KushaalGannoju/distilbert-pseudo-model

license: mittags:text-classificationdistilbertcustompseudo-labelingMy DistilBERT Pseudo-Labeled ModelModel DescriptionThis is a fine-tuned DistilBERT model, specifically trained for text classification. It was trained using a pseudo-labeling strategy, where the model was initially trained on a small labeled dataset and then used to generate "pseudo-labels" for a larger unlabeled dataset, which was then used for further training.Intended UsesThis model is intended for classifying text.Training DataThe model was trained on:An initial labeled dataset: a custom dataset of movie reviews.An unlabeled dataset for pseudo-labeling: additional unlabeled movie reviews.How to UseYou can load and use this model with the Hugging Face transformers library:from transformers import AutoTokenizer, AutoModelForSequenceClassification import torch

Load tokenizer and model

tokenizer = AutoTokenizer.from_pretrained("KushaalGannoju/distilbert-pseudo-model") model = AutoModelForSequenceClassification.from_pretrained("KushaalGannoju/distilbert-pseudo-model")

Example inference

text = "This is an example sentence for classification." inputs = tokenizer(text, return_tensors="pt")

with torch.no_grad(): outputs = model(**inputs) logits = outputs.logits predictions = torch.argmax(logits, dim=-1)

print(f"Predicted class ID: {predictions.item()}") Limitations and BiasPerformance may degrade on text outside of the domain it was trained on.As with all models trained on large text corpora, this model may inherit biases present in the training data.Model ArchitectureThis model utilizes the DistilBERT base architecture, fine-tuned for sequence classification.ContactFor any questions or issues, please open an issue on the model's Hugging Face repository page.