urban-environments
urban-environments copied to clipboard
adrianalbert
Code for constructing the urban environments dataset and for land use classification via convolutional neural networks
**For context**: I'm doing a research project and plan to use machine learning to forecast last-mile delivery times using the dataset from Amazon that they made available for their last-mile...
In [Pipeline-to-create-Urban-Environments-dataset.ipynb](https://github.com/adrianalbert/urban-environments/blob/master/dataset-collection/Pipeline-to-create-Urban-Environments-dataset.ipynb) In this block: ``` # compute ground traster for given window size raster, locations_grid, cur_classes = mycity_crop.extract_class_raster(grid_size=grid_size) myraster = np.zeros(grid_size + (len(classes),)) idx = [class2label[c] for k,c in...
It doesn't work properly for build the urbanatlas. File "", line 52 print "%d polygons | %d land use classes" % (len(self._gdf), len(self._classes)) ^ SyntaxError: invalid syntax
Hi, I am building ipython code for experiment purpose. But I am not able to build "Similar images using CNN features.ipynb" due to the reason classes.pickle.gz is not found. Please...
 Great work, I am curious about the code for visualizing features using the t-SNE algorithm. I wonder how this visualization is done? Could you provide me with some advice...
Hi [Adrian](https://github.com/adrianalbert) I would like to try your model. Are you interested to try it on some aerial drone images?
Metadata
Owner
Metadata
Code for constructing the urban environments dataset and for land use classification via convolutional neural networks