metadata
tags:
- generated_from_keras_callback
model-index:
- name: GeoBERT
results: []
GeoBERT_Analyzer
GeoBERT_Analyzer is a Text Classification model that was fine-tuned from GeoBERT on the Geoscientific Corpus dataset. The model was trained on the Labeled Geoscientific & Non-Geosceintific Corpus dataset (21416 x 2 sentences).
Intended uses
The train aims to make the Language Model have the ability to distinguish between Geoscience and Non – Geoscience (General) corpus
Training hyperparameters
The following hyperparameters were used during training:
- optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 14000, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}
- training_precision: mixed_float16
Framework versions
- Transformers 4.22.1
- TensorFlow 2.10.0
- Datasets 2.4.0
- Tokenizers 0.12.1
Model performances (metric: seqeval)
| entity | precision | recall | f1 |
|---|---|---|---|
| General | 0.9976 | 0.9980 | 0.9978 |
| Geoscience | 0.9980 | 0.9984 | 0.9982 |
How to use GeoBERT with HuggingFace
Load GeoBERT and its sub-word tokenizer :
from transformers import AutoTokenizer, AutoModelForSequenceClassification
tokenizer = AutoTokenizer.from_pretrained("botryan96/GeoBERT_analyzer")
model = AutoModelForTokenClassification.from_pretrained("botryan96/GeoBERT_analyzer")
#Define the pipeline
from transformers import pipeline
anlyze_machine=pipeline('text-classification',model = model_checkpoint2)
#Define the sentences
sentences = ['the average iron and sulfate concentrations were calculated to be 19 . 6 5 . 2 and 426 182 mg / l , respectively .',
'She first gained media attention as a friend and stylist of Paris Hilton']
#Deploy the machine
anlyze_machine(sentences)