From 387f659962e28c042013c2dc63ae22bc1f47a8df Mon Sep 17 00:00:00 2001 From: Wei Ouyang Date: Fri, 25 May 2018 11:56:26 +0200 Subject: [PATCH] Update README.md --- README.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/README.md b/README.md index ae16589..6641b46 100644 --- a/README.md +++ b/README.md @@ -21,15 +21,15 @@ For more details, please checkout the [ANNA-PALM project website](https://annapa Sample images can be downloaded from here: [STORM image 1000 frames (used as figure 4 in the paper), cropped to 512x512](./img/figure4c-PALM-1000frames-512crop.tif), [full size version (2560x2560)](./img/figure4c-PALM-1000frames.tif), more sample images can be exported from [https://annapalm.pasteur.fr](https://annapalm.pasteur.fr/#/repository). ## Use your own model -A set of pre-trained models is provided, a shortcut to download these models is to use the `download` button in the plugin. You can check directly in a folder named `anet-models` in your ImageJ folder (next to the `plugin` folder). You can also download it manually from [here](https://github.com/imodpasteur/Anet-ImageJ/releases/download/0.2.2/models_v0.1.zip). +A set of pre-trained models are provided, a shortcut to download these models is to use the `download` button in the plugin. You can check directly in a folder named `anet-models` in your ImageJ folder (next to the `plugin` folder). You can also download a zip containing the models manually from [here](https://github.com/imodpasteur/Anet-ImageJ/releases/download/0.2.2/models_v0.1.zip). ![model folder structure](./img/model-folder-structure.png) Model files for the plugin consist of two files: a `tensorflow_model.pb` file for the frozen model, and a `config.json` file for the configuration which describe the GUI, input and output of the model. -You can create a new folder (e.g. named "mito_model_v2.3") inside the `anet-models` folder in ImageJ. Then place your own model file(`tensorflow_model.pb`) generated by the `freeze.py` file in [ANNA-PALM](https://github.com/imodpasteur/ANNA-PALM). To generate the GUI, you need to manually add a `config.json` file next to your model file `tensorflow_model.pb`. To simplify the process, you can just copy an existing `config.json` file provided by us. In case you have a different input image size, you will need to replace all the `512` into your own size in `config.json`. +You can create a new folder (e.g. named "mito_model_v2.3") inside the `anet-models` folder in ImageJ. Then place your own model file(`tensorflow_model.pb`) generated with the `freeze.py` script in [ANNA-PALM](https://github.com/imodpasteur/ANNA-PALM). To define the GUI and describe the model, you need to manually add a `config.json` file next to your model file `tensorflow_model.pb`. To simplify the process, you can just copy an existing `config.json` file provided by us. In case you have a different input image size, you will need to replace all the `512` into your own size in `config.json`. -Once you done that, run `Setup A-net`, and you should be able to see your model shown in the model list. Then select it and process images with your own model. +Once you're done, run `Setup A-net` from the menu and you should be able to see your model shown in the model list. Then select it and process images with your own model. # Citation Please cite our paper: `Ouyang et al., Nat. Biotechnol. 2018, doi:10.1038/nbt.4106`