julien-c commited on
Commit
dafc05b
1 Parent(s): f32a3b1

Migrate model card from transformers-repo

Browse files

Read announcement at https://discuss.huggingface.co/t/announcement-all-model-cards-will-be-migrated-to-hf-co-model-repos/2755
Original file history: https://github.com/huggingface/transformers/commits/master/model_cards/a-ware/roberta-large-squad-classification/README.md

Files changed (1) hide show
  1. README.md +48 -0
README.md ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ datasets:
3
+ - squad_v2
4
+ ---
5
+
6
+ # Roberta-LARGE finetuned on SQuADv2
7
+
8
+ This is roberta-large model finetuned on SQuADv2 dataset for question answering answerability classification
9
+
10
+ ## Model details
11
+ This model is simply an Sequenceclassification model with two inputs (context and question) in a list.
12
+ The result is either [1] for answerable or [0] if it is not answerable.
13
+ It was trained over 4 epochs on squadv2 dataset and can be used to filter out which context is good to give into the QA model to avoid bad answers.
14
+
15
+ ## Model training
16
+ This model was trained with following parameters using simpletransformers wrapper:
17
+ ```
18
+ train_args = {
19
+ 'learning_rate': 1e-5,
20
+ 'max_seq_length': 512,
21
+ 'overwrite_output_dir': True,
22
+ 'reprocess_input_data': False,
23
+ 'train_batch_size': 4,
24
+ 'num_train_epochs': 4,
25
+ 'gradient_accumulation_steps': 2,
26
+ 'no_cache': True,
27
+ 'use_cached_eval_features': False,
28
+ 'save_model_every_epoch': False,
29
+ 'output_dir': "bart-squadv2",
30
+ 'eval_batch_size': 8,
31
+ 'fp16_opt_level': 'O2',
32
+ }
33
+ ```
34
+
35
+ ## Results
36
+ ```{"accuracy": 90.48%}```
37
+ ## Model in Action 🚀
38
+ ```python3
39
+ from simpletransformers.classification import ClassificationModel
40
+
41
+ model = ClassificationModel('roberta', 'a-ware/roberta-large-squadv2', num_labels=2, args=train_args)
42
+
43
+ predictions, raw_outputs = model.predict([["my dog is an year old. he loves to go into the rain", "how old is my dog ?"]])
44
+ print(predictions)
45
+ ==> [1]
46
+ ```
47
+
48
+ > Created with ❤️ by A-ware UG [![Github icon](https://cdn0.iconfinder.com/data/icons/octicons/1024/mark-github-32.png)](https://github.com/aware-ai)