andi611

11 models • 1 total models in database
Sort by:

distilbert-base-uncased-ner-agnews

license:apache-2.0
24
3

Distilbert Base Uncased Ner Conll2003

This model is a fine-tuned version of distilbert-base-uncased on the conll2003 dataset. It achieves the following results on the evaluation set: - Loss: 0.0664 - Precision: 0.9332 - Recall: 0.9423 - F1: 0.9377 - Accuracy: 0.9852 The following hyperparameters were used during training: - learningrate: 3e-05 - trainbatchsize: 16 - evalbatchsize: 32 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lrschedulertype: linear - numepochs: 5 | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| | 0.2042 | 1.0 | 878 | 0.0636 | 0.9230 | 0.9253 | 0.9241 | 0.9822 | | 0.0428 | 2.0 | 1756 | 0.0577 | 0.9286 | 0.9370 | 0.9328 | 0.9841 | | 0.0199 | 3.0 | 2634 | 0.0606 | 0.9364 | 0.9401 | 0.9383 | 0.9851 | | 0.0121 | 4.0 | 3512 | 0.0641 | 0.9339 | 0.9380 | 0.9360 | 0.9847 | | 0.0079 | 5.0 | 4390 | 0.0664 | 0.9332 | 0.9423 | 0.9377 | 0.9852 | - Transformers 4.8.2 - Pytorch 1.8.1+cu111 - Datasets 1.8.0 - Tokenizers 0.10.3

license:apache-2.0
20
1

bert-large-uncased-whole-word-masking-squad2-with-ner-conll2003-with-neg-with-repeat

license:cc-by-4.0
3
0

bert-large-uncased-whole-word-masking-squad2-with-ner-mit-movie-with-neg-with-repeat

dataset:mit_movie
3
0

distilbert-base-uncased-qa-boolq

license:apache-2.0
3
0

distilbert-base-uncased-ner-mit-restaurant

dataset:mit_restaurant
2
1

bert-large-uncased-whole-word-masking-ner-conll2003

license:apache-2.0
2
0

bert-large-uncased-whole-word-masking-squad2-with-ner-Pwhatisthe-conll2003-with-neg-with-repeat

license:cc-by-4.0
2
0

bert-base-uncased-ner-conll2003

license:apache-2.0
1
0

distilbert-base-uncased-squad2-with-ner-with-neg-with-multi

—
1
0

roberta-base-ner-conll2003

license:mit
0
1