更新

.gitignore

@@ -25,6 +25,12 @@ dist/
 # Pycharm editor settings
 .idea
 
+# vscode editor settings
+.vscode
+
+# MacOS
+.DS_Store
+
 # project dirs
 /datasets
 /models

INSTALL.md

@@ -61,7 +61,58 @@ unset INSTALL_DIR
 # or if you are on macOS
 # MACOSX_DEPLOYMENT_TARGET=10.9 CC=clang CXX=clang++ python setup.py build develop
 ```
+#### Windows 10 
+```bash
+open a cmd and change to desired installation directory 
+from now on will be refered as INSTALL_DIR 
+conda create --name maskrcnn_benchmark
+conda activate maskrcnn_benchmark
+
+# this installs the right pip and dependencies for the fresh python
+conda install ipython
+
+# maskrcnn_benchmark and coco api dependencies
+pip install ninja yacs cython matplotlib tqdm opencv-python
+
+# follow PyTorch installation in https://pytorch.org/get-started/locally/
+# we give the instructions for CUDA 9.0
+## Important : check the cuda version installed on your computer by running the command in the cmd : 
+nvcc -- version 
+conda install -c pytorch pytorch-nightly torchvision cudatoolkit=9.0
+
+git clone https://github.com/cocodataset/cocoapi.git
+
+    #To prevent installation error do the following after commiting cocooapi : 
+    #using file explorer  naviagate to cocoapi\PythonAPI\setup.py and change line 14 from:
+    #extra_compile_args=['-Wno-cpp', '-Wno-unused-function', '-std=c99'],
+    #to
+    #extra_compile_args={'gcc': ['/Qstd=c99']},
+    #Based on  https://github.com/cocodataset/cocoapi/issues/51
+
+cd cocoapi/PythonAPI
+python setup.py build_ext install
+
+# navigate back to INSTALL_DIR
+cd ..
+cd .. 
+# install apex
+
+git clone https://github.com/NVIDIA/apex.git
+cd apex
+python setup.py install --cuda_ext --cpp_ext
+# navigate back to INSTALL_DIR
+cd .. 
+# install PyTorch Detection
 
+git clone https://github.com/Idolized22/maskrcnn-benchmark.git
+cd maskrcnn-benchmark
+
+# the following will install the lib with
+# symbolic links, so that you can modify
+# the files if you want and won't need to
+# re-build it
+python setup.py build develop
+```
 ### Option 2: Docker Image (Requires CUDA, Linux only)
 
 Build image with defaults (`CUDA=9.0`, `CUDNN=7`, `FORCE_CUDA=1`):

README.md

@@ -10,6 +10,7 @@ creating detection and segmentation models using PyTorch 1.0.
 - **Very fast**: up to **2x** faster than [Detectron](https://github.com/facebookresearch/Detectron) and **30%** faster than [mmdetection](https://github.com/open-mmlab/mmdetection) during training. See [MODEL_ZOO.md](MODEL_ZOO.md) for more details.
 - **Memory efficient:** uses roughly 500MB less GPU memory than mmdetection during training
 - **Multi-GPU training and inference**
+- **Mixed precision training:** trains faster with less GPU memory on [NVIDIA tensor cores](https://developer.nvidia.com/tensor-cores).
 - **Batched inference:** can perform inference using multiple images per batch per GPU
 - **CPU support for inference:** runs on CPU in inference time. See our [webcam demo](demo) for an example
 - Provides pre-trained models for almost all reference Mask R-CNN and Faster R-CNN configurations with 1x schedule.
@@ -152,6 +153,15 @@ python -m torch.distributed.launch --nproc_per_node=$NGPUS /path_to_maskrcnn_ben
 ```
 Note we should set `MODEL.RPN.FPN_POST_NMS_TOP_N_TRAIN` follow the rule in Single-GPU training.
 
+### Mixed precision training
+We currently use [APEX](https://github.com/NVIDIA/apex) to add [Automatic Mixed Precision](https://developer.nvidia.com/automatic-mixed-precision) support. To enable, just do Single-GPU or Multi-GPU training and set `DTYPE "float16"`.
+
+```bash
+export NGPUS=8
+python -m torch.distributed.launch --nproc_per_node=$NGPUS /path_to_maskrcnn_benchmark/tools/train_net.py --config-file "path/to/config/file.yaml" MODEL.RPN.FPN_POST_NMS_TOP_N_TRAIN images_per_gpu x 1000 DTYPE "float16"
+```
+If you want more verbose logging, set `AMP_VERBOSE True`. See [Mixed Precision Training guide](https://docs.nvidia.com/deeplearning/sdk/mixed-precision-training/index.html) for more details.
+
 ## Evaluation
 You can test your model directly on single or multiple gpus. Here is an example for Mask R-CNN R-50 FPN with the 1x schedule on 8 GPUS:
 ```bash

configs/test_time_aug/e2e_mask_rcnn_R_50_FPN_1x.yaml

@@ -0,0 +1,48 @@
+MODEL:
+  META_ARCHITECTURE: "GeneralizedRCNN"
+  WEIGHT: "catalog://ImageNetPretrained/MSRA/R-50"
+  BACKBONE:
+    CONV_BODY: "R-50-FPN"
+  RESNETS:
+    BACKBONE_OUT_CHANNELS: 256
+  RPN:
+    USE_FPN: True
+    ANCHOR_STRIDE: (4, 8, 16, 32, 64)
+    PRE_NMS_TOP_N_TRAIN: 2000
+    PRE_NMS_TOP_N_TEST: 1000
+    POST_NMS_TOP_N_TEST: 1000
+    FPN_POST_NMS_TOP_N_TEST: 1000
+  ROI_HEADS:
+    USE_FPN: True
+  ROI_BOX_HEAD:
+    POOLER_RESOLUTION: 7
+    POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125)
+    POOLER_SAMPLING_RATIO: 2
+    FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor"
+    PREDICTOR: "FPNPredictor"
+  ROI_MASK_HEAD:
+    POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125)
+    FEATURE_EXTRACTOR: "MaskRCNNFPNFeatureExtractor"
+    PREDICTOR: "MaskRCNNC4Predictor"
+    POOLER_RESOLUTION: 14
+    POOLER_SAMPLING_RATIO: 2
+    RESOLUTION: 28
+    SHARE_BOX_FEATURE_EXTRACTOR: False
+  MASK_ON: True
+DATASETS:
+  TRAIN: ("coco_2014_train", "coco_2014_valminusminival")
+  TEST: ("coco_2014_minival",)
+DATALOADER:
+  SIZE_DIVISIBILITY: 32
+SOLVER:
+  BASE_LR: 0.02
+  WEIGHT_DECAY: 0.0001
+  STEPS: (60000, 80000)
+  MAX_ITER: 90000
+TEST:
+  BBOX_AUG:
+    ENABLED: True
+    H_FLIP: True
+    SCALES: (400, 500, 600, 700, 900, 1000, 1100, 1200)
+    MAX_SIZE: 2000
+    SCALE_H_FLIP: True