lorenzoscottb commited on
Commit
2dc7f43
1 Parent(s): bc87400

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +12 -4
README.md CHANGED
@@ -15,21 +15,29 @@ The current dataset contains five sampled splits, used in the supervised experim
15
 
16
  ## Data Structure
17
 
18
- The data are organised around the five `Train/test_split#`, each containing a training and test set of circa 60K and 2K.
19
 
20
  ### Features
21
 
22
  Each entrance has 6 features: `seq, label, Adj_Class, Adj, Nn, Hy`
23
  - `seq`:test sequense
24
- - `label`: ground truth (1:entail, 0:no-entailment)
25
  - `Adj_Class`: the class of the sequence adjectives
26
- - `Adj`: the adjective of the sequence (I:intersective, S:subsective, O:intensional)
27
  - `N`n: the noun
28
  - `Hy`: the noun's hypericum
29
 
 
 
 
 
 
 
 
 
30
  ### Trained Model
31
 
32
- You can find a pre-trend BERT model (on the 2nd out-of-distribution split) [here](https://huggingface.co/lorenzoscottb/bert-base-cased-PLANE-ood-2?text=A+fake+smile+is+a+smile).
33
 
34
  ### Cite
35
 
15
 
16
  ## Data Structure
17
 
18
+ The `dataset` is organised around five `Train/test_split#`, each containing a training and test set of circa 60K and 2K.
19
 
20
  ### Features
21
 
22
  Each entrance has 6 features: `seq, label, Adj_Class, Adj, Nn, Hy`
23
  - `seq`:test sequense
24
+ - `label`: ground truth (1:entialment, 0:no-entailment)
25
  - `Adj_Class`: the class of the sequence adjectives
26
+ - `Adj`: the adjective of the sequence (I: intersective, S: subsective, O: intensional)
27
  - `N`n: the noun
28
  - `Hy`: the noun's hypericum
29
 
30
+ Each sample in `seq` can take one of three forms (or inference types, in paper):
31
+
32
+ - An *Adjective-Noun* is a *Noun* (e.g. A red car is a car)
33
+ - An *Adjective-Noun* is a *Hypernym(Noun)* (e.g. A red car is a vehicle)
34
+ - An *Adjective-Noun* is a *Adjective-Hypernym(Noun)* (e.g. A red car is a red vehicle)
35
+
36
+ Please note that, as specified in the paper, the ground truth is automatically assigned based on the linguistic rule that governs the interaction between each adjective class and inference type – see the paper for more detail.
37
+
38
  ### Trained Model
39
 
40
+ You can find a tuned BERT-base model (tuned and validated using the 2nd split) [here](https://huggingface.co/lorenzoscottb/bert-base-cased-PLANE-ood-2?text=A+fake+smile+is+a+smile).
41
 
42
  ### Cite
43