- [Continue Training on the Pretrained Model](#continue-training-on-the-pretrained-model)
- [Description of Random Situation](#description-of-random-situation)
- [ModelZoo Homepage](#modelzoo-homepage)
# [Unet Description](#contents)
Unet Medical model for 2D image segmentation. This implementation is as described in the original paper [UNet: Convolutional Networks for Biomedical Image Segmentation](https://arxiv.org/abs/1505.04597). Unet, in the 2015 ISBI cell tracking competition, many of the best are obtained. In this paper, a network model for medical image segmentation is proposed, and a data enhancement method is proposed to effectively use the annotation data to solve the problem of insufficient annotation data in the medical field. A U-shaped network structure is also used to extract the context and location information.
[Paper](https://arxiv.org/abs/1505.04597): Olaf Ronneberger, Philipp Fischer, Thomas Brox. "U-Net: Convolutional Networks for Biomedical Image Segmentation." * conditionally accepted at MICCAI 2015*. 2015.
Specifically, the U network structure is proposed in UNET, which can better extract and fuse high-level features and obtain context information and spatial location information. The U network structure is composed of encoder and decoder. The encoder is composed of two 3x3 conv and a 2x2 max pooling iteration. The number of channels is doubled after each down sampling. The decoder is composed of a 2x2 deconv, concat layer and two 3x3 convolutions, and then outputs after a 1x1 convolution.
- Description: The training and test datasets are two stacks of 30 sections from a serial section Transmission Electron Microscopy (ssTEM) data set of the Drosophila first instar larva ventral nerve cord (VNC). The microcube measures 2 x 2 x 1.5 microns approx., with a resolution of 4x4x50 nm/pixel.
- License: You are free to use this data set for the purpose of generating or testing non-commercial image segmentation software. If any scientific publications derive from the usage of this data set, you must cite TrakEM2 and the following publication: Cardona A, Saalfeld S, Preibisch S, Schmid B, Cheng A, Pulokas J, Tomancak P, Hartenstein V. 2010. An Integrated Micro- and Macroarchitectural Analysis of the Drosophila Brain by Computer-Assisted Serial Section Electron Microscopy. PLoS Biol 8(10): e1000502. doi:10.1371/journal.pbio.1000502.
- Dataset size：22.5M，
- Train：15M, 30 images (Training data contains 2 multi-page TIF files, each containing 30 2D-images. train-volume.tif and train-labels.tif respectly contain data and label.)
- Val：(We randomly divde the training data into 5-fold and evaluate the model by across 5-fold cross-validation.)
- Test：7.5M, 30 images (Testing data contains 1 multi-page TIF files, each containing 30 2D-images. test-volume.tif respectly contain data.)
- Data format：binary files(TIF file)
- Note：Data will be processed in src/data_loader.py
# [Environment Requirements](#contents)
- Hardware（Ascend）
- Prepare hardware environment with Ascend processor. If you want to try Ascend, please send the [application form](https://obs-9be7.obs.cn-east-2.myhuaweicloud.com/file/other/Ascend%20Model%20Zoo%E4%BD%93%E9%AA%8C%E8%B5%84%E6%BA%90%E7%94%B3%E8%AF%B7%E8%A1%A8.docx) to ascend@huawei.com. Once approved, you can get the resources.
The above shell script will run distribute training in the background. You can view the results through the file `logs/device[X]/log.log`. The loss value will be achieved as follows:
```
# grep "loss is" logs/device0/log.log
step: 1, loss is 0.70524895, fps is 0.15914689861221412
step: 2, loss is 0.6925452, fps is 56.43668656967454
...
step: 299, loss is 0.20551169, fps is 58.4039329983891
step: 300, loss is 0.18949677, fps is 57.63118508760329
```
## [Evaluation Process](#contents)
### Evaluation
- evaluation on ISBI dataset when running on Ascend
Before running the command below, please check the checkpoint path used for evaluation. Please set the checkpoint path to be the absolute full path, e.g., "username/unet/ckpt_unet_medical_adam-48_600.ckpt".
The above python command will run in the background. You can view the results through the file "eval.log". The accuracy of the test dataset will be as follows:
```
# grep "Cross valid dice coeff is:" eval.log
============== Cross valid dice coeff is: {'dice_coeff': 0.9085704886070473}
If you need to use the trained model to perform inference on multiple hardware platforms, such as Ascend 910 or Ascend 310, you can refer to this [Link](https://www.mindspore.cn/tutorial/training/en/master/advanced_use/migrate_3rd_scripts.html). Following the steps below, this is a simple example:
