|Project Name||Stars||Downloads||Repos Using This||Packages Using This||Most Recent Commit||Total Releases||Latest Release||Open Issues||License||Language|
|Tensorflow||174,977||327||77||20 hours ago||46||October 23, 2019||2,150||apache-2.0||C++|
|An Open Source Machine Learning Framework for Everyone|
|Transformers||101,817||64||911||16 hours ago||91||June 21, 2022||719||apache-2.0||Python|
|🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.|
|Pytorch||67,250||146||16 hours ago||23||August 10, 2022||12,004||other||Python|
|Tensors and Dynamic neural networks in Python with strong GPU acceleration|
|Keras||58,439||330||a day ago||68||May 13, 2022||397||apache-2.0||Python|
|Deep Learning for humans|
|Cs Video Courses||55,887||13 days ago||16|
|List of Computer Science courses with video lectures.|
|Faceswap||44,904||12 days ago||24||gpl-3.0||Python|
|Deepfakes Software For All|
|D2l Zh||43,808||1||2 days ago||45||March 25, 2022||33||apache-2.0||Python|
|Tensorflow Examples||42,312||7 months ago||218||other||Jupyter Notebook|
|TensorFlow Tutorial and Examples for Beginners (support TF v1 & v2)|
|100 Days Of Ml Code||40,344||2 months ago||61||mit|
|100 Days of ML Coding|
|Deepfacelab||39,560||5 days ago||531||gpl-3.0||Python|
|DeepFaceLab is the leading software for creating deepfakes.|
The Merlin Models library provides standard models for recommender systems with an aim for high-quality implementations that range from classic machine learning models to highly-advanced deep learning models.
The goal of this library is to make it easy for users in the industry to train and deploy recommender models with the best practices that are already baked into the library. The library simplifies how users in the industry can train standard models against their dataset and put high-performance, GPU-accelerated models into production. The library also enables researchers to build custom models by incorporating standard components of deep learning recommender models and then researchers can benchmark the new models on example offline datasets.
In our initial releases, Merlin Models features a TensorFlow API. The PyTorch API is initiated, but incomplete. We have PyTorch support for transformer-based, session-based recommender systems in the Transformer4Rec library.
RecSys model implementations - The library provides a high-level API for classic and state-of-the-art deep learning architectures for recommender models. These models include both retrieval (e.g. Matrix Factorization, Two tower, YouTube DNN, ..) and ranking (e.g. DLRM, DCN-v2, DeepFM, ...) models.
Building blocks - Within Merlin Models, recommender models are built on reusable building blocks. The design makes it easy to combine the blocks to define new architectures. The library provides model definition blocks (MLP layers, factorization layers, input blocks, negative samplers, loss functions), training models (data loaders from Parquet files), and evaluation (e.g. ranking metrics).
Integration with Merlin platform - Merlin Models is deeply integrated with the other Merlin components. For example, models depend on NVTabular for pre-processing and integrate easily with Merlin Systems for inference. The thoughtfully-designed integration makes it straightforward to build performant end-to-end RecSys pipelines.
Merlin Models DataLoaders - Merlin provides seamless integration with common deep learning frameworks, such as TensorFlow, PyTorch, and HugeCTR. When training deep learning recommender system models, data loading can be a bottleneck. To address the challenge, Merlin has custom, highly-optimized dataloaders to accelerate existing TensorFlow and PyTorch training pipelines. The Merlin dataloaders can lead to a speedup that is nine times faster than the same training pipeline used with the GPU.
With the Merlin dataloaders, you can:
To learn about the core features of Merlin Models, see the Models Overview page.
Merlin Models can be installed with
pip by running the following command:
pip install merlin-models
Installing Merlin Models with
pipdoes not install some additional GPU dependencies, such as the CUDA Toolkit. When you run Merlin Models in one of our Docker containers, the dependencies are already installed.
Merlin Models is included in the Merlin Containers.
Refer to the Merlin Containers documentation page for information about the Merlin container names, URLs to the container images on the NVIDIA GPU Cloud catalog, and key Merlin components.
Merlin Models can be installed from source by running the following commands:
git clone https://github.com/NVIDIA-Merlin/models cd models && pip install -e .
Merlin Models makes it straightforward to define architectures that adapt to different input features.
This adaptability is provided by building on a core feature of the NVTabular library.
When you use NVTabular for feature engineering, NVTabular creates a schema that identifies the input features.
You can see the
Schema object in action by looking at the From ETL to Training RecSys models - NVTabular and Merlin Models integrated example example notebook.
You can easily build popular RecSys architectures like DLRM, as shown in the following code sample. After you define the model, you can train and evaluate it with a typical Keras model.
import merlin.models.tf as mm from merlin.io.dataset import Dataset train = Dataset(PATH_TO_TRAIN_DATA) valid = Dataset(PATH_TO_VALID_DATA) model = mm.DLRMModel( train.schema, # 1 embedding_dim=64, bottom_block=mm.MLPBlock([128, 64]), # 2 top_block=mm.MLPBlock([128, 64, 32]), prediction_tasks=mm.BinaryClassificationTask(train.schema) # 3 ) model.compile(optimizer="adagrad", run_eagerly=False) model.fit(train, validation_data=valid, batch_size=1024) eval_metrics = model.evaluate(valid, batch_size=1024, return_dict=True)
BinaryClassificationTaskin this example. The target binary feature is also inferred from the schema (i.e., tagged as 'TARGET').
You can find more details and information about a low-level API in our overview of the Deep Learning Recommender Model.
View the example notebooks in the documentation to help you become familiar with Merlin Models.
The same notebooks are available in the
examples directory from the Merlin Models GitHub repository.
If you'd like to contribute to the library directly, see the CONTRIBUTING.md file. We're particularly interested in contributions or feature requests for our feature engineering and preprocessing operations. To further advance our Merlin Roadmap, we encourage you to share all the details regarding your recommender system pipeline in this survey.