shuttie commited on
Commit
f695438
1 Parent(s): 4d7fbd8

initial commit

Browse files
Files changed (3) hide show
  1. README.md +30 -0
  2. test.csv +0 -0
  3. train.csv +0 -0
README.md CHANGED
@@ -1,3 +1,33 @@
1
  ---
2
  license: apache-2.0
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: apache-2.0
3
  ---
4
+
5
+ # Dad Jokes dataset
6
+
7
+ This dataset is generated from the [Kaggle Reddit Dad Jokes](https://www.kaggle.com/datasets/oktayozturk010/reddit-dad-jokes) by [Oktay Ozturk](https://www.kaggle.com/oktayozturk010), with the following modifications:
8
+
9
+ * Only jokes with 5+ votes were sampled. Less upvoted jokes are too cringe.
10
+ * With a set of heuristics, each joke was split into two parts: base and the punchline.
11
+
12
+ ## Format
13
+
14
+ The dataset is formatted as a CSV, and is split into train/test parts:
15
+
16
+ * train: 52000 samples
17
+ * test: 1400 samples
18
+
19
+ ```csv
20
+ "question","response"
21
+ "I asked my priest how he gets holy water","He said it’s just regular water, he just boils the hell out of it"
22
+ "Life Hack: If you play My Chemical Romance loud enough in your yard","your grass will cut itself"
23
+ "Why did Mr. Potato Head get pulled over","He was baked"
24
+ "How did the Mexican John Wick taste his Burrito","He took Juan Lick"
25
+ ```
26
+
27
+ ## Usage
28
+
29
+ With a base/punchline split, this dataset can be used for a joke prediction task with any LLM.
30
+
31
+ ## License
32
+
33
+ Apache 2.0.
test.csv ADDED
The diff for this file is too large to render. See raw diff
 
train.csv ADDED
The diff for this file is too large to render. See raw diff