File size: 715 Bytes
b179af1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
---

license: mit
tags:
- audio-video
- curriculum
- mixup-cutmix
- moderate
- nlp-sentiment
- npy-sharded
- progressive
- self-training
- stratified-90-10
---


# load_data.py



## Dataset Summary



A **nlp sentiment** dataset with **audio video** modality, stored in **npy sharded** format.



## Preprocessing & Augmentation



- **Preprocessing**: progressive

- **Augmentation**: mixup cutmix



## Splits & Sampling



- **Split strategy**: stratified 90 10

- **Sampling**: curriculum



## Quality & Labeling



- **Quality filtering**: moderate

- **Labeling**: self training



## Files



- `load_data.py` — main artifact of this repository

## License

See the license field above.