File size: 1,259 Bytes
8b76812
 
1db282d
8b76812
 
1db282d
8b76812
 
1db282d
8b76812
 
ee57c6d
1db282d
ee57c6d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
44
45
46
47
48
49
50
51
52
---
license: mit

datasets:
- PolyAI/banking77

language:
- en

tags:
- autoencoder
---

# VAE trained on Banking 77 Open Intent Classification Dataset
This is a Variational Autoencoder (VAE) trained on the [PolyAI/banking77](https://huggingface.co/datasets/PolyAI/banking77) dataset.

### Architecture
- **input_dim**: 768
- **hidden_dim**: 256
- **latent_dim**: 64

#### Encoder
The encoder maps the input to a latent space distribution.

```python
encoder = nn.Sequential(
            nn.Linear(input_dim, hidden_dim),
            nn.ReLU()
        )

mu = nn.Linear(hidden_dim, latent_dim)
logvar = nn.Linear(hidden_dim, latent_dim)
```

#### Decoder
The decoder reconstructs the input from a sample of the latent space.

```python
decoder = nn.Sequential(
            nn.Linear(latent_dim, hidden_dim),
            nn.ReLU(),
            nn.Linear(hidden_dim, input_dim)
        )
```

#### Metrics
The model was trained and evaluated using the following metrics:
1. Training set: VAE Loss
   * 50% reconstruction loss between original input vs reconstructed output
   * 50% KL divergence between Latent Z vs standard normal distribution
2. Validation set: 100% reconstruction loss -> used to find the best model (with the lowest reconstruction loss)