Meena commited on
Commit
ee9e171
1 Parent(s): 2b8d9a0

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +11 -3
README.md CHANGED
@@ -1,8 +1,9 @@
1
- ## TABLE QUESTION ANSWEING
2
 
3
- ## Model
4
  TAPAS, the model learns an inner representation of the English language used in tables and associated texts, which can then be used to extract features useful for downstream tasks such as answering questions about a table, or determining whether a sentence is entailed or refuted by the contents of a table.
5
 
 
6
  - It is a BERT-based model specifically designed (and pre-trained) for answering questions about tabular data
7
  - TAPAS uses relative position embeddings and has 7 token types that encode tabular structure.
8
  - It is pre-trained on the masked language modeling (MLM) objective on a large dataset comprising millions of tables from English Wikipedia and corresponding texts.
@@ -11,4 +12,11 @@ The model has been fine-tuned on several datasets
11
 
12
  1. SQA (Sequential Question Answering by Microsoft)
13
  2. WTQ (Wiki Table Questions by Stanford University)
14
- 3. WikiSQL (by Salesforce).
 
 
 
 
 
 
 
 
1
+ ## TABLE QUESTION ANSWERING
2
 
3
+ ## TAPAS model
4
  TAPAS, the model learns an inner representation of the English language used in tables and associated texts, which can then be used to extract features useful for downstream tasks such as answering questions about a table, or determining whether a sentence is entailed or refuted by the contents of a table.
5
 
6
+ ## Model description
7
  - It is a BERT-based model specifically designed (and pre-trained) for answering questions about tabular data
8
  - TAPAS uses relative position embeddings and has 7 token types that encode tabular structure.
9
  - It is pre-trained on the masked language modeling (MLM) objective on a large dataset comprising millions of tables from English Wikipedia and corresponding texts.
 
12
 
13
  1. SQA (Sequential Question Answering by Microsoft)
14
  2. WTQ (Wiki Table Questions by Stanford University)
15
+ 3. WikiSQL (by Salesforce).
16
+
17
+ ## Uses and limitations
18
+
19
+ ## Usage
20
+
21
+
22
+ ## Evaluation Results