A from-scratch ML framework in C++20 with a PyTorch-like Python API.
Features
- Automatic differentiation -- reverse-mode autograd with gradient checking on every op
- Strided tensor engine -- zero-copy views, NumPy interop
- Hand-written AVX2 SIMD matmul -- ~8x faster than naive on x86
- Full transformer stack -- multi-head attention, LayerNorm, GELU, GPT
- CUDA backend -- optional GPU kernels (
--device cudain Python,-DTIRAMISU_ENABLE_CUDA=ONin CMake) - PyTorch-familiar API --
Tensor,requires_grad,backward(),nn.Linear,optim.Adam
Installation
pip install tiramisu-ml
Requires Python 3.10+. Wheels ship for Linux (x86_64, aarch64), macOS (x86_64, arm64), and Windows (amd64); other platforms build from sdist and need CMake + a C++20 compiler.
Quick Start
import numpy as np
import tiramisu as tr
# Random batch of 32 samples, 10 features, 3 classes
x = tr.from_numpy(np.random.randn(32, 10).astype(np.float32))
targets = tr.from_numpy(np.random.randint(0, 3, size=(32,)).astype(np.float32))
layer1 = tr.nn.Linear(10, 64)
layer2 = tr.nn.Linear(64, 3)
# Forward pass
h = tr.relu(layer1.forward(x))
logits = layer2.forward(h)
loss = tr.nn.cross_entropy_loss(logits, targets)
# Backward pass
loss.backward()
# Optimize
params = layer1.parameters() + layer2.parameters()
optimizer = tr.optim.Adam(params, lr=1e-3)
optimizer.step()
optimizer.zero_grad()
API Reference
Tensor
import numpy as np
import tiramisu as tr
# Creation
tr.Tensor([2, 3]) # zero-filled float32
tr.from_numpy(np.zeros((2, 3), np.float32))
# Arithmetic (with autograd)
tr.add(a, b) tr.add(a, 2.0) # addition
tr.sub(a, b) # subtraction
tr.mul(a, b) tr.mul(a, 2.0) # multiplication
tr.div(a, b) # division
tr.neg(a) # negation
# Activations
tr.relu(a) tr.gelu(a) # ReLU, GELU
tr.softmax(a) # softmax
# Reductions
tr.sum(a) tr.mean(a) # sum / mean over all elements
# Layout
a.reshape([6]) # reshape
tr.transpose(a) # swap last two dims
a.contiguous() # ensure contiguous memory
# Linear algebra
tr.matmul(a, b) # matrix multiplication
# Autograd
x = tr.from_numpy(np.array([2.0], np.float32))
x.requires_grad = True
y = tr.mul(x, x)
y.backward()
x.grad # gradient tensor
# NumPy interop
arr = np.asarray(a) # zero-copy when contiguous
a.numpy() # same, via method
Neural Network Modules
import tiramisu as tr
linear = tr.nn.Linear(in_features=784, out_features=10)
layernorm = tr.nn.LayerNorm(features=128)
gpt = tr.nn.GPT(
vocab_size=65,
d_model=128,
num_heads=4,
num_layers=2,
max_seq_len=256,
)
out = linear.forward(x)
params = gpt.parameters()
cfg = gpt.config()
Functional Operations
import tiramisu as tr
loss = tr.nn.cross_entropy_loss(logits, targets)
sm = tr.softmax(logits)
g = tr.gelu(x)
Optimizers
import tiramisu as tr
optimizer = tr.optim.Adam(parameters, lr=1e-3)
optimizer = tr.optim.AdamW(parameters, lr=3e-4, weight_decay=0.1)
optimizer = tr.optim.SGD(parameters, lr=0.01)
optimizer.step()
optimizer.zero_grad()
tr.optim.clip_grad_norm_(parameters, max_norm=1.0)
scheduler = tr.optim.CosineAnnealingLR(base_lr=3e-4, total_steps=1000)
lr = scheduler.step()
Serialize
import tiramisu as tr
tr.serialize.save_gpt("model.ckpt", model, step=100, epoch=1)
step, epoch = tr.serialize.load_gpt("model.ckpt", model)
Device
import tiramisu as tr
tr.cuda_available() # True if built with CUDA
x = tr.from_numpy(arr, device="cuda") # place tensor on GPU
loss.cpu().numpy() # move back to CPU for NumPy
Token indices are stored as float32 tensors today (the C++ embedding path reads float indices).
Architecture
Python API (tiramisu)
│
└─→ pybind11 (_C)
│
├─→ core/ Tensor, Storage, dtype, device
├─→ ops/ CPU kernels (AVX2); optional CUDA
├─→ autograd/ backward tape, gradcheck
├─→ nn/ Linear, GPT, LayerNorm, loss
└─→ optim/ Adam, AdamW, SGD, schedulers
Examples
Runnable scripts live in examples/python/ and examples/ (C++).
| Script | Description |
|---|---|
examples/python/autograd_demo.py |
Minimal autograd: y = x² + 3x |
examples/python/linear_forward.py |
Forward pass + NumPy interop |
examples/python/train_mnist.py |
2-layer MLP on MNIST |
examples/python/gpt_step.py |
Single GPT training step |
examples/python/train_shakespeare.py |
Char-level GPT on Tiny Shakespeare |
# Python (after pip install)
python examples/python/train_mnist.py --data-dir data --epochs 5
python examples/python/train_shakespeare.py --preset tiny --epochs 3
C++ examples require a local build (see below):
cmake --build build --target mnist && ./build/examples/mnist
cmake --build build --target train_shakespeare
./build/examples/train_shakespeare --preset tiny --epochs 3
Building from source
Only needed for C++ development, CUDA, or contributing. Requires CMake 3.20+ and a C++20 compiler.
cmake -S . -B build -G Ninja -DCMAKE_BUILD_TYPE=Release
cmake --build build --parallel
ctest --test-dir build --output-on-failure
CUDA: add -DTIRAMISU_ENABLE_CUDA=ON. Debug builds enable ASan + UBSan by default (TIRAMISU_ENABLE_SANITIZERS=ON).
Python editable install from the repo root:
pip install -e ".[dev]"
pytest tests/python -v
See python/README.md for CMake Python extension build notes.
License
MIT
1 maintainer
Metadata
1 maintainer
Release files for tiramisu-ml 0.2.2
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| tiramisu_ml-0.2.2.tar.gz | 2.1 MB | Details |
Built distributions (wheels)
Total release size: 7.6 MB
Release files / tiramisu_ml-0.2.2.tar.gz
| Download URL | tiramisu_ml-0.2.2.tar.gz |
|---|---|
| Size | 2.1 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
684ad8554c56422b3d8d41c3c9aaf28956090c0681af933ce441933241d830be
|
|
BLAKE2b-256 checksum How to use checksums |
c28dfe6993f438d39789cd38ddc733e8e8075024cb4115bf6bbbe0860f2c82e4
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 7, 2026.
Transparency logRelease files / tiramisu_ml-0.2.2-cp313-cp313-win_amd64.whl
| Download URL | tiramisu_ml-0.2.2-cp313-cp313-win_amd64.whl |
|---|---|
| Size | 204.3 kB |
| Tags | CPython 3.13 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
dbccf04fb073578da218a05305add8fac3e139a17d2158e9f0bea58bf12a7038
|
|
BLAKE2b-256 checksum How to use checksums |
4d737bf5c0367ef5dc86b9fb8e0764c61cfd5b9a35546eb52a34d74556c88547
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 7, 2026.
Transparency logRelease files / tiramisu_ml-0.2.2-cp313-cp313-manylinux_2_28_x86_64.whl
| Download URL | tiramisu_ml-0.2.2-cp313-cp313-manylinux_2_28_x86_64.whl |
|---|---|
| Size | 379.7 kB |
| Tags | CPython 3.13 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
8e27b99eb525851ef7b62e61df13171e735d9b3473551879e91263cbbd512fa0
|
|
BLAKE2b-256 checksum How to use checksums |
df273e3de2b8263b872b6d92d66aa8682b51380523aca30cd3c176902458e7f8
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 7, 2026.
Transparency logRelease files / tiramisu_ml-0.2.2-cp313-cp313-manylinux_2_28_aarch64.whl
| Download URL | tiramisu_ml-0.2.2-cp313-cp313-manylinux_2_28_aarch64.whl |
|---|---|
| Size | 347.5 kB |
| Tags | CPython 3.13 Linux glibc 2.28+ ARM64 |
|
SHA-256 checksum How to use checksums |
c776df8f551b2ca7535a4ddfa9280e03e0131c4f7e02b43be3134b5659a76240
|
|
BLAKE2b-256 checksum How to use checksums |
4ec654be505a85da2ac607a257c49e7076dddd493d51c22b8b6939e2ef36a3d0
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 7, 2026.
Transparency logRelease files / tiramisu_ml-0.2.2-cp313-cp313-macosx_13_0_x86_64.whl
| Download URL | tiramisu_ml-0.2.2-cp313-cp313-macosx_13_0_x86_64.whl |
|---|---|
| Size | 235.0 kB |
| Tags | CPython 3.13 macOS 13.0+ x86-64 |
|
SHA-256 checksum How to use checksums |
718b095ac91b1b371bda17a615c8a37f0aa076e99ac1e9ac776ba5eedbc08f3a
|
|
BLAKE2b-256 checksum How to use checksums |
6d1e13eda777f726ac0c3915e8f9b348064f7973ea4b37351e06452d931b55b5
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 7, 2026.
Transparency logRelease files / tiramisu_ml-0.2.2-cp313-cp313-macosx_13_0_arm64.whl
| Download URL | tiramisu_ml-0.2.2-cp313-cp313-macosx_13_0_arm64.whl |
|---|---|
| Size | 211.6 kB |
| Tags | CPython 3.13 macOS 13.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
990cef795f6879e67e543fafabacdea9d51d719705c62fecee67ef81dc195b72
|
|
BLAKE2b-256 checksum How to use checksums |
f9129317b5f071436ae4d4939aebcf9061ba226cf3c171033938da2bbe5045ab
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 7, 2026.
Transparency logRelease files / tiramisu_ml-0.2.2-cp312-cp312-win_amd64.whl
| Download URL | tiramisu_ml-0.2.2-cp312-cp312-win_amd64.whl |
|---|---|
| Size | 204.4 kB |
| Tags | CPython 3.12 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
743bde25001ad2250c07f80ca4a9444f1c7ea8d2ccdaf62ed10f31fc3a73315f
|
|
BLAKE2b-256 checksum How to use checksums |
d585b8a8b5e946a488c261cb6652d28054eb7503d09b67b95388f2618d608f65
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 7, 2026.
Transparency logRelease files / tiramisu_ml-0.2.2-cp312-cp312-manylinux_2_28_x86_64.whl
| Download URL | tiramisu_ml-0.2.2-cp312-cp312-manylinux_2_28_x86_64.whl |
|---|---|
| Size | 379.7 kB |
| Tags | CPython 3.12 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
1ad33f8db60916c36667a0970c147182c34afd73fc055838f32a1843111ddedf
|
|
BLAKE2b-256 checksum How to use checksums |
25eab0bbcb003056afeacd8e328574c09429e8062c08f50d1a2effe43015eb52
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 7, 2026.
Transparency logRelease files / tiramisu_ml-0.2.2-cp312-cp312-manylinux_2_28_aarch64.whl
| Download URL | tiramisu_ml-0.2.2-cp312-cp312-manylinux_2_28_aarch64.whl |
|---|---|
| Size | 347.7 kB |
| Tags | CPython 3.12 Linux glibc 2.28+ ARM64 |
|
SHA-256 checksum How to use checksums |
977f5351e511f32e74843824f9d9e1997a8536ef6ac31f3a5553dd754a3e6b63
|
|
BLAKE2b-256 checksum How to use checksums |
868bcddc908e5dbe88106fb86f2705e06742228b511386d2b8642e2d9bb3ca09
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 7, 2026.
Transparency logRelease files / tiramisu_ml-0.2.2-cp312-cp312-macosx_13_0_x86_64.whl
| Download URL | tiramisu_ml-0.2.2-cp312-cp312-macosx_13_0_x86_64.whl |
|---|---|
| Size | 234.9 kB |
| Tags | CPython 3.12 macOS 13.0+ x86-64 |
|
SHA-256 checksum How to use checksums |
b82f00b625561a6b08b05c1684f3fcc5615ef75f95583cac7ab973caa47f7d7a
|
|
BLAKE2b-256 checksum How to use checksums |
de8b6895a1411c78e154c85ae6033f21d7e3b41048222932a31ae1054994e3e9
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 7, 2026.
Transparency logRelease files / tiramisu_ml-0.2.2-cp312-cp312-macosx_13_0_arm64.whl
| Download URL | tiramisu_ml-0.2.2-cp312-cp312-macosx_13_0_arm64.whl |
|---|---|
| Size | 211.5 kB |
| Tags | CPython 3.12 macOS 13.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
1e9635545d24737c3495a0b5178d49318a2ede559d80efe568c1923c44e20890
|
|
BLAKE2b-256 checksum How to use checksums |
7e5868f10fce43fb65d22d6ea6bc0b09399422a044ec4e4172781d3776e7826f
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 7, 2026.
Transparency logRelease files / tiramisu_ml-0.2.2-cp311-cp311-win_amd64.whl
| Download URL | tiramisu_ml-0.2.2-cp311-cp311-win_amd64.whl |
|---|---|
| Size | 203.7 kB |
| Tags | CPython 3.11 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
4d5ad11ada32e943cd5da87dc373674caa376993a72b0772ab19d792e45b2eb5
|
|
BLAKE2b-256 checksum How to use checksums |
4c9fc5ff039c67c775a5be80a5d454bd102cbf8e592d83ca5ac50f667a09e76e
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 7, 2026.
Transparency logRelease files / tiramisu_ml-0.2.2-cp311-cp311-manylinux_2_28_x86_64.whl
| Download URL | tiramisu_ml-0.2.2-cp311-cp311-manylinux_2_28_x86_64.whl |
|---|---|
| Size | 379.6 kB |
| Tags | CPython 3.11 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
91629de7d52ed704959c74ad30fd278be7c0aac61a0730ba4896e5e9cb030846
|
|
BLAKE2b-256 checksum How to use checksums |
4468faba5f05fcf66737faee90a06732a3c14b66b9e0fa117956d52039b4c0ec
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 7, 2026.
Transparency logRelease files / tiramisu_ml-0.2.2-cp311-cp311-manylinux_2_28_aarch64.whl
| Download URL | tiramisu_ml-0.2.2-cp311-cp311-manylinux_2_28_aarch64.whl |
|---|---|
| Size | 348.9 kB |
| Tags | CPython 3.11 Linux glibc 2.28+ ARM64 |
|
SHA-256 checksum How to use checksums |
eafd47733c3ca4befbebbd73e5667273456d04b1d4c22d36d7d866f2b1ee9c1d
|
|
BLAKE2b-256 checksum How to use checksums |
03e11a4eb85bc39939bb25949e26150e3ce6621fec4baa665454496759e8df41
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 7, 2026.
Transparency logRelease files / tiramisu_ml-0.2.2-cp311-cp311-macosx_13_0_x86_64.whl
| Download URL | tiramisu_ml-0.2.2-cp311-cp311-macosx_13_0_x86_64.whl |
|---|---|
| Size | 233.3 kB |
| Tags | CPython 3.11 macOS 13.0+ x86-64 |
|
SHA-256 checksum How to use checksums |
c163ae4928b431e5b7a0ce25d0addba4d489925cb0b11e2cd0b20429d75081c6
|
|
BLAKE2b-256 checksum How to use checksums |
5db248fb726c9a128e5843b61baf2f35f5bf2d1b532ca3f6fae2ccdfab50447d
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 7, 2026.
Transparency logRelease files / tiramisu_ml-0.2.2-cp311-cp311-macosx_13_0_arm64.whl
| Download URL | tiramisu_ml-0.2.2-cp311-cp311-macosx_13_0_arm64.whl |
|---|---|
| Size | 211.4 kB |
| Tags | CPython 3.11 macOS 13.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
ec27edb9f3bdea2c2238dc3c46b18ade98bb7faa8d6b86a8d4e9a1c39b6553ba
|
|
BLAKE2b-256 checksum How to use checksums |
6dba2a513547eb66e97c8d5188d894a060f9791b06bf7423c91260fc545f1023
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 7, 2026.
Transparency logRelease files / tiramisu_ml-0.2.2-cp310-cp310-win_amd64.whl
| Download URL | tiramisu_ml-0.2.2-cp310-cp310-win_amd64.whl |
|---|---|
| Size | 202.7 kB |
| Tags | CPython 3.10 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
b8257cb2261031ae94a9e6da49cdff9629d193cd20969a9afeae5d08eafffb7e
|
|
BLAKE2b-256 checksum How to use checksums |
fab6b9f17c2648cd2fb6f5159dbd7d5b4340e8ab2a28a50a99520a7c250dbd98
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 7, 2026.
Transparency logRelease files / tiramisu_ml-0.2.2-cp310-cp310-manylinux_2_28_x86_64.whl
| Download URL | tiramisu_ml-0.2.2-cp310-cp310-manylinux_2_28_x86_64.whl |
|---|---|
| Size | 378.8 kB |
| Tags | CPython 3.10 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
722d9c5558fbe3448d95949179f401fe7bafe6b8f72c2cf815733e4139fe3102
|
|
BLAKE2b-256 checksum How to use checksums |
7e020cbd1eee9f1f5531caeed6c2b11355c11859e199956bbbd007a9e17dcd34
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 7, 2026.
Transparency logRelease files / tiramisu_ml-0.2.2-cp310-cp310-manylinux_2_28_aarch64.whl
| Download URL | tiramisu_ml-0.2.2-cp310-cp310-manylinux_2_28_aarch64.whl |
|---|---|
| Size | 348.3 kB |
| Tags | CPython 3.10 Linux glibc 2.28+ ARM64 |
|
SHA-256 checksum How to use checksums |
911baa26347a876fe8776c7b50e316938d25c7ce6fcfcca02d81836c00e6e07c
|
|
BLAKE2b-256 checksum How to use checksums |
29270eb73c3b85ba3a035a048fee3fa7f0c744e506d9773399ef6ed81051c21e
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 7, 2026.
Transparency logRelease files / tiramisu_ml-0.2.2-cp310-cp310-macosx_13_0_x86_64.whl
| Download URL | tiramisu_ml-0.2.2-cp310-cp310-macosx_13_0_x86_64.whl |
|---|---|
| Size | 231.6 kB |
| Tags | CPython 3.10 macOS 13.0+ x86-64 |
|
SHA-256 checksum How to use checksums |
6a15295450640dc44dadf54ae8d48ef63dc1195e60b54a52b48a3a0899691be1
|
|
BLAKE2b-256 checksum How to use checksums |
3074b34167f244546ad58dfeee37fec62b0f91d415416f9119f9b65ae0e8362e
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 7, 2026.
Transparency logRelease files / tiramisu_ml-0.2.2-cp310-cp310-macosx_13_0_arm64.whl
| Download URL | tiramisu_ml-0.2.2-cp310-cp310-macosx_13_0_arm64.whl |
|---|---|
| Size | 210.0 kB |
| Tags | CPython 3.10 macOS 13.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
848407a7cb49024b30c72605c7cc4ff0defe078b970f687397cdb81ef4f21324
|
|
BLAKE2b-256 checksum How to use checksums |
f481bf55f42fd59acd991023ac6ef151619223b8137e21166e65a5be9056c2d8
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.12
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 7, 2026.
Transparency log
