Skip to content

Commit 6cbc9f5

Browse files
tcpdumpfbacketcpdumpfbacke
authored
Merge pull request #4500 from ovh/MB-ai-notebooks-img-seg-tuto
Mb ai notebooks img seg tuto
2 parents 308413c + 2f3d68e commit 6cbc9f5

18 files changed

+1729
-0
lines changed

‎pages/index.md

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -785,6 +785,7 @@
785785
+ [AI Notebooks - Tutorial - Build your spam classifier](platform/ai/notebook_tuto_09_spam_classifier)
786786
+ [AI Notebooks - Tutorial - Create and train a Rasa chatbot](platform/ai/notebook_tuto_10_create_chatbot)
787787
+ [AI Notebooks - Tutorial - Train YOLOv7 for sign language recognition](platform/ai/notebook_tuto_11_yolov7)
788+
+ [AI Notebooks - Tutorial - Brain tumor segmentation using U-Net](platform/ai/notebook_tuto_12_image-segmentation-unet-tumors)
788789
+ [AI Training](public-cloud-ai-and-machine-learning-ai-training)
789790
+ [Guides](public-cloud-ai-and-machine-learning-ai-training-guides)
790791
+ [AI Training - Features, Capabilities and Limitations](platform/ai/training_guide_01_capabilities)

‎pages/platform/ai/notebook_tuto_12_image-segmentation-unet-tumors/guide.de-de.md

Lines changed: 116 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,116 @@
1+
---
2+
title: AI Notebooks - Tutorial - Brain tumor segmentation using U-Net
3+
slug: notebooks/tuto-image-segmentation-unet-tumors
4+
excerpt: Implementing a Convolutional Neural Network for Brain Tumor Segmentation in Medical Imaging
5+
section: AI Notebooks - Tutorials
6+
order: 12
7+
routes:
8+
canonical: 'https://docs.ovh.com/gb/en/publiccloud/ai/notebooks/tuto-image-segmentation-unet-tumors/'
9+
updated: 2023-04-13
10+
---
11+
12+
**Last updated 13th April, 2023.**
13+
14+
## Objective
15+
16+
Over the past few years, the field of **computer vision** has experienced a significant growth. It encompasses a wide range of methods for acquiring, processing, analyzing and understanding digital images.
17+
18+
Among these methods, one is called **image segmentation**.
19+
20+
The purpose of this tutorial is to show you how it is possible to build and train a brain tumor segmentation model with **OVHcloud AI Notebooks**. You will be able to learn the concepts of medical imaging, image segmentation, model evaluation, and much more. We will use a popular convolutional neural network named **U-Net**.
21+
22+
At the end of this tutorial, you will have learnt the principal methods to **segment brain tumors**.
23+
24+
![image](images/mri_orig_pred_segmentation.gif){.thumbnail}
25+
26+
> [!primary]
27+
>
28+
> We will train this model on the *[BraTS2020 dataset](http://braintumorsegmentation.org/)*. We will show you how you can easily download the dataset in the notebook tutorial.
29+
>
30+
31+
## Requirements
32+
33+
- Access to the [OVHcloud Control Panel](https://www.ovh.com/auth/?action=gotomanager&from=https://www.ovh.de/&ovhSubsidiary=de)
34+
- An AI Notebooks project created inside a [Public Cloud project](https://www.ovhcloud.com/de/public-cloud/) in your OVHcloud account
35+
- A user for AI Notebooks
36+
- A [Kaggle](https://www.kaggle.com/) account to download the dataset
37+
38+
## Instructions
39+
40+
You can launch the notebook from the [OVHcloud Control Panel](https://www.ovh.com/auth/?action=gotomanager&from=https://www.ovh.de/&ovhSubsidiary=de) or via the ovhai [CLI](https://docs.ovh.com/de/publiccloud/ai/cli/getting-started-cli/).
41+
42+
### Launching a Jupyter notebook with "Miniconda" via UI (Control Panel)
43+
44+
To launch your notebook from the [OVHcloud Control Panel](https://www.ovh.com/auth/?action=gotomanager&from=https://www.ovh.de/&ovhSubsidiary=de), refer to the following steps.
45+
46+
#### Code editor
47+
48+
Choose the `Jupyterlab` code editor.
49+
50+
#### Framework
51+
52+
In this tutorial, the `tensorflow` framework is used. If you use another environment, there may be some compatibility problems such as missing libraries.
53+
54+
#### Resources
55+
56+
Using GPUs is recommended because medical imaging is a training intensive task.
57+
58+
> [!primary]
59+
>
60+
> Here, using `1 GPU` is sufficient.
61+
>
62+
63+
### Launching a Jupyter notebook with "Miniconda" via CLI
64+
65+
*If you do not use our CLI yet, follow [this guide](https://docs.ovh.com/de/publiccloud/ai/cli/install-client/) to install it.*
66+
67+
If you want to launch your notebook with the OVHcloud AI CLI, choose the `jupyterlab` editor and the `tensorflow` framework.
68+
69+
To access the different versions of `tensorflow` available, run the following command:
70+
71+
```console
72+
ovhai capabilities framework list -o yaml
73+
```
74+
75+
> [!primary]
76+
>
77+
> If you do not specify a version, your notebook starts with the default version of `tensorflow`.
78+
>
79+
80+
You will also need to choose the number of CPUs/GPUs (`<nb-cpus>` or `<nb-gpus>`) to use in your notebook.
81+
82+
> [!primary]
83+
>
84+
> Here we recommend using `1 GPU`.
85+
>
86+
87+
To launch your notebook, run the following command:
88+
89+
```console
90+
ovhai notebook run tensorflow jupyterlab \
91+
--name <notebook-name> \
92+
--framework-version <tensorflow-version> \
93+
--gpu <nb-gpus>
94+
```
95+
96+
You can then reach your notebook’s URL once the notebook is running.
97+
98+
### Accessing the notebooks
99+
100+
Once our [AI examples repository](https://github.com/ovh/ai-training-examples/) has been cloned in your environment, find your notebook by following this path: `ai-training-examples` > `notebooks` > `computer-vision` > `image-segmentation` > `tensorflow` > `brain-tumor-segmentation-unet` > `notebook_image_segmentation_unet.ipynb`.
101+
102+
A preview of this notebook can be found on GitHub [here](https://github.com/ovh/ai-training-examples/blob/main/notebooks/computer-vision/image-segmentation/tensorflow/brain-tumor-segmentation-unet/notebook_image_segmentation_unet.ipynb).
103+
104+
## Go further
105+
106+
There are many other tasks that exist in the computer vision field. Check our other tutorials to learn how to:
107+
108+
- [Use Transfer Learning with ResNet50 for image classification](https://github.com/ovh/ai-training-examples/blob/main/notebooks/computer-vision/image-classification/tensorflow/resnet50/notebook-resnet-transfer-learning-image-classification.ipynb)
109+
110+
- [Train YOLOv7 for American Sign Language recognition](https://github.com/ovh/ai-training-examples/blob/main/notebooks/computer-vision/object-detection/miniconda/yolov7/notebook_object_detection_yolov7_asl.ipynb)
111+
112+
## Feedback
113+
114+
Please send us your questions, feedback and suggestions to improve the service:
115+
116+
- On the OVHcloud [Discord server](https://discord.com/invite/vXVurFfwe9)

‎pages/platform/ai/notebook_tuto_12_image-segmentation-unet-tumors/guide.en-asia.md

Lines changed: 114 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,114 @@
1+
---
2+
title: AI Notebooks - Tutorial - Brain tumor segmentation using U-Net
3+
slug: notebooks/tuto-image-segmentation-unet-tumors
4+
excerpt: Implementing a Convolutional Neural Network for Brain Tumor Segmentation in Medical Imaging
5+
section: AI Notebooks - Tutorials
6+
order: 12
7+
updated: 2023-04-13
8+
---
9+
10+
**Last updated 13th April, 2023.**
11+
12+
## Objective
13+
14+
Over the past few years, the field of **computer vision** has experienced a significant growth. It encompasses a wide range of methods for acquiring, processing, analyzing and understanding digital images.
15+
16+
Among these methods, one is called **image segmentation**.
17+
18+
The purpose of this tutorial is to show you how it is possible to build and train a brain tumor segmentation model with **OVHcloud AI Notebooks**. You will be able to learn the concepts of medical imaging, image segmentation, model evaluation, and much more. We will use a popular convolutional neural network named **U-Net**.
19+
20+
At the end of this tutorial, you will have learnt the principal methods to **segment brain tumors**.
21+
22+
![image](images/mri_orig_pred_segmentation.gif){.thumbnail}
23+
24+
> [!primary]
25+
>
26+
> We will train this model on the *[BraTS2020 dataset](http://braintumorsegmentation.org/)*. We will show you how you can easily download the dataset in the notebook tutorial.
27+
>
28+
29+
## Requirements
30+
31+
- Access to the [OVHcloud Control Panel](https://www.ovh.com/auth/?action=gotomanager&from=https://www.ovh.com/asia/&ovhSubsidiary=asia)
32+
- An AI Notebooks project created inside a [Public Cloud project](https://www.ovhcloud.com/asia/public-cloud/) in your OVHcloud account
33+
- A user for AI Notebooks
34+
- A [Kaggle](https://www.kaggle.com/) account to download the dataset
35+
36+
## Instructions
37+
38+
You can launch the notebook from the [OVHcloud Control Panel](https://www.ovh.com/auth/?action=gotomanager&from=https://www.ovh.com/asia/&ovhSubsidiary=asia) or via the ovhai [CLI](https://docs.ovh.com/asia/en/publiccloud/ai/cli/getting-started-cli/).
39+
40+
### Launching a Jupyter notebook with "Miniconda" via UI (Control Panel)
41+
42+
To launch your notebook from the [OVHcloud Control Panel](https://www.ovh.com/auth/?action=gotomanager&from=https://www.ovh.com/asia/&ovhSubsidiary=asia), refer to the following steps.
43+
44+
#### Code editor
45+
46+
Choose the `Jupyterlab` code editor.
47+
48+
#### Framework
49+
50+
In this tutorial, the `tensorflow` framework is used. If you use another environment, there may be some compatibility problems such as missing libraries.
51+
52+
#### Resources
53+
54+
Using GPUs is recommended because medical imaging is a training intensive task.
55+
56+
> [!primary]
57+
>
58+
> Here, using `1 GPU` is sufficient.
59+
>
60+
61+
### Launching a Jupyter notebook with "Miniconda" via CLI
62+
63+
*If you do not use our CLI yet, follow [this guide](https://docs.ovh.com/asia/en/publiccloud/ai/cli/install-client/) to install it.*
64+
65+
If you want to launch your notebook with the OVHcloud AI CLI, choose the `jupyterlab` editor and the `tensorflow` framework.
66+
67+
To access the different versions of `tensorflow` available, run the following command:
68+
69+
```console
70+
ovhai capabilities framework list -o yaml
71+
```
72+
73+
> [!primary]
74+
>
75+
> If you do not specify a version, your notebook starts with the default version of `tensorflow`.
76+
>
77+
78+
You will also need to choose the number of CPUs/GPUs (`<nb-cpus>` or `<nb-gpus>`) to use in your notebook.
79+
80+
> [!primary]
81+
>
82+
> Here we recommend using `1 GPU`.
83+
>
84+
85+
To launch your notebook, run the following command:
86+
87+
```console
88+
ovhai notebook run tensorflow jupyterlab \
89+
--name <notebook-name> \
90+
--framework-version <tensorflow-version> \
91+
--gpu <nb-gpus>
92+
```
93+
94+
You can then reach your notebook’s URL once the notebook is running.
95+
96+
### Accessing the notebooks
97+
98+
Once our [AI examples repository](https://github.com/ovh/ai-training-examples/) has been cloned in your environment, find your notebook by following this path: `ai-training-examples` > `notebooks` > `computer-vision` > `image-segmentation` > `tensorflow` > `brain-tumor-segmentation-unet` > `notebook_image_segmentation_unet.ipynb`.
99+
100+
A preview of this notebook can be found on GitHub [here](https://github.com/ovh/ai-training-examples/blob/main/notebooks/computer-vision/image-segmentation/tensorflow/brain-tumor-segmentation-unet/notebook_image_segmentation_unet.ipynb).
101+
102+
## Go further
103+
104+
There are many other tasks that exist in the computer vision field. Check our other tutorials to learn how to:
105+
106+
- [Use Transfer Learning with ResNet50 for image classification](https://github.com/ovh/ai-training-examples/blob/main/notebooks/computer-vision/image-classification/tensorflow/resnet50/notebook-resnet-transfer-learning-image-classification.ipynb)
107+
108+
- [Train YOLOv7 for American Sign Language recognition](https://github.com/ovh/ai-training-examples/blob/main/notebooks/computer-vision/object-detection/miniconda/yolov7/notebook_object_detection_yolov7_asl.ipynb)
109+
110+
## Feedback
111+
112+
Please send us your questions, feedback and suggestions to improve the service:
113+
114+
- On the OVHcloud [Discord server](https://discord.com/invite/vXVurFfwe9)

‎pages/platform/ai/notebook_tuto_12_image-segmentation-unet-tumors/guide.en-au.md

Lines changed: 114 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,114 @@
1+
---
2+
title: AI Notebooks - Tutorial - Brain tumor segmentation using U-Net
3+
slug: notebooks/tuto-image-segmentation-unet-tumors
4+
excerpt: Implementing a Convolutional Neural Network for Brain Tumor Segmentation in Medical Imaging
5+
section: AI Notebooks - Tutorials
6+
order: 12
7+
updated: 2023-04-13
8+
---
9+
10+
**Last updated 13th April, 2023.**
11+
12+
## Objective
13+
14+
Over the past few years, the field of **computer vision** has experienced a significant growth. It encompasses a wide range of methods for acquiring, processing, analyzing and understanding digital images.
15+
16+
Among these methods, one is called **image segmentation**.
17+
18+
The purpose of this tutorial is to show you how it is possible to build and train a brain tumor segmentation model with **OVHcloud AI Notebooks**. You will be able to learn the concepts of medical imaging, image segmentation, model evaluation, and much more. We will use a popular convolutional neural network named **U-Net**.
19+
20+
At the end of this tutorial, you will have learnt the principal methods to **segment brain tumors**.
21+
22+
![image](images/mri_orig_pred_segmentation.gif){.thumbnail}
23+
24+
> [!primary]
25+
>
26+
> We will train this model on the *[BraTS2020 dataset](http://braintumorsegmentation.org/)*. We will show you how you can easily download the dataset in the notebook tutorial.
27+
>
28+
29+
## Requirements
30+
31+
- Access to the [OVHcloud Control Panel](https://www.ovh.com/auth/?action=gotomanager&from=https://www.ovh.com.au/&ovhSubsidiary=au)
32+
- An AI Notebooks project created inside a [Public Cloud project](https://www.ovhcloud.com/en-au/public-cloud/) in your OVHcloud account
33+
- A user for AI Notebooks
34+
- A [Kaggle](https://www.kaggle.com/) account to download the dataset
35+
36+
## Instructions
37+
38+
You can launch the notebook from the [OVHcloud Control Panel](https://www.ovh.com/auth/?action=gotomanager&from=https://www.ovh.com.au/&ovhSubsidiary=au) or via the ovhai [CLI](https://docs.ovh.com/au/en/publiccloud/ai/cli/getting-started-cli/).
39+
40+
### Launching a Jupyter notebook with "Miniconda" via UI (Control Panel)
41+
42+
To launch your notebook from the [OVHcloud Control Panel](https://www.ovh.com/auth/?action=gotomanager&from=https://www.ovh.com.au/&ovhSubsidiary=au), refer to the following steps.
43+
44+
#### Code editor
45+
46+
Choose the `Jupyterlab` code editor.
47+
48+
#### Framework
49+
50+
In this tutorial, the `tensorflow` framework is used. If you use another environment, there may be some compatibility problems such as missing libraries.
51+
52+
#### Resources
53+
54+
Using GPUs is recommended because medical imaging is a training intensive task.
55+
56+
> [!primary]
57+
>
58+
> Here, using `1 GPU` is sufficient.
59+
>
60+
61+
### Launching a Jupyter notebook with "Miniconda" via CLI
62+
63+
*If you do not use our CLI yet, follow [this guide](https://docs.ovh.com/au/en/publiccloud/ai/cli/install-client/) to install it.*
64+
65+
If you want to launch your notebook with the OVHcloud AI CLI, choose the `jupyterlab` editor and the `tensorflow` framework.
66+
67+
To access the different versions of `tensorflow` available, run the following command:
68+
69+
```console
70+
ovhai capabilities framework list -o yaml
71+
```
72+
73+
> [!primary]
74+
>
75+
> If you do not specify a version, your notebook starts with the default version of `tensorflow`.
76+
>
77+
78+
You will also need to choose the number of CPUs/GPUs (`<nb-cpus>` or `<nb-gpus>`) to use in your notebook.
79+
80+
> [!primary]
81+
>
82+
> Here we recommend using `1 GPU`.
83+
>
84+
85+
To launch your notebook, run the following command:
86+
87+
```console
88+
ovhai notebook run tensorflow jupyterlab \
89+
--name <notebook-name> \
90+
--framework-version <tensorflow-version> \
91+
--gpu <nb-gpus>
92+
```
93+
94+
You can then reach your notebook’s URL once the notebook is running.
95+
96+
### Accessing the notebooks
97+
98+
Once our [AI examples repository](https://github.com/ovh/ai-training-examples/) has been cloned in your environment, find your notebook by following this path: `ai-training-examples` > `notebooks` > `computer-vision` > `image-segmentation` > `tensorflow` > `brain-tumor-segmentation-unet` > `notebook_image_segmentation_unet.ipynb`.
99+
100+
A preview of this notebook can be found on GitHub [here](https://github.com/ovh/ai-training-examples/blob/main/notebooks/computer-vision/image-segmentation/tensorflow/brain-tumor-segmentation-unet/notebook_image_segmentation_unet.ipynb).
101+
102+
## Go further
103+
104+
There are many other tasks that exist in the computer vision field. Check our other tutorials to learn how to:
105+
106+
- [Use Transfer Learning with ResNet50 for image classification](https://github.com/ovh/ai-training-examples/blob/main/notebooks/computer-vision/image-classification/tensorflow/resnet50/notebook-resnet-transfer-learning-image-classification.ipynb)
107+
108+
- [Train YOLOv7 for American Sign Language recognition](https://github.com/ovh/ai-training-examples/blob/main/notebooks/computer-vision/object-detection/miniconda/yolov7/notebook_object_detection_yolov7_asl.ipynb)
109+
110+
## Feedback
111+
112+
Please send us your questions, feedback and suggestions to improve the service:
113+
114+
- On the OVHcloud [Discord server](https://discord.com/invite/vXVurFfwe9)

0 commit comments

Comments
 (0)