Instructions to use koen430/Phi2_stockprediction_news with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use koen430/Phi2_stockprediction_news with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("koen430/Phi2_stockprediction_news", dtype="auto") - Notebooks
- Google Colab
- Kaggle
Model Card for Model ID
Predicts the price movements of a stock based on news articles for the day after the article is published. DO NOT USE in its current form.
Model Details
Model Description
- Developed by: K. van der Leij
- Model type: Causal
- Language(s) (NLP): English
- License: The model is licensed under the MIT license.
- Finetuned from model [optional]: Phi-2
Training Details
Training Data
https://huggingface.co/datasets/koen430/relevant_selected_stock_news
Use
Prompt with:
Instruct: Given the article below, please indicate whether the price of the stock with ticker {ticker} will go up or down tomorrow. Only answer with UP or DOWN. \n {Article text}. \n \n Output:
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support