Update README.md
Browse files
README.md
CHANGED
@@ -2,7 +2,9 @@
|
|
2 |
license: gpl-3.0
|
3 |
---
|
4 |
|
5 |
-
|
|
|
|
|
6 |
|
7 |
This repo contains the trained model of Convolutional autoencoder for image denoising on MNIST Dataset mixed with random noise.
|
8 |
|
@@ -10,6 +12,26 @@ Spaces Link:- https://huggingface.co/spaces/keras-io/conv_autoencoder
|
|
10 |
|
11 |
Keras Example Link:- https://keras.io/examples/vision/autoencoder/
|
12 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
13 |
<details>
|
14 |
<summary>View Model Plot</summary>
|
15 |
|
|
|
2 |
license: gpl-3.0
|
3 |
---
|
4 |
|
5 |
+
## Model Description
|
6 |
+
|
7 |
+
### Keras Implementation of Convolutional autoencoder for image denoising
|
8 |
|
9 |
This repo contains the trained model of Convolutional autoencoder for image denoising on MNIST Dataset mixed with random noise.
|
10 |
|
|
|
12 |
|
13 |
Keras Example Link:- https://keras.io/examples/vision/autoencoder/
|
14 |
|
15 |
+
## Intended uses & limitations
|
16 |
+
- The trained model can be used to remove noise from any grayscale image.
|
17 |
+
- Since this model is trained on MNIST Data added with random noise, so this model can be used only for images with shape 28 * 28.
|
18 |
+
|
19 |
+
## Training and evaluation data
|
20 |
+
- Original mnist train & test dataset were loaded from tensorflow datasets.
|
21 |
+
- Then Some noise was added to train & test images.
|
22 |
+
- Noisy images were used as input images and original clean images were used as output images for training.
|
23 |
+
|
24 |
+
## Training procedure
|
25 |
+
### Training hyperparameter
|
26 |
+
The following hyperparameters were used during training:
|
27 |
+
- optimizer: 'adam'
|
28 |
+
- loss: 'binary_crossentropy'
|
29 |
+
- epochs: 100
|
30 |
+
- batch_size: 128
|
31 |
+
- ReLU was used as activation function in all layers except last layer where Sigmoid was used as activation function.
|
32 |
+
|
33 |
+
## Model Plot
|
34 |
+
|
35 |
<details>
|
36 |
<summary>View Model Plot</summary>
|
37 |
|