Follow original repo's guide
https://github.com/thtrieu/darkflow
Python3, tensorflow 1.0, numpy, opencv 3.
You can choose one of the following three ways to get started with darkflow.
-
Just build the Cython extensions in place. NOTE: If installing this way you will have to use
./flowin the cloned darkflow directory instead offlowas darkflow is not installed globally.python3 setup.py build_ext --inplace -
Let pip install darkflow globally in dev mode (still globally accessible, but changes to the code immediately take effect)
pip install -e . -
Install with pip globally
pip install .
Load tiny-yolo.weights
-
DL tiny-yolo.weights
https://pjreddie.com/darknet/yolo/ -
Copy & paste tiny-yolo.cfg
-
change self.offset = 16 --> 20
the file is darkflow/utils/loader.py line 121
$ flow --model cfg/tiny-yolo.cfg --load bin/tiny-yolo.weights
If the commands works properly, you will get below.
Parsing ./cfg/tiny-yolo.cfg
Parsing cfg/tiny-yolo.cfg
Loading bin/tiny-yolo.weights ...
Successfully identified 44948600 bytes
Finished in 0.009058952331542969s
Model has a coco model name, loading coco labels.
Building net ...
Source | Train? | Layer description | Output size
-------+--------+----------------------------------+---------------
| | input | (?, 416, 416, 3)
Load | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 416, 416, 16)
Load | Yep! | maxp 2x2p0_2 | (?, 208, 208, 16)
Load | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 208, 208, 32)
Load | Yep! | maxp 2x2p0_2 | (?, 104, 104, 32)
Load | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 104, 104, 64)
Load | Yep! | maxp 2x2p0_2 | (?, 52, 52, 64)
Load | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 52, 52, 128)
Load | Yep! | maxp 2x2p0_2 | (?, 26, 26, 128)
Load | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 26, 26, 256)
Load | Yep! | maxp 2x2p0_2 | (?, 13, 13, 256)
Load | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 13, 13, 512)
Load | Yep! | maxp 2x2p0_1 | (?, 13, 13, 512)
Load | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 13, 13, 1024)
Load | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 13, 13, 512)
Load | Yep! | conv 1x1p0_1 linear | (?, 13, 13, 425)
-------+--------+----------------------------------+---------------
Running entirely on CPU
2017-11-17 21:01:11.893502: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations.
2017-11-17 21:01:11.893533: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations.
2017-11-17 21:01:11.893540: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.
2017-11-17 21:01:11.893545: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX2 instructions, but these are available on your machine and could speed up CPU computations.
2017-11-17 21:01:11.893550: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use FMA instructions, but these are available on your machine and could speed up CPU computations.
Finished in 4.230991840362549s
Forwarding 8 inputs ...
Total time = 4.750931024551392s / 8 inps = 1.6838804770387932 ips
Post processing 8 inputs ...
Total time = 0.674644947052002s / 8 inps = 11.858089258590944 ips
test images
$ flow --imgdir sample_img/ --model cfg/tiny-yolo.cfg --load bin/tiny-yolo.weightsMy mac is 4GB and core i5, so not good one but it just took 20 secs and the results are awesome.