File size: 1,754 Bytes
957fbda
 
df0b87a
 
 
 
 
 
 
 
e470798
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
license: odbl
task_categories:
- tabular-regression
language:
- en
tags:
- climate
size_categories:
- 100K<n<1M
---

## Cluj Napoca Weather Dataset

A weather dataset of Cluj Napoca taken from the OpenWheather History API. It was scraped from the Open Weather
Map using their Weather API. The data collected was from January 1st 2008 until May 2023
at an hourly rate. The columns are presented in the image below:

![image/png](https://cdn-uploads.huggingface.co/production/uploads/6376481a48baeb86b0c3038b/DnLOpVnoBFJANEb7B44ny.png)

The data is licensed per OpenWeather under the [odbl license](https://opendatacommons.org/licenses/odbl/).  

### Long Time Series Predictions

Clone the [Time-Series-Library](https://github.com/thuml/Time-Series-Library) repository.
```
git clone https://github.com/thuml/Time-Series-Library.git
```

For training the model create a folder called dataset. After that you can either modify the .sh files (for me it didn't work), or run the predict.ipynb file found in the Files Folder.
Alternatively, check the HuggingFace implementations for [Informer](https://huggingface.co/docs/transformers/model_doc/informer),
[Autoformer](https://huggingface.co/docs/transformers/model_doc/autoformer), [Vanilla Transformer](https://huggingface.co/docs/transformers/model_doc/time_series_transformer),
[PatchTST](https://huggingface.co/docs/transformers/model_doc/patchtst) and [PatchTSMixer](https://huggingface.co/docs/transformers/model_doc/patchtsmixer)

### Results

Without any hyperparameter tuning the results on this dataset were the following: 

![image/png](https://cdn-uploads.huggingface.co/production/uploads/6376481a48baeb86b0c3038b/MQ3G4h7Y04ttSlXvdXCSw.png)

A description of waht I did can be found [here]()