|
|
--- |
|
|
license: apache-2.0 |
|
|
language: |
|
|
- de |
|
|
- en |
|
|
metrics: |
|
|
- accuracy |
|
|
- roc_auc |
|
|
library_name: tf-keras |
|
|
pipeline_tag: image-classification |
|
|
--- |
|
|
# Brain tumor classification using CNN |
|
|
|
|
|
<!-- Provide a quick summary of what the model is/does. |
|
|
|
|
|
This modelcard aims to be a base template for new models. It has been generated using [this raw template](https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/templates/modelcard_template.md?plain=1). |
|
|
--> |
|
|
## Model Details |
|
|
|
|
|
### Model Description |
|
|
|
|
|
```python |
|
|
model.summary() |
|
|
Out: |
|
|
Model: "sequential" |
|
|
_________________________________________________________________ |
|
|
Layer (type) Output Shape Param # |
|
|
================================================================= |
|
|
rescaling (Rescaling) (None, 200, 200, 1) 0 |
|
|
|
|
|
conv2d (Conv2D) (None, 200, 200, 16) 160 |
|
|
|
|
|
max_pooling2d (MaxPooling2D (None, 100, 100, 16) 0 |
|
|
) |
|
|
|
|
|
conv2d_1 (Conv2D) (None, 100, 100, 32) 4640 |
|
|
|
|
|
max_pooling2d_1 (MaxPooling (None, 50, 50, 32) 0 |
|
|
2D) |
|
|
|
|
|
conv2d_2 (Conv2D) (None, 50, 50, 64) 18496 |
|
|
|
|
|
max_pooling2d_2 (MaxPooling (None, 25, 25, 64) 0 |
|
|
2D) |
|
|
|
|
|
flatten (Flatten) (None, 40000) 0 |
|
|
|
|
|
dense (Dense) (None, 128) 5120128 |
|
|
|
|
|
dense_1 (Dense) (None, 64) 8256 |
|
|
|
|
|
dense_2 (Dense) (None, 128) 8320 |
|
|
|
|
|
dense_3 (Dense) (None, 64) 8256 |
|
|
|
|
|
dense_4 (Dense) (None, 32) 2080 |
|
|
|
|
|
dense_5 (Dense) (None, 96) 3168 |
|
|
|
|
|
dense_6 (Dense) (None, 96) 9312 |
|
|
|
|
|
dense_7 (Dense) (None, 128) 12416 |
|
|
|
|
|
dense_8 (Dense) (None, 1) 129 |
|
|
|
|
|
================================================================= |
|
|
Total params: 5,195,361 |
|
|
Trainable params: 5,195,361 |
|
|
Non-trainable params: 0 |
|
|
_________________________________________________________________ |
|
|
|
|
|
``` |
|
|
|
|
|
### Dataset |
|
|
The dataset is composed of [Brain Tumor Classification (MRI)](https://www.kaggle.com/datasets/sartajbhuvaji/brain-tumor-classification-mri) and |
|
|
[Brain MRI Images for Brain Tumor Detection](https://www.kaggle.com/datasets/navoneel/brain-mri-images-for-brain-tumor-detection) |
|
|
|
|
|
Using image data augmentation we get 199.632 files belonging to 2 classes. |
|
|
|
|
|
train-test-split: 80/20 |
|
|
|
|
|
Training (159705 files): |
|
|
- Using 143735 files for training |
|
|
- Using 15970 files for validation |
|
|
|
|
|
Test/Validation: |
|
|
- 39927 files |
|
|
|
|
|
### Training |
|
|
Coming Soon |
|
|
|
|
|
### Validation |
|
|
Coming Soon |
|
|
|
|
|
#### Hardware |
|
|
|
|
|
Lenovo Thinkpad P14s |
|
|
- CPU (# Cores/Threads): AMD Ryzen 7 PRO 5850U (8/16) |
|
|
- RAM: 32 GB |
|
|
|
|
|
#### Software |
|
|
|
|
|
DataSpell 2023.1.2 |
|
|
- Python 3.10.9 |
|
|
- Tensorflow 2.12.0 |
