Datasets:
NYTK
/

Languages:
Hungarian
License:
ligetinagy commited on
Commit
36047c9
1 Parent(s): 54ae770

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +10 -13
README.md CHANGED
@@ -52,7 +52,7 @@ The BCP-47 code for Hungarian, the only represented language in this dataset, is
52
 
53
  ### Data Instances
54
 
55
- For each instance, there is a unique id ('Sent_id', a sentence ('Sent') and a label (0 for wrong, 1 for good sentences).
56
 
57
  An example:
58
 
@@ -63,22 +63,19 @@ An example:
63
  ```
64
 
65
  ### Data Fields
66
- - id: unique id of the instances, an integer between 1 and 1000;
67
- - question: "cause" or "effect". It suggests what kind of causal relation are we looking for: in the case of "cause" we search for the more plausible alternative that may be a cause of the premise. In the case of "effect" we are looking for a plausible result of the premise;
68
- - premise: the premise, a sentence;
69
- - choice1: the first alternative, a sentence;
70
- - choice2: the second alternative, a sentence;
71
- - label: the number of the more plausible alternative (1 or 2).
72
 
73
  ### Data Splits
74
 
75
- HuCoPA has 3 splits: *train*, *validation* and *test*.
76
 
77
- | Dataset split | Number of instances in the split |
78
- |---------------|----------------------------------|
79
- | train | 400 |
80
- | validation | 100 |
81
- | test | 500 |
82
 
83
  The test data is distributed without the labels. To evaluate your model, please [contact us](mailto:ligeti-nagy.noemi@nytud.hu), or check [HuLU's website](hulu.nlp.nytud.hu) for an automatic evaluation (this feature is under construction at the moment).
84
 
 
52
 
53
  ### Data Instances
54
 
55
+ For each instance, there is aN id, a sentence and a label.
56
 
57
  An example:
58
 
 
63
  ```
64
 
65
  ### Data Fields
66
+ - Sent_id: unique id of the instances, an integer between 1 and 1000;
67
+ - Sent: a Hungarian sentence;
68
+ - label: '0' for wrong, '1' for good sentences.
 
 
 
69
 
70
  ### Data Splits
71
 
72
+ HuCOLA has 3 splits: *train*, *validation* and *test*.
73
 
74
+ | Dataset split | Number of sentences in the split | Proportion of the split
75
+ |---------------|----------------------------------| ---------|
76
+ | train | 7214 | 80%|
77
+ | validation | 900 |10%|
78
+ | test | 900 |10%|
79
 
80
  The test data is distributed without the labels. To evaluate your model, please [contact us](mailto:ligeti-nagy.noemi@nytud.hu), or check [HuLU's website](hulu.nlp.nytud.hu) for an automatic evaluation (this feature is under construction at the moment).
81