firobeid/L3_financial_news_headlines_Doc2Vec

Custom Doc2Vec Embeddings

This model contains custom word embeddings trained using Gensim's Doc2Vec implementation.

Model details

• Trained using Gensim's Word2Vec
• Includes custom n-grams as tokens
• Vector size: 256
• Context window: 5
• Training algorithm: Skip-gram

Doc2Vec Model Components

This folder contains all components of the Doc2Vec model:

• doc2vec_model.model: The complete model that can be loaded with Doc2Vec.load()
• model_parameters.pkl: Dictionary of all model parameters
• word_vocabulary.pkl: Dictionary mapping words to indices
• word_vectors.npy: NumPy array of word vectors
• word_list.pkl: List of words corresponding to word_vectors.npy
• doc_vectors.npy: NumPy array of document vectors (if available)
• doc_tags.pkl: List of document tags corresponding to doc_vectors.npy

To load the model:

from gensim.models.doc2vec import Doc2Vec
model = Doc2Vec.load("doc2vec_model.model")