OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Developer Tools & SDKs

Tamnun Ml

Tamnun is an open-source Python machine- and deep-learning framework with NLP and transfer-learning utilities, including scikit-learn-style BERT classifiers, a PyTorch module estimator and a model-distillation utility.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Tamnun Ml — OpenViST project profile in AI Developer Tools & SDKs.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Tamnun within a Python machine-learning workflow when its documented BERT, PyTorch fitting or distillation interfaces match the task.

Who it is for

  • Python developers building NLP classifiers with BERT.
  • Machine-learning practitioners who want to fit PyTorch modules through an estimator interface.
  • Developers exploring teacher-to-student model distillation.

Documented uses

  • Fine-tune BERT for binary or multiclass classification using a scikit-learn pipeline.
  • Fit a PyTorch module for classification with TorchEstimator.
  • Distill a larger teacher model into a smaller student model.
Before you choose

Requirements & limits

The project is presented as an open-source library on GitHub. Its README gives installation instructions through PyPI or a source checkout and lists Python 3.6 and 3.7.

  • The README lists Python 3.6 and 3.7; check the project instructions against the Python environment you intend to use.
  • Tamnun depends on machine-learning frameworks including PyTorch and Keras, with additional dependencies.
  • The BERT classifier is documented for binary and multiclass classification.
  • The repository provides a fine-tuning script and notebooks, including examples for BERT fine-tuning and distillation.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Fine-tune BERT for binary classification

Fit a BERT classifier with Tamnun's scikit-learn-style pipeline and predict labels for test inputs.

Before you start

  • Tamnun and its dependencies, including PyTorch-Pretrained-BERT and scikit-learn.
  • Training inputs train_X and labels train_y, plus test inputs test_X.

Steps

  1. Import BertClassifier and BertVectorizer from tamnun.bert, and make_pipeline from sklearn.pipeline.

  2. Create and fit a pipeline with make_pipeline(BertVectorizer(), BertClassifier(num_of_classes=2)).fit(train_X, train_y).

  3. Call clf.predict(test_X) to produce predictions.

Expected outcome

The pipeline produces predictions from a BERT classifier fitted for binary classification.

Guide limitations

  • The README states support for binary and multiclass classification; this recipe covers binary classification only.
  • The README points to a notebook for a full example; it does not define the training-data format here.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
18 October 2026
Original website title
hiredscorelabs/tamnun-ml · Official homepage title · retained in its source language
How we review sourcesSuggest a correction