adding link to pretrained weights

README.md

@@ -39,9 +39,9 @@ Scripts assisting in the computation of heatmap visualizations can be found in f
 - The folder `model_definitions` contains the `*.prototxt` files for Caffe, i.e. a description of the model architecture each. Here, a naming pattern `[target]_[init]_[arch][_preproc]` applies, where
   + `target` is from `{age, gender}` and describes the prediction problem
   + `init` is from `{fromscratch, finetuning, imdbwiki}` and describes random initialization, a weight intialization from ImageNet pretraining, and a weight initialization from ImageNet pretraining followed by [IMDB-WIKI](https://data.vision.ee.ethz.ch/cvl/rrothe/imdb-wiki/) pretraining, respectively.
-  + `arch` is from `{caffereference, googlenet, vgg16, net_definitions}` and describes the architecture of the model. Here, `net_definitions` refers to the model architecture used in [Gil Levi's repo](https://github.com/GilLevi/AgeGenderDeepLearning).
-  + The `_preproc` suffix is optional and refers to `_unaligned` images (i.e. training images only under rotation alignment), aligned training images (landmark-based alignment, so suffix) or `_mixed` alignment, (i.e. both images under landmark-based and rotation-based alignment are used for trainng)
-  + [ ] TODO add starting weights to datacloud
+  + `arch` is from `{caffereference, googlenet, vgg16, net_definitions}` and describes the architecture of the model. Here, `net_definitions` refers to the model architecture used in [Gil Levi's repo](https://github.com/GilLevi/AgeGenderDeepLearning). The `net_definitions` models do not have an `init` block within the folder name.
+  + The `_preproc` suffix is optional and refers to `_unaligned` images (i.e. training images only under rotation alignment), aligned training images (landmark-based alignment, so suffix) or `_mixed` alignment, (i.e. both images under landmark-based and rotation-based alignment are used for training)
+  + The [pretrained](https://datacloud.hhi.fraunhofer.de/nextcloud/s/rakPNNWAQbeGCPx) models used as starting points (`init`) for training can be downloaded [here](https://datacloud.hhi.fraunhofer.de/nextcloud/s/rakPNNWAQbeGCPx). The model weights behind this link have been downloaded from the [Caffe repo](https://github.com/BVLC/caffe/tree/master/models) (`caffereference`, `googlenet`), the [IMDB-WIKI project page](https://data.vision.ee.ethz.ch/cvl/rrothe/imdb-wiki/) (`vgg16` on `imdbwiki`) and the [Caffe Model Zoo](https://github.com/BVLC/caffe/wiki/Model-Zoo#models-used-by-the-vgg-team-in-ilsvrc-2014) (`vgg16` on `imagenet`).
 - [The `lmdb` files](https://datacloud.hhi.fraunhofer.de/nextcloud/s/n6BLLnGPzinbe55) used for model training, validation testing can be downloaded [here](https://datacloud.hhi.fraunhofer.de/nextcloud/s/n6BLLnGPzinbe55).
 -  [The model weights](https://datacloud.hhi.fraunhofer.de/nextcloud/s/TQnGNJmQZLWkQ7X) (i.e. the `*.caffemodel` files) to the neural network descriptions contained in this repository can be downloaded [here](https://datacloud.hhi.fraunhofer.de/nextcloud/s/TQnGNJmQZLWkQ7X). These files match the model definitions in folder `model_definitions`
 - `heatmap_drawing` contains scripts generating configuration files for computing LRP heatmaps using the [LRP Toolbox](https://github.com/sebastian-lapuschkin/lrp_toolbox) for Caffe.