Notebooks
H
Hugging Face Sagemaker Notebook

Sagemaker Notebook

hf-notebookssagemaker14_train_and_push_to_hub
alph-notebooks/hf-notebooks / sagemaker-notebook.ipynb

HuggingFace Hub meets Amazon SageMaker

Fine-tune a Multi-Class Classification with Trainer and emotion dataset and push it to the Hugging Face Hub

Introduction

Welcome to our end-to-end multi-class Text-Classification example. In this demo, we will use the Hugging Faces transformers and datasets library together with a custom Amazon sagemaker-sdk extension to fine-tune a pre-trained transformer for multi-class text classification. In particular, the pre-trained model will be fine-tuned using the emotion dataset. To get started, we need to set up the environment with a few prerequisite steps, for permissions, configurations, and so on.

emotion-widget.png

NOTE: You can run this demo in Sagemaker Studio, your local machine or Sagemaker Notebook Instances

Development Environment and Permissions

Installation

Note: we only install the required libraries from Hugging Face and AWS. You also need PyTorch or Tensorflow, if you haven´t it installed

[ ]
[ ]

Permissions

If you are going to use Sagemaker in a local environment. You need access to an IAM Role with the required permissions for Sagemaker. You can find here more about it.

[ ]

Preprocessing

We are using the datasets library to download and preprocess the emotion dataset. After preprocessing, the dataset will be uploaded to our sagemaker_session_bucket to be used within our training job. The emotion dataset consists of 16000 training examples, 2000 validation examples, and 2000 testing examples.

Tokenization

[3]
[4]
Downloading: 3.62kB [00:00, 629kB/s]                    
Downloading: 3.28kB [00:00, 998kB/s]                    
Using custom data configuration default
Reusing dataset emotion (/Users/philipp/.cache/huggingface/datasets/emotion/default/0.0.0/348f63ca8e27b3713b6c04d723efe6d824a56fb3d1449794716c0f0296072705)
100%|██████████| 2/2 [00:00<00:00, 99.32it/s]
100%|██████████| 16/16 [00:02<00:00,  7.47ba/s]
100%|██████████| 2/2 [00:00<00:00,  6.47ba/s]

Uploading data to sagemaker_session_bucket

After we processed the datasets we are going to use the new FileSystem integration to upload our dataset to S3.

[5]

Creating an Estimator and start a training job

List of supported models: https://huggingface.co/models?library=pytorch,transformers&sort=downloads

setting up push_to_hub for our model.

The train.py scripts implements the push_to_hub using the Trainer and TrainingArguments. To push our model to the Hub we need to define the push_to_hub. hyperparameter and set it to True and provide out Hugging Face Token. Additionally, we can configure the repository name and saving strategy using the hub_model_id, hub_strategy.

You can find documentation to those parameters here.

We are going to provide our HF Token securely with out exposing it to the public using notebook_login from the huggingface_hub SDK. But be careful your token will still be visible insight the logs of the training job. If you run huggingface_estimator.fit(...,wait=True) you will see the token in the logs. A better way of providing your HF_TOKEN to your training jobs would be using AWS Secret Manager

You can also directly find your token at https://hf.co/settings/token.

[17]
VBox(children=(HTML(value="<center>\n<img src=https://huggingface.co/front/assets/huggingface_logo-noborder.sv…
[2]
[10]
[11]
[ ]

Accessing the model on hf.co/models

we can access the model on hf.co/models using the hub_model_id and our username.

[6]
https://huggingface.co/philschmid/sagemaker-distilbert-emotion

Deploying the model from Hugging Face to a SageMaker Endpoint

To deploy our model to Amazon SageMaker we can create a HuggingFaceModel and provide the Hub configuration (HF_MODEL_ID & HF_TASK) to deploy it. Alternatively, we can use the hugginface_estimator to deploy our model from S3 with huggingface_estimator.deploy().

[ ]

Then, we use the returned predictor object to call the endpoint.

[ ]

Finally, we delete the inference endpoint.

[ ]