erinmikail
commited on
Commit
•
05d3de9
1
Parent(s):
54a843b
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
tags:
|
3 |
+
- Tutorial
|
4 |
+
size_categories:
|
5 |
+
- n<1K
|
6 |
+
---
|
7 |
+
|
8 |
+
## Zero to One: Label Studio Tutorial Dataset
|
9 |
+
|
10 |
+
This dataset is used in the [Label Studio Zero to One Tutorial](https://hubs.ly/Q01CNlyy0). This dataset was originally provided by [Andrew Maas](https://ai.stanford.edu/~amaas/)([ref](https://ai.stanford.edu/~amaas/papers/wvSent_acl2011.bib)). This is an open and well-known dataset. The original dataset did have over 100,000 reviews.
|
11 |
+
|
12 |
+
### Parsing down 100,000 reviews to 100 reviews
|
13 |
+
|
14 |
+
To parse this dataset down to 100 reviews, (Chris Hoge)[https://huggingface.co/hogepodge] and myself((Erin Mikail Staples)[https://huggingface.co/erinmikail]) took the following steps.
|
15 |
+
|
16 |
+
It started by (writing a script)[https://s3.amazonaws.com/labelstud.io/datasets/IMDB_collect.py] that walked the directory structure to capture the data and metadata as rows of data. The data was written in randomized batches with rows corresponding to:
|
17 |
+
|
18 |
+
- 0 - 25,000: Labeled training data, with positive and negative sentiment mixed.
|
19 |
+
- 25,001 - 75000: Unlabeled training data.
|
20 |
+
- 75001 - 100,000: Labeled testing data, with positive and negative sentiment mixed.
|
21 |
+
|
22 |
+
These batches were also written out as separate files for convenience. Finally, the first 100 rows of each batch were written out as separate files to support faster loading for a streamlined learning experience.
|
23 |
+
|
24 |
+
Our thanks to Andrew Maas for having provided this free data set from their research.
|
25 |
+
|
26 |
+
## Did you try your hand at this tutorial?
|
27 |
+
|
28 |
+
We'd love to hear you share your results and how it worked out for you!
|
29 |
+
|
30 |
+
Did you build something else with the data?
|
31 |
+
|
32 |
+
Let us know! Join us in the (Label Studio Slack Community)[https://hubs.ly/Q01CNprb0] or drop us an (email)[mailto:community@labelstud.io]
|
33 |
+
|
34 |
+
## Enjoy what we're working on?
|
35 |
+
|
36 |
+
Drop us a star on (GitHub!)[https://hubs.ly/Q01CNp4W0]
|
37 |
+
|