-[Fill-Mask](#fill-mask)

-[Vector Database](#vector-database)

-[LLM Fine-tuning](#llm-fine-tuning)

-[Text Classification](#llm-fine-tuning-text-classification)

-[Text Classification - 2 classes](#text-classification-2-classes)

-[Text Classification - 9 classes](#text-classification-9-classes)

2. Obtain a Hugging Face API token to push the fine-tuned model to the Hugging Face Model Hub. Follow the instructions on the[Hugging Face website](https://huggingface.co/settings/tokens) to get your API token.

By following these steps, you can effectively restart trainingfrom a previously trained model, allowing for further refinementand adaptation of the model basedon new requirementsor insights. Adjust parametersas needed for your specific use caseand dataset.

By following these steps, you can leverage PostgresML to seamlessly integrate fine-tuning of Language Models fortext classification directly within your PostgreSQL database. Adjust the dataset, model,and hyperparameters to suit your specific requirements.

In this section, webegin by loading the FinGPT sentiment analysis dataset using the`pgml.load_dataset` function. The dataset is then processedand organized into a shuffled view (pgml.fingpt_sentiment_shuffled_view), ensuring a randomized order of records. This step is crucial for preventing biases introduced by the original data orderingand enhancing the training process.

Once the dataset is loadedand shuffled, we delve into understanding the distribution of sentiment classes within the data. By querying the shuffled view, we obtain valuable insights into thenumber of instances for each sentiment class. This exploration is essential for gaining a comprehensive understanding of the datasetand its inherent class imbalances.

To facilitate the training process, we create distinct views for trainingand testing purposes. The training view (pgml.fingpt_sentiment_train_view) contains80% of the shuffled dataset, enabling the model to learn patternsand associations. Simultaneously, the test view (pgml.fingpt_sentiment_test_view) encompasses the remaining20% of the data, providing a reliable evaluationset to assess the model's performance.

In the final section, we kick off the fine-tuning process using the `pgml.tune` function. The model will be internally configured for sentiment analysis with 9 classes. The training is executed on the 80% of the train view and evaluated on the remaining 20% of the train view. The test view is reserved for evaluating the model's accuracy after training is completed. Please note that the option`hub_private_repo: true` is used to push the model to a private Hugging Face repository.