tf-semantic-segmentation
Semantic Segmentation using Tensorflow on popular Datasets like Ade20k, Camvid, Coco, PascalVoc
File Explorer
Download Latest Version (.zip)- Dockerfile
- Dockerfile
- activations.md
- datasets.md
- docker.md
- index.md
- metrics.md
- models.md
- predict.md
- requirements.txt
- tflite.md
- usage.md
- Demo.ipynb
- FindLR.ipynb
- FindOptimalBatchSize.ipynb
- Predict.ipynb
- Train.ipynb
- docker-compose-install.sh
- nvidia-docker-setup.sh
- __init__.py
- tfrecord_writer_test.py
- train_test.py
- __init__.py
- test_dataset.py
- test_shapes.py
- test_tfrecord.py
- test_utils.py
- __init__.py
- test_export_dataset.py
- __init__.py
- test_utils.py
- __init__.py
- dataset_test.py
- image_test.py
- __init__.py
- mask_test.py
- __init__.py
- data.py
- fixtures.py
- requirements.txt
- test.png
- test_activations.py
- test_apps.py
- test_losses.py
- test_metrics.py
- test_models.py
- test_threading.py
- test_utils.py
- __init__.py
- __init__.py
- convert_tflite.py
- download.py
- model_server_config_writer.py
- tfrecord_analyser.py
- tfrecord_download.py
- tfrecord_writer.py
- train.py
- train_all.py
- __init__.py
- ade20k.py
- bioimage.py
- camvid.py
- cityscapes.py
- cub.py
- cvc_clinicdb.py
- dataset.py
- directory.py
- isic.py
- mapping_challenge.py
- mots_challenge.py
- ms_coco.py
- pascal.py
- shapes.py
- sun.py
- taco.py
- tfrecord.py
- toy.py
- utils.py
- __init__.py
- dataset_export.py
- dataset_vis.py
- devices.py
- export_saved_model.py
- model_parameters.py
- preprocessing_vis.py
- record_vis.py
- tflite_test.py
- __init__.py
- compare_models.py
- eval_loss.py
- predict.py
- video.py
- viewer.py
- __init__.py
- conv.py
- minibatchstddev.py
- pixel_norm.py
- subpixel.py
- utils.py
- __init__.py
- ce.py
- combined.py
- dice.py
- focal.py
- lovasz.py
- ssim.py
- utils.py
- __init__.py
- f_scores.py
- iou_score.py
- kmetrics.py
- precision.py
- psnr.py
- recall.py
- ssim.py
- __init__.py
- inception.py
- mobilenet.py
- resnet50.py
- utils.py
- __init__.py
- attention_unet.py
- deeplabv3.py
- deeplabv3plus.py
- erfnet.py
- fcn.py
- imagenet_unet.py
- multires_unet.py
- nested_unet.py
- psp.py
- satellite_unet.py
- u2net.py
- unet.py
- __init__.py
- __init__.py
- dataset.py
- image.py
- __init__.py
- masks.py
- show.py
- __init__.py
- callbacks.py
- serving.py
- settings.py
- threading.py
- utils.py
- version.py
- .codecov.xml
- .gitignore
- .readthedocs.yml
- .travis.yml
- LICENSE
- MANIFEST.in
- mkdocs.yml
- README.md
- requirements.txt
- ROADMAP.md
- setup.cfg
- setup.py
- sweep.example.yaml
# Installation Guide
git clone https://github.com/baudcode/tf-semantic-segmentation
Downloads the entire project code from GitHub to your computer.
cd tf-semantic-segmentation
Moves into the project folder you just downloaded.
2. Official Install Script
Easy Recommended- APT (Debian/Ubuntu κ³μ΄) Built into Debian/Ubuntu-based Linux distributions.
- Python 3 Python is required to use pip.
sudo apt-get install libsm6 libxext6 libxrender-dev libyaml-dev libpython3-dev
Installs directly from the APT package repository (Debian/Ubuntu-based).
pip install tensorflow-gpu==2.4.0 --upgrade
Installs the package published on PyPI directly β no need to clone the source.
pip install tensorflow-addons==0.12.0 --upgrade
Installs the package published on PyPI directly β no need to clone the source.
pip install tf-semantic-segmentation
Installs the package published on PyPI directly β no need to clone the source.
Pulled directly from this repo's README.
3. Docker
Easy- Git Needed to download the project code from GitHub.
- Docker Desktop Needed to build and run containers. Install it and keep it running in the background.
docker build -t tf_semantic_segmentation -f docker/Dockerfile ./
Builds a runnable image based on the Dockerfile.
docker pull baudcode/tf_semantic_segmentation:latest
Type this command into your terminal and run it.
Pulled directly from this repo's README.
4. Python
Easypip install tensorflow-gpu==2.4.0 --upgrade
Installs the package published on PyPI directly β no need to clone the source.
pip install tensorflow-addons==0.12.0 --upgrade
Installs the package published on PyPI directly β no need to clone the source.
pip install tf-semantic-segmentation
Installs the package published on PyPI directly β no need to clone the source.
python -m tf_semantic_segmentation.bin.train -ds 'tacobinary' -bs 8 -e 100 \
Runs the Python script (or module).
python -m tf_semantic_segmentation.bin.train --record_dir=records/cityscapes-512x256-rgb/ \
Runs the Python script (or module).
Pulled directly from this repo's README.