The ICEBERG Penguin colony usecase package
Project description
Quality Metrics
Prerequisites
- Linux
- Python 3
- CPU or NVIDIA GPU + CUDA CuDNN
Software Dependencies
- scipy==1.2.1
- Pillow==4.3.0
- torch
- scikit-learn==0.19.1
- torchvision==0.2.0'
- opencv-python
- rasterio
- future
Installation
PSC Bridges
From source:
$ git clone https://github.com/iceberg-project/Penguins.git
$ module load cuda
$ module load python3
$ virtualenv iceberg_penguins
$ source iceberg_penguins/bin/activate
[iceberg_penguins] $ export PYTHONPATH=<path>/iceberg_penguins/lib/python3.5/site-packages
[iceberg_penguins] $ pip install . --upgrade
From PyPi:
$ module load cuda
$ module load python3
$ virtualenv iceberg_penguins
$ source iceberg_penguins/bin/activate
[iceberg_penguins] $ export PYTHONPATH=<path>/iceberg_penguins/lib/python3.5/site-packages
[iceberg_penguins] $ pip install iceberg_penguins.search
To test
[iceberg_penguins] $ iceberg_penguins.detect
Prediction
- Download a pre-trained model at:
https://drive.google.com/file/d/149j5rlynkO1jQTLOMpL5lextHY0ozw6N/view?usp=sharing
Please put the model file to: <checkpoints_dir>/<model_name>/
The one provided here is at the epoch 300 of the model named "v3weakly_unetr_bs96_main_model_ignore_bad"
- The script to run the testing for a single PNG image:
iceberg_penguins.detect [--params ...]
params:
- --name: name of the model used for testing
- --gpu_ids: the gpu used for testing
- --checkpoints_dir: path to the folder containing the trained models
- --epoch: which epoch we use to test the model
- --input_im: path to the input image
- --output: directory to save the outputs
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Close
Hashes for iceberg_penguins.search-0.2.7.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | 08674577e98137758774d07193865e645cb2e91e946ca70d8d6ad1e058b1a30b |
|
MD5 | a41852e9a627cb3cdb1eddc2205e7ebb |
|
BLAKE2b-256 | a36996041fbdb79c6adbb4691347c1224b328c947f44e83cd0a03ed32b72e82b |
Close
Hashes for iceberg_penguins.search-0.2.7-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 457c7357ca9b086d666a82b6269163e64ff657951946abca4f1338aebf377ffd |
|
MD5 | 42295e6ab4f8ca32e7ac36a4d83c811f |
|
BLAKE2b-256 | 03c9d833d27c984e82b1d458a6ae82de1453258671adee123cfa788764e2ffcf |