No project description provided
Project description
DepthML: A Deep Learning Framework
DepthML is a high-performance, Pythonic deep learning framework, which provides a hybrid structure (Keras-like lazy initialization combined with PyTorch-like imperative programming), while delegating all low-level tensor calculus to the DepthTensor backend.
Features
-
Hybrid Object-Oriented API: Incorporates design paradigms present in both Keras and PyTorch.
-
Hardware Agnostic: Easily switches between the CPU (numpy) and the GPU (cupy) via the DepthTensor backend.
Benchmarks
DepthML is capable of training standard architectures.
Task: MNIST Digit Classifcation (60k samples)
Architecture: 3-Layer MLP (784 -> 32 -> 10)
Hardware: NVIDIA GPU
| Metric | Result |
|---|---|
| Final Accuracy | 93.86% |
| Training Time (5 epochs) | ~45 seconds |
| Average Step Time | 7.4 ms |
| Convergence | < 2 epochs |
This framework achieves within 3x the step-latency of optimized C++ frameworks for small-scale MLPs.
Installation
pip install depthml
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file depthml-0.2.0.tar.gz.
File metadata
- Download URL: depthml-0.2.0.tar.gz
- Upload date:
- Size: 13.2 kB
- Tags: Source
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.1.0 CPython/3.13.7
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
d0ffdaedd93f4c31abaaf0c97fe952e4b1662f901b1fe741cf77bd841553a21f
|
|
| MD5 |
206d0845000231672882643daa219147
|
|
| BLAKE2b-256 |
49cb65bdb51cf143ddc40dc4e5357110ce7e745d59007bb840bea5bbc8060977
|
Provenance
The following attestation bundles were made for depthml-0.2.0.tar.gz:
Publisher:
publish.yml on l-h-ha/DepthML
-
Statement:
-
Statement type:
https://in-toto.io/Statement/v1 -
Predicate type:
https://docs.pypi.org/attestations/publish/v1 -
Subject name:
depthml-0.2.0.tar.gz -
Subject digest:
d0ffdaedd93f4c31abaaf0c97fe952e4b1662f901b1fe741cf77bd841553a21f - Sigstore transparency entry: 813622055
- Sigstore integration time:
-
Permalink:
l-h-ha/DepthML@2cea54ed2ec07df9538153d890dbbcb88f75829b -
Branch / Tag:
refs/tags/v0.2.0 - Owner: https://github.com/l-h-ha
-
Access:
public
-
Token Issuer:
https://token.actions.githubusercontent.com -
Runner Environment:
github-hosted -
Publication workflow:
publish.yml@2cea54ed2ec07df9538153d890dbbcb88f75829b -
Trigger Event:
release
-
Statement type:
File details
Details for the file depthml-0.2.0-py3-none-any.whl.
File metadata
- Download URL: depthml-0.2.0-py3-none-any.whl
- Upload date:
- Size: 22.1 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.1.0 CPython/3.13.7
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
43f7fc8aa15c4c0d949f98f6b17527d7347fd3b556c123bc68a226f0de14ecfa
|
|
| MD5 |
349e958e182840325d86f09899b12b6a
|
|
| BLAKE2b-256 |
c72ce72ea75cdca9fb101dc4216fa363a4c6a53e6276c14b7a644a9107e96460
|
Provenance
The following attestation bundles were made for depthml-0.2.0-py3-none-any.whl:
Publisher:
publish.yml on l-h-ha/DepthML
-
Statement:
-
Statement type:
https://in-toto.io/Statement/v1 -
Predicate type:
https://docs.pypi.org/attestations/publish/v1 -
Subject name:
depthml-0.2.0-py3-none-any.whl -
Subject digest:
43f7fc8aa15c4c0d949f98f6b17527d7347fd3b556c123bc68a226f0de14ecfa - Sigstore transparency entry: 813622056
- Sigstore integration time:
-
Permalink:
l-h-ha/DepthML@2cea54ed2ec07df9538153d890dbbcb88f75829b -
Branch / Tag:
refs/tags/v0.2.0 - Owner: https://github.com/l-h-ha
-
Access:
public
-
Token Issuer:
https://token.actions.githubusercontent.com -
Runner Environment:
github-hosted -
Publication workflow:
publish.yml@2cea54ed2ec07df9538153d890dbbcb88f75829b -
Trigger Event:
release
-
Statement type:
