Loading SBERT_Model.py→src/SBERT_Model.py +0 −1 Changes for src/SBERT_Model.py: 0 added lines, 1 removed line. Original line number Diff line number Diff line Loading @@ -10,7 +10,6 @@ import numpy as np import time from datetime import datetime import os import copy import spacy Loading STSB_Dataset.py→src/STSB_Dataset.py +0 −0 File moved. View file attention_visualization.py→src/attention_visualization.py +4 −4 Changes for src/attention_visualization.py: 4 added lines, 4 removed lines. Original line number Diff line number Diff line Loading @@ -3,12 +3,12 @@ from transformers import BertConfig, BertTokenizer, BertModel import seaborn as sns import matplotlib.pyplot as plt config = BertConfig.from_pretrained('./models/stsb-bert-large/0_BERT') config = BertConfig.from_pretrained('../models/stsb-bert-large/0_BERT') config.output_attentions = True model = BertModel.from_pretrained('./models/stsb-bert-large/0_BERT', config=config).to('cpu') tokenizer = BertTokenizer.from_pretrained('./models/stsb-bert-large/0_BERT') model = BertModel.from_pretrained('../models/stsb-bert-large/0_BERT', config=config).to('cpu') tokenizer = BertTokenizer.from_pretrained('../models/stsb-bert-large/0_BERT') text = 'US drone strike kills 10 in Pakistan' text = '[CLS] US drone strike kills 10 in Pakistan' tok = tokenizer.tokenize(text) pos = 3 Loading paws_evaluation.py→src/paws_evaluation.py +2 −2 Changes for src/paws_evaluation.py: 2 added lines, 2 removed lines. Original line number Diff line number Diff line Loading @@ -5,8 +5,8 @@ test_dataset = load_dataset('paws', 'labeled_final', split='test') test_dataset.set_format(type='pandas') test_dataset = test_dataset[:] zero_model = SBERT_Model("Zero Model", 'models/nli-bert-large/', test_dataset) ft_model = SBERT_Model("Fine-tuned Model", 'models/stsb-bert-large/', test_dataset) zero_model = SBERT_Model("Zero Model", '../models/nli-bert-large/', test_dataset) ft_model = SBERT_Model("Fine-tuned Model", '../models/stsb-bert-large/', test_dataset) paws_ft_model = SBERT_Model("Fine-tuned on PAWS Model", "models/paws-stsb-bert-large", test_dataset) zero_model.print_statistics() Loading paws_training.py→src/paws_training.py +1 −1 Changes for src/paws_training.py: 1 added line, 1 removed line. Original line number Diff line number Diff line Loading @@ -13,7 +13,7 @@ num_epochs = 1 model_save_path = 'output/training_paws_continue_training-'+model_name+'-'+datetime.now().strftime("%Y-%m-%d_%H-%M-%S") # Load a pre-trained sentence transformer model model = SentenceTransformer('models/stsb-bert-large') model = SentenceTransformer('../models/stsb-bert-large') # Convert the dataset to a DataLoader ready for training train_samples = [] Loading Loading
SBERT_Model.py→src/SBERT_Model.py +0 −1 Changes for src/SBERT_Model.py: 0 added lines, 1 removed line. Original line number Diff line number Diff line Loading @@ -10,7 +10,6 @@ import numpy as np import time from datetime import datetime import os import copy import spacy Loading
attention_visualization.py→src/attention_visualization.py +4 −4 Changes for src/attention_visualization.py: 4 added lines, 4 removed lines. Original line number Diff line number Diff line Loading @@ -3,12 +3,12 @@ from transformers import BertConfig, BertTokenizer, BertModel import seaborn as sns import matplotlib.pyplot as plt config = BertConfig.from_pretrained('./models/stsb-bert-large/0_BERT') config = BertConfig.from_pretrained('../models/stsb-bert-large/0_BERT') config.output_attentions = True model = BertModel.from_pretrained('./models/stsb-bert-large/0_BERT', config=config).to('cpu') tokenizer = BertTokenizer.from_pretrained('./models/stsb-bert-large/0_BERT') model = BertModel.from_pretrained('../models/stsb-bert-large/0_BERT', config=config).to('cpu') tokenizer = BertTokenizer.from_pretrained('../models/stsb-bert-large/0_BERT') text = 'US drone strike kills 10 in Pakistan' text = '[CLS] US drone strike kills 10 in Pakistan' tok = tokenizer.tokenize(text) pos = 3 Loading
paws_evaluation.py→src/paws_evaluation.py +2 −2 Changes for src/paws_evaluation.py: 2 added lines, 2 removed lines. Original line number Diff line number Diff line Loading @@ -5,8 +5,8 @@ test_dataset = load_dataset('paws', 'labeled_final', split='test') test_dataset.set_format(type='pandas') test_dataset = test_dataset[:] zero_model = SBERT_Model("Zero Model", 'models/nli-bert-large/', test_dataset) ft_model = SBERT_Model("Fine-tuned Model", 'models/stsb-bert-large/', test_dataset) zero_model = SBERT_Model("Zero Model", '../models/nli-bert-large/', test_dataset) ft_model = SBERT_Model("Fine-tuned Model", '../models/stsb-bert-large/', test_dataset) paws_ft_model = SBERT_Model("Fine-tuned on PAWS Model", "models/paws-stsb-bert-large", test_dataset) zero_model.print_statistics() Loading
paws_training.py→src/paws_training.py +1 −1 Changes for src/paws_training.py: 1 added line, 1 removed line. Original line number Diff line number Diff line Loading @@ -13,7 +13,7 @@ num_epochs = 1 model_save_path = 'output/training_paws_continue_training-'+model_name+'-'+datetime.now().strftime("%Y-%m-%d_%H-%M-%S") # Load a pre-trained sentence transformer model model = SentenceTransformer('models/stsb-bert-large') model = SentenceTransformer('../models/stsb-bert-large') # Convert the dataset to a DataLoader ready for training train_samples = [] Loading