README.md

metadata

base_model:
  - AIMH/mental-longformer-base-4096
language:
  - en
library_name: transformers
license: mit
metrics:
  - name: F1 Score
    type: f1
    value: 0.5524
    verified: false
  - name: Accuracy
    type: accuracy
    value: 0.6064
    verified: false
  - name: Precision
    type: precision
    value: 0.602
    verified: false
  - name: Recall
    type: recall
    value: 0.5385
    verified: false
pipeline_tag: text-classification

About This Model

This model is fine-tuned from the checkpoint of AIMH/mental-longformer-base-4096 using drmuskangarg/CAMS dataset. For more information about the base Longformer model, please visit their model page. We used the same configuration as AIMH/mental-longformer-base-4096 including their tokenizer.

Usage

If you wish to use my model to infer your dataset or maybe pre-train it further, you can import my model in a Python script/notebook.

from transformers import LongformerTokenizer, LongformerForSequenceClassification

tokenizer = LongformerTokenizer.from_pretrained("aimh/mental-longformer-base-4096")
model = LongformerForSequenceClassification.from_pretrained("stackofsugar/mentallongformer-cams-finetuned")  

If you prefer to use the high-level HuggingFace pipeline to make predictions, you can also do it in a Python script/notebook.

from transformers import pipeline

pipe = pipeline("text-classification", model="stackofsugar/mentallongformer-cams-finetuned", tokenizer="aimh/mental-longformer-base-4096")     

More Information

For more information, visit my GitHub Repo.