24)


Docs

* Downloads numbers ([13cc624](https://github.com/unum-cloud/usearch/commit/13cc624a59627da424f79120da4582b22a2c450d))

Fix

* SimSIMD dispatch ([c8dc3b7](https://github.com/unum-cloud/usearch/commit/c8dc3b79945a3b5c282cb6384bd9d52b7052c1b5)), closes [#320](https://github.com/unum-cloud/usearch/issues/320)

Make

* Fix Node version for SemVer ([a68fca6](https://github.com/unum-cloud/usearch/commit/a68fca6d51b3ce157d9959e3b7f9aefae25bb506))
* Upgrade dependencies ([6ab2150](https://github.com/unum-cloud/usearch/commit/6ab2150d55c676f6cd836851dbf353725c2ca8b9))
* Upgrade Node environment ([f5b6750](https://github.com/unum-cloud/usearch/commit/f5b67501210736fc80dbc4752e600df7153a9e08))

[2.8.15](https://github.com/unum-cloud/usearch/compare/v2.8.14...v2.8.15) (2024-01-09)


Fix

* exporting quantization enums in Go (336) ([93c74c3](https://github.com/unum-cloud/usearch/commit/93c74c338ea31e6a2181f030c7389f5e2ef0da85)), closes [#336](https://github.com/unum-cloud/usearch/issues/336)




Hashes
* docs.tar.gz : `1083f4feafdcb4e2f58df423302cf0bf66cf55e3d03b14d902e462760a055a9f`
* usearch-v2.8.15.tar.gz : `551909e77acb8a83dff948ba7ae2689a8c87f11cb99c190b69b24d613e42ff1d`
* usearch-v2.8.15.zip : `1e4ed46061cf2b9bf426ed0b230cffd91ec8115a8744682d47350695100b2384`
* usearch_linux_amd_2.8.15.deb : `cdf91489cd1d74eaa8cdb4e599f2f10e93b7fa2ad9f06f31353d350ee573d43e`
* usearch_linux_arm_2.8.15.deb : `09090bfbf4d164995279022e4f0f47ac78cb1a20f4b07bbccdb8a078f5e6cb6b`
* usearch_macOS_arm64_2.8.15.zip : `8eb411e10183273ca5dfea133c4406f24d537b9e593293100786616da9085940`
* usearch_macOS_x86_64_2.8.15.zip : `08075a3bbd6ff327a92cfc8729d16f50113689fd689452ef7059fb87d16850c9`
* usearch_wasm_linux_arm64_2.8.15.tar.gz : `fdd51bc4ee2e8e9409dd1dce050d0c29938d05e10f9d353e55d0aa288a133ca8`
* usearch_wasm_linux_x86_64_2.8.15.tar.gz : `77624e1a46b2e34cdd10a0eb772fe3754b35867025543760812cb4c553542b60`
* usearch_wasm_macos_arm64_2.8.15.zip : `466700de9092d412651f526a93d597756387e9a5cc6f92ba72f057c2af903e68`
* usearch_wasm_macos_x86_64_2.8.15.zip : `44f60623ca6a48f0d63030a72f4a876c65fc43c2887ca8b9a7a6c7c94f035903`
* usearch_wasm_windows_x64_2.8.15.tar.gz : `834bb4f96bf557e8d167dc8c0cffe39348a2dc71e914c62b7a2c8acda9f54fd7`
* usearch_wasm_windows_x86_2.8.15.tar.gz : `b1eea6d42c47f49a0cc305e7e7bb31339360a6f4ed30f6b39924cf39704327b2`
* usearch_windows_x64_2.8.15.tar : `91ddee43f7aca3c8c2305bfc18a141d5581a7589b6262b2de99dc05f046dc53f`
* usearch_windows_x86_2.8.15.tar : `75895cc4cfe9a999fd3d448726a5f9b24c29c240dc4a6d8b29d51b70cc65f3a7`

As was discussed in the [SciPy integration thread](https://github.com/scipy/scipy/issues/19454), Python libraries use double-precision floating-point numbers by default.
So, in this release, I've extended the spatial distance functions in the underlying [SimSIMD](https://github.com/ashvardanian/simsimd) - `Cos`, `L2sq`, `IP` with support for `double` arguments with specialized implementations on AVX-512-capable x86 CPUs and SVE-capable Arm CPUs.

Benchmarking SimSIMD vs. SciPy on Intel Sapphire Rapids CPU

- Vector dimensions: 1536
- Vectors count: 1000
- Hardware capabilities: serial, x86_avx2, x86_avx512, x86_avx2fp16, x86_avx512fp16, x86_avx512vpopcntdq, x86_avx512vnni
- NumPy BLAS dependency: openblas64
- NumPy LAPACK dependency: dep140640983012528

Between 2 Vectors, Batch Size: 1

| Datatype | Method | Ops/s | SimSIMD Ops/s | SimSIMD Improvement |
| :------- | :-------------------- | -------------------: | -------------------: | ------------------: |
| `f64` | `scipy.cosine` | 63,612 | 572,605 | 9.00 x |
| `f64` | `scipy.sqeuclidean` | 238,547 | 915,596 | 3.84 x |
| `f64` | `numpy.inner` | 449,499 | 986,522 | 2.19 x |

Between 2 Vectors, Batch Size: 1,000

| Datatype | Method | Ops/s | SimSIMD Ops/s | SimSIMD Improvement |
| :------- | :-------------------- | -------------------: | -------------------: | ------------------: |
| `f64` | `scipy.cosine` | 68,962 | 1,457,172 | 21.13 x |
| `f64` | `scipy.sqeuclidean` | 247,727 | 1,535,547 | 6.20 x |
| `f64` | `numpy.inner` | 463,509 | 1,512,004 | 3.26 x |


Benchmarking SimSIMD vs. SciPy on AWS Graviton 3

- Vector dimensions: 1536
- Vectors count: 1000
- Hardware capabilities: serial, arm_neon, arm_sve
- NumPy BLAS dependency: openblas64
- NumPy LAPACK dependency: openblas64

Between 2 Vectors, Batch Size: 1

| Datatype | Method | Ops/s | SimSIMD Ops/s | SimSIMD Improvement |
| :------- | :-------------------- | -------------------: | -------------------: | ------------------: |
| `f64` | `scipy.cosine` | 40,729 | 725,382 | 17.81 x |
| `f64` | `scipy.sqeuclidean` | 160,812 | 728,114 | 4.53 x |
| `f64` | `numpy.inner` | 473,443 | 767,374 | 1.62 x |
| `f64` | `scipy.jensenshannon` | 15,684 | 38,528 | 2.46 x |
| `f64` | `scipy.kl_div` | 49,983 | 61,811 | 1.24 x |

Between 2 Vectors, Batch Size: 1,000

| Datatype | Method | Ops/s | SimSIMD Ops/s | SimSIMD Improvement |
| :------- | :-------------------- | -------------------: | -------------------: | ------------------: |
| `f64` | `scipy.cosine` | 41,130 | 1,460,850 | 35.52 x |
| `f64` | `scipy.sqeuclidean` | 162,147 | 1,486,255 | 9.17 x |
| `f64` | `numpy.inner` | 473,856 | 1,580,136 | 3.33 x |

Hashes
* docs.tar.gz : `def474428a4d67076e68dfd16b660a53bf51fad12af7e4c6ee77e1555b220b8f`
* usearch-v2.8.14.tar.gz : `6ae186618120b6c710ff3ed1bf31e9a58610e7b837bccdeae79000247c2b24a3`
* usearch-v2.8.14.zip : `a1da6b34bc23111926b16be9d36a7403988405a00c833417b27b8ccd9c70227f`
* usearch_linux_amd_2.8.14.deb : `ae4995f9504a9ab90921e3091a5aa6af432de647f1a6c835ee5cb2622dd2f8a3`
* usearch_linux_arm_2.8.14.deb : `dafcee294630b7c17adaed9aebb668d4cbfc5fe269f35f66b10bf458d66d899d`
* usearch_macOS_arm64_2.8.14.zip : `65e0a8d0259400e35de692c9afb3406d6bd4db5dd2f46632676890c41ba1537c`
* usearch_macOS_x86_64_2.8.14.zip : `1fa9a6e6983f5b6fe5dd2b82ce566262dcbd1d1e8671ec39b9d01ceddc3b80dd`
* usearch_wasm_linux_arm64_2.8.14.tar.gz : `edb5846ab0b38b1095b12f2b8ca771748394911bf6efd74df6d5ba66f36328d9`
* usearch_wasm_linux_x86_64_2.8.14.tar.gz : `84e4162db1dc83157f7a4032278c9cfc01a68ac59c4b1af1b5fc35e43905c515`
* usearch_wasm_macos_arm64_2.8.14.zip : `8c966ef2f5e425cf82f472bd4b74f0e7cd7fa39b860dd9f25f488370afc035fe`
* usearch_wasm_macos_x86_64_2.8.14.zip : `7e48cc5d0e34b586f36b86671c8410b2b2deca31962beb0ccf6230546ac42621`
* usearch_wasm_windows_x64_2.8.14.tar.gz : `02b1382c68fe8ef52d55da639a38ad15c7449d579230ef5ae78465776fd689fa`
* usearch_wasm_windows_x86_2.8.14.tar.gz : `cf922953ca61c8f9101fac33a2aa06d70fd09ef8b8c73d64a6623f73dcfe09cc`
* usearch_windows_x64_2.8.14.tar : `8a74a122fafae229f65df8de5a6e264eb16a7e5eba691629d6439b0b54ea8b74`
* usearch_windows_x86_2.8.14.tar : `96c6625fc2ca723cc44e864a75f5c72dc3ab9aeb5fe5f0f80ba32783121f6766`

[2.8.13](https://github.com/unum-cloud/usearch/compare/v2.8.12...v2.8.13) (2023-11-18)


Docs

* badges, comparisons with clustering tools ([05b3109](https://github.com/unum-cloud/usearch/commit/05b310959ae564bd7720c3fbcaa3695e23648b37))

Improve

* Faster exports for non-multi Indexes ([48655a3](https://github.com/unum-cloud/usearch/commit/48655a3201fc99b907593396683cdb41ca48fe41))




Hashes
* docs.tar.gz : `1da23a52fc6c7a458176e9cc35f76ee781bcd25b23b8cdd09cb37cbe015de8fa`
* usearch-v2.8.13.tar.gz : `7600c3a29761cce942ded60ab60596a1065d2a561399c56ab56fde78037bb1a5`
* usearch-v2.8.13.zip : `d7d39a8d954d45d901dd9975086e05d4db58ddf15a93573caa13d2953e801f3a`
* usearch_linux_amd_2.8.13.deb : `2b966a5d77c25c640e81f16ca831dc9673ac8ee24371e5349e94761f44f8f001`
* usearch_linux_arm_2.8.13.deb : `c0b147adfd9427b14547ac6de2f7bec02bcfde9ef2badadf4ec96867a39b9944`
* usearch_macOS_arm64_2.8.13.zip : `cd2de8d9656e8c69307945b7ef2fbc648a24ce06c13d08ca66cb7b532a42eb78`
* usearch_macOS_x86_64_2.8.13.zip : `2b44e47829edca468b4e0e5e4832ba0baefbf90bc402eba97078462f20519ced`
* usearch_wasm_linux_arm64_2.8.13.tar.gz : `0be59109c46c80a14323ab5131a75f4fe71c7f9537bdf2ff23370daa2808fab4`
* usearch_wasm_linux_x86_64_2.8.13.tar.gz : `175f6234bd52e76501336de8401d0d6dabcf0274b7ca32e7998cbea65159c54c`
* usearch_wasm_macos_arm64_2.8.13.zip : `41e4d2afa660f96f2d4c5316c12d5a1bd2e067997f9727364f7531909197ad8f`
* usearch_wasm_macos_x86_64_2.8.13.zip : `c42edb8e87b0b1ced4bdc6684f7b081004ff8f295e56faedd41b4e2113809ad3`
* usearch_wasm_windows_x64_2.8.13.tar.gz : `bffd155710a385980b0b551173594ea3255a3bb3f358d89f083c2bcb71941e0d`
* usearch_wasm_windows_x86_2.8.13.tar.gz : `f8fcd52e46bfc694c175e725f904745077d736741f11cc69df259a8d3a794e12`
* usearch_windows_x64_2.8.13.tar : `c9df87ae09fc2819a9ed905050015faf06f4af2d4aacea8efec5ec07dc3e23a1`
* usearch_windows_x86_2.8.13.tar : `c9df87ae09fc2819a9ed905050015faf06f4af2d4aacea8efec5ec07dc3e23a1`

[2.8.12](https://github.com/unum-cloud/usearch/compare/v2.8.11...v2.8.12) (2023-11-13)


Docs

* Typo in Java README (307) ([494b9f0](https://github.com/unum-cloud/usearch/commit/494b9f046cab632e573c50ad5e4fe2dd51579364)), closes [#307](https://github.com/unum-cloud/usearch/issues/307)
* Update Java's step-by-step build instructions (306) ([63d8027](https://github.com/unum-cloud/usearch/commit/63d8027ffce9db7499abcf9d04514a944f81952d)), closes [#306](https://github.com/unum-cloud/usearch/issues/306)

Fix

* Dumping to uninitialized memory in Java ([9eaa10d](https://github.com/unum-cloud/usearch/commit/9eaa10d9790d0f809357210d2deffe57a4ac1d2e))
* Key type mismatch in Java ([25dc6fc](https://github.com/unum-cloud/usearch/commit/25dc6fcdeb5f8c8561f1f8c3e51f13b6be8256c4))
* Normalize bit-vector length for SimSIMD (311) ([8a77b3e](https://github.com/unum-cloud/usearch/commit/8a77b3e28d96add165e9436483b8dab7f54b917b)), closes [#311](https://github.com/unum-cloud/usearch/issues/311)
* OOM error in Java ([2589979](https://github.com/unum-cloud/usearch/commit/25899790a6cd3e7ca540d9f6f8764c659081aaa3)), closes [#308](https://github.com/unum-cloud/usearch/issues/308)

Improve

* TypeScript annotations (309) ([6a5e009](https://github.com/unum-cloud/usearch/commit/6a5e00946441db20fc17f5850ecb0243add80c57)), closes [#309](https://github.com/unum-cloud/usearch/issues/309)

Make

* Trace used compiler ([7d17eab](https://github.com/unum-cloud/usearch/commit/7d17eabdb97754a6d4cb56ce7739ae9ab77f363a))




Hashes
* docs.tar.gz : `2d939471b122e76d02f0ce06a200cd70f0f3e549992264dbcdb7bd289b456af2`
* usearch-v2.8.12.tar.gz : `e8b272f8d9b544123ddc574fcd925c60b67d8ce94568f30fe80a72ec86139257`
* usearch-v2.8.12.zip : `0c8d8ef7144913f2fcd6190135880137e16cca46d822acf3b8e985c7c7445e52`
* usearch_linux_amd_2.8.12.deb : `646a26438035b4d698eca8c259a4454bb15d1270dd554fb47d68ffa791f51135`
* usearch_linux_arm_2.8.12.deb : `68fd077160ec6669716a296481d0a2b74a7e9ed17d76a41663a8212ecd361f5a`
* usearch_macOS_arm64_2.8.12.zip : `f2e81ef8f3913028aa6237a8773d93da4875a86c5bd0111e825e7201a900980e`
* usearch_macOS_x86_64_2.8.12.zip : `5389c72e3c91012ba70389885e7260b9deebfe94cf284a8c5c36f3eedc0991d1`
* usearch_wasm_linux_arm64_2.8.12.tar.gz : `323631cae148ab13ade9aab3158d23249979fb3af238aad5e04c1e6a28cf16aa`
* usearch_wasm_linux_x86_64_2.8.12.tar.gz : `0d387642e06c49aa7c3e47f15a0b8d4ceda161a25b50e5f301b3270f55f478f7`
* usearch_wasm_macos_arm64_2.8.12.zip : `b10c7fa633f491069737f64d3c39fbdb11b8565a25932d6e3d9acea20512c980`
* usearch_wasm_macos_x86_64_2.8.12.zip : `ecc6b08c1dd42d0024274f966e91cfaf41b9db35214f7ec0c54c13586b3d9b77`
* usearch_wasm_windows_x64_2.8.12.tar.gz : `9f0a85be9190d2e026575041f8fce4992688e110dd53ee037774f61dfcd0e87f`
* usearch_wasm_windows_x86_2.8.12.tar.gz : `697a723af5e9640c01f30f28fc007ec0c837b62dbbef6bdf95edf3879fc2e165`
* usearch_windows_x64_2.8.12.tar : `e3239a9f2d0a8353ee539d480d9cbed28b911fa590164cdbede4ce362db9ff2f`
* usearch_windows_x86_2.8.12.tar : `dc830356ca3c7cf50ee9c0633f37f68fe298dff95ea10fb6512112803815692a`

[2.8.11](https://github.com/unum-cloud/usearch/compare/v2.8.10...v2.8.11) (2023-11-11)


Fix

* loading of native libs from Jar with NativeUtils.java (305) ([40bb2fb](https://github.com/unum-cloud/usearch/commit/40bb2fbbbbbd2acdce08928d17912c58cdb29612)), closes [#305](https://github.com/unum-cloud/usearch/issues/305)