| --- |
| license: mit |
| --- |
| # Model Card for DistilRoBERTaEmotionClassifier |
|
|
| This model was created to demonstrate several MLOps practices and was for education purposes only. Please see the following [GitHub repo](https://github.com/teg-lad/CA4015-MLOPSPipelineImplementation) covering the material |
|
|
| ## Model Details |
|
|
| ### Model Description |
|
|
| This model was trained on the [Kaggle Emotions](https://www.kaggle.com/datasets/nelgiriyewithana/emotions/data) dataset, which has 6 classes. |
|
|
| + Sadness (0) |
| + Joy (1) |
| + Love (2) |
| + Anger (3) |
| + Fear (4) |
| + Surprise (5) |
|
|
| ### Model Usage |
|
|
| ``` |
| import torch |
| from transformers import AutoModelForSequenceClassification, AutoTokenizer |
| |
| # Load the model and tokenizer |
| model = AutoModelForSequenceClassification.from_pretrained("teglad/DistilRoBERTaEmotionClassifier") |
| tokenizer = AutoTokenizer.from_pretrained("teglad/DistilRoBERTaEmotionClassifier") |
| |
| # Tokenize the input text, returning PyTorch tensors. |
| input_ids = tokenizer("Deep Learning models can be so difficult to understand, how do they even work?", return_tensors="pt") |
| |
| # Pass the input_ids and attention_masks into the model |
| output = model(**input_ids) |
| |
| # Get the position of the largest logit, this is the predicted class |
| prediction = torch.argmax(output.logits, dim=1).tolist() |
| ``` |
|
|
