Text Classification
Transformers
Safetensors
bert
Merge
mergekit
lazymergekit
xshubhamx/legal-bert-base-uncased
xshubhamx/InLegalBERT
text-embeddings-inference
Instructions to use xshubhamx/legal-InLegal-merge-linear with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use xshubhamx/legal-InLegal-merge-linear with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="xshubhamx/legal-InLegal-merge-linear")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("xshubhamx/legal-InLegal-merge-linear") model = AutoModelForSequenceClassification.from_pretrained("xshubhamx/legal-InLegal-merge-linear", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Metrics
- loss: 0.9426
- accuracy: 0.8273
- precision: 0.8402
- recall: 0.8273
- precision_macro: 0.8502
- recall_macro: 0.7573
- macro_fpr: 0.0151
- weighted_fpr: 0.0147
- weighted_specificity: 0.9778
- macro_specificity: 0.9870
- weighted_sensitivity: 0.8273
- macro_sensitivity: 0.7573
- f1_micro: 0.8273
- f1_macro: 0.7739
- f1_weighted: 0.8274
- runtime: 29.7266
- samples_per_second: 43.4290
- steps_per_second: 5.4500
legal-InLegal-merge-linear
legal-InLegal-merge-linear is a merge of the following models using mergekit:
🧩 Configuration
models:
- model: xshubhamx/legal-bert-base-uncased
parameters:
weight: 1
- model: xshubhamx/InLegalBERT
parameters:
weight: 1
merge_method: linear
dtype: bfloat16
- Downloads last month
- 3