udacity / carnd-object-detection-lab Goto Github PK

View Code? Open in Web Editor NEW

100.0 14.0 339.0 25.47 MB

License: MIT License

Jupyter Notebook 95.05% Python 4.95%

carnd-object-detection-lab's Introduction

CarND Object Detection Lab

In lab this you will:

Learn about MobileNets and separable depthwise convolutions.
The SSD (Single Shot Detection) architecture used for object detection
Use pretrained TensorFlow object detection inference models to detect objects
Use different architectures and weigh the tradeoffs.
Apply an object detection pipeline to a video.

Open the notebook and work through it!

Requirements

Install environment with Anaconda:

conda env create -f environment.yml

Change TensorFlow pip installation from tensorflow-gpu to tensorflow if you don't have a GPU available.

The environment should be listed via conda info --envs:

# conda environments:
#
carnd-advdl-odlab        /usr/local/anaconda3/envs/carnd-advdl-odlab
root                  *  /usr/local/anaconda3

Further documentation on working with Anaconda environments.

Particularly useful sections:

https://conda.io/docs/using/envs.html#change-environments-activate-deactivate https://conda.io/docs/using/envs.html#remove-an-environment

Resources

TensorFlow object detection model zoos: V1 and V2.
Driving video

Tips

Some windows users have reported the driving video as playable only in Jupyter Notebook operating in Chrome browser, and not in media player or Jupyter Notebook operating in other browsers. In contrast the post-segmentation video appears to be operating across players and browsers.

carnd-object-detection-lab's People

Contributors

Stargazers

Watchers

Forkers

l1aoxingyu arvindshekar07 artxj vguerra amintahmasbi d13sl0w dparamoshkin anshoomehra goddard brian-mchugh hristo-vrigazov ashispapu gojira naushad-rahman svanimisetti jjjjohnson xcarnd shukla-san1 boxmein liaochiheng christianmarzahl prasshantg gaylordmarville jinchenglee subhash darienmt jamdang goo00d omnigeeker bcchangebfe yubrshen jorvidas ricgu8086 ra9hur steffens21 calsaviour sameersg dlzhangxg cesare-montresor bhushan017 wmrifenb ajeet-repos sonofram cybercoderbot piliwilliam0306 meghamaheshwari wycg1984 msrks lostinet lookingin97 sunnyshah0903 vishalkharjul mreichelt logic4twisted hecspc gantners guyk1971 mplaza dmitrykudinov deepdriving ttakahashi765 nishanthjois jlee21 asimay nanfengpo apalvai kamil-k yifenghuang geminijun bigdan88 drganjoo samipshah abhaycl m00nd00r koalabella kanhua pabloelizalde atolmid momr hexiangquan mummummum bkanaki jolen217 babatakahiro1983 ykirpichev ahubi lajos rstanchak volkodava dushshantha clockworkbits bryan-rathos tianhao-david-hu vladpaunescu mark5udacity polarbeargo yuxiangjohn flourishlily frankcarey geek101

carnd-object-detection-lab's Issues

code in this environment not responding

When I run the code cell, it's taking forever to finish and fail in the end.

computation cost is wrong

Instead of Df * Df * M * N * Dk * Dk, it should be Dg * Dg * M * N * Dk * Dk with Dg replacing Df.

Same applies to depthwise and pointwise convolution.

Anaconda build fail

conda env create -f environment.yml

`Found conflicts! Looking for incompatible packages.
This can take several minutes. Press CTRL-C to abort.
failed

UnsatisfiableError: The following specifications were found to be incompatible with each other:

Package python conflicts for:
imageio==2.1.2 -> numpy -> mkl-service[version='>=2,<3.0a0'] -> six -> python[version='3.4.|>=3.7,<3.8.0a0|>=3.8,<3.9.0a0']
imageio==2.1.2 -> numpy -> python[version='>=2.7,<2.8.0a0|>=3.5,<3.6.0a0|>=3.6,<3.7.0a0']
python==3.6
imageio==2.1.2 -> python[version='2.7.|3.5.|3.6.']
jupyter -> ipykernel -> tornado[version='>=4.0'] -> certifi -> python=3.4
matplotlib -> cycler[version='>=0.10'] -> python[version='2.7.|3.5.|3.6.|<3']
jupyter -> ipywidgets -> nbformat[version='>=4.2.0'] -> jsonschema[version='>=2.0,!=2.5.0'] -> functools32 -> python[version='<3']
numpy -> python[version='2.7.|3.4.|3.5.|3.6.|>=2.7,<2.8.0a0|>=3.5,<3.6.0a0|>=3.6,<3.7.0a0|>=3.7,<3.8.0a0|>=3.8,<3.9.0a0']
pillow -> python[version='2.7.|3.4.|3.5.|3.6.|>=2.7,<2.8.0a0|>=3.5,<3.6.0a0|>=3.6,<3.7.0a0|>=3.7,<3.8.0a0|>=3.8,<3.9.0a0']
matplotlib -> cycler[version='>=0.10'] -> six -> python=3.4
imageio==2.1.2 -> pillow -> olefile -> python
pip -> python[version='2.7.|3.4.|3.5.|3.6.|>=2.7,<2.8.0a0|>=3.5,<3.6.0a0|>=3.6,<3.7.0a0|>=3.7,<3.8.0a0|>=3.8,<3.9.0a0']
pillow -> olefile -> python
jupyter -> python[version='2.7.|3.5.|3.6.|>=2.7,<2.8.0a0|>=3.5,<3.6.0a0|>=3.6,<3.7.0a0|>=3.7,<3.8.0a0']
scipy -> python[version='2.7.|3.4.|3.5.|3.6.|>=2.7,<2.8.0a0|>=3.5,<3.6.0a0|>=3.6,<3.7.0a0|>=3.7,<3.8.0a0|>=3.8,<3.9.0a0']
Package gtk2 conflicts for:
jupyter -> qtconsole -> pyqt[version='>=5.9.2,<5.10.0a0'] -> qt[version='>=4.8.6,<5.0'] -> gtk2
matplotlib -> pyqt -> qt[version='>=4.8.6,<5.0'] -> gtk2
Package tk conflicts for:
scipy -> python[version='>=3.6,<3.7.0a0'] -> tk[version='8.5.|8.6.|>=8.6.10,<8.7.0a0|>=8.6.7,<8.7.0a0|>=8.6.8,<8.7.0a0|>=8.6.9,<8.7.0a0']
pillow -> tk[version='8.6.|>=8.6.7,<8.7.0a0|>=8.6.8,<8.7.0a0|>=8.6.9,<8.7.0a0']
matplotlib -> tk[version='8.6.|>=8.6.7,<8.7.0a0|>=8.6.8,<8.7.0a0']
jupyter -> python[version='>=2.7,<2.8.0a0'] -> tk[version='8.6.|>=8.6.10,<8.7.0a0|>=8.6.7,<8.7.0a0|>=8.6.8,<8.7.0a0|>=8.6.9,<8.7.0a0']
matplotlib -> python[version='>=3.7,<3.8.0a0'] -> tk[version='>=8.6.10,<8.7.0a0|>=8.6.9,<8.7.0a0']
pillow -> python[version='>=2.7,<2.8.0a0'] -> tk[version='8.5.|>=8.6.10,<8.7.0a0']
python==3.6 -> tk=8.5
imageio==2.1.2 -> pillow -> tk[version='8.5.|8.6.|>=8.6.7,<8.7.0a0|>=8.6.8,<8.7.0a0|>=8.6.9,<8.7.0a0']
pip -> python=3.5 -> tk[version='8.5.|8.6.|>=8.6.10,<8.7.0a0|>=8.6.7,<8.7.0a0|>=8.6.8,<8.7.0a0|>=8.6.9,<8.7.0a0']
numpy -> python[version='>=3.5,<3.6.0a0'] -> tk[version='8.5.|8.6.|>=8.6.10,<8.7.0a0|>=8.6.7,<8.7.0a0|>=8.6.8,<8.7.0a0|>=8.6.9,<8.7.0a0']
Package readline conflicts for:
pillow -> python[version='>=2.7,<2.8.0a0'] -> readline[version='6.2.|7.|7.0.|7.0|>=7.0,<8.0a0|>=8.0,<9.0a0']
jupyter -> python[version='>=2.7,<2.8.0a0'] -> readline[version='7.|7.0.|7.0|>=7.0,<8.0a0|>=8.0,<9.0a0']
matplotlib -> python[version='>=3.7,<3.8.0a0'] -> readline[version='7.|7.0.|7.0|>=7.0,<8.0a0|>=8.0,<9.0a0']
pip -> python=3.5 -> readline[version='6.2.|7.|7.0.|7.0|>=7.0,<8.0a0|>=8.0,<9.0a0']
imageio==2.1.2 -> python=3.6 -> readline[version='6.2.|7.|7.0.|7.0|>=7.0,<8.0a0|>=8.0,<9.0a0']
scipy -> python[version='>=3.6,<3.7.0a0'] -> readline[version='6.2.|7.|7.0.|7.0|>=7.0,<8.0a0|>=8.0,<9.0a0']
numpy -> python[version='>=3.5,<3.6.0a0'] -> readline[version='6.2.|7.|7.0.|7.0|>=7.0,<8.0a0|>=8.0,<9.0a0']
python==3.6 -> readline=6.2
Package atk conflicts for:
matplotlib -> pyqt -> qt[version='>=4.8.6,<5.0'] -> gtk2 -> atk[version='>=2.32.0,<3.0a0']
jupyter -> qtconsole -> pyqt[version='>=5.9.2,<5.10.0a0'] -> qt[version='>=4.8.6,<5.0'] -> gtk2 -> atk[version='>=2.32.0,<3.0a0']
Package ipython conflicts for:
jupyter -> ipykernel -> ipython[version='>=4.0|>=4.0.0|>=5.0']
Package sqlite conflicts for:
pillow -> python[version='>=2.7,<2.8.0a0'] -> sqlite[version='3.13.|3.20.|3.9.|>=3.20.1,<4.0a0|>=3.22.0,<4.0a0|>=3.23.1,<4.0a0|>=3.24.0,<4.0a0|>=3.25.1,<4.0a0|>=3.25.2,<4.0a0|>=3.25.3,<4.0a0|>=3.26.0,<4.0a0|>=3.27.2,<4.0a0|>=3.28.0,<4.0a0|>=3.29.0,<4.0a0|>=3.30.0,<4.0a0|>=3.30.1,<4.0a0']
pip -> python=3.5 -> sqlite[version='3.13.|3.20.|3.9.|>=3.20.1,<4.0a0|>=3.22.0,<4.0a0|>=3.23.1,<4.0a0|>=3.24.0,<4.0a0|>=3.25.1,<4.0a0|>=3.25.2,<4.0a0|>=3.25.3,<4.0a0|>=3.26.0,<4.0a0|>=3.27.2,<4.0a0|>=3.28.0,<4.0a0|>=3.29.0,<4.0a0|>=3.30.0,<4.0a0|>=3.30.1,<4.0a0']
numpy -> python[version='>=3.5,<3.6.0a0'] -> sqlite[version='3.13.|3.20.|3.9.|>=3.20.1,<4.0a0|>=3.22.0,<4.0a0|>=3.23.1,<4.0a0|>=3.24.0,<4.0a0|>=3.25.1,<4.0a0|>=3.25.2,<4.0a0|>=3.25.3,<4.0a0|>=3.26.0,<4.0a0|>=3.27.2,<4.0a0|>=3.28.0,<4.0a0|>=3.29.0,<4.0a0|>=3.30.0,<4.0a0|>=3.30.1,<4.0a0']
imageio==2.1.2 -> python=3.6 -> sqlite[version='3.13.|3.20.|3.9.|>=3.20.1,<4.0a0|>=3.22.0,<4.0a0|>=3.23.1,<4.0a0|>=3.24.0,<4.0a0|>=3.25.2,<4.0a0|>=3.25.3,<4.0a0|>=3.26.0,<4.0a0|>=3.27.2,<4.0a0|>=3.28.0,<4.0a0|>=3.29.0,<4.0a0|>=3.30.0,<4.0a0|>=3.30.1,<4.0a0']
jupyter -> python[version='>=2.7,<2.8.0a0'] -> sqlite[version='3.20.|>=3.20.1,<4.0a0|>=3.22.0,<4.0a0|>=3.23.1,<4.0a0|>=3.24.0,<4.0a0|>=3.25.1,<4.0a0|>=3.25.2,<4.0a0|>=3.25.3,<4.0a0|>=3.26.0,<4.0a0|>=3.27.2,<4.0a0|>=3.28.0,<4.0a0|>=3.29.0,<4.0a0|>=3.30.0,<4.0a0|>=3.30.1,<4.0a0']
matplotlib -> python[version='>=3.7,<3.8.0a0'] -> sqlite[version='3.20.|>=3.20.1,<4.0a0|>=3.22.0,<4.0a0|>=3.23.1,<4.0a0|>=3.24.0,<4.0a0|>=3.25.1,<4.0a0|>=3.25.2,<4.0a0|>=3.25.3,<4.0a0|>=3.26.0,<4.0a0|>=3.27.2,<4.0a0|>=3.28.0,<4.0a0|>=3.29.0,<4.0a0|>=3.30.0,<4.0a0|>=3.30.1,<4.0a0']
scipy -> python[version='>=3.6,<3.7.0a0'] -> sqlite[version='3.13.|3.20.|3.9.|>=3.20.1,<4.0a0|>=3.22.0,<4.0a0|>=3.23.1,<4.0a0|>=3.24.0,<4.0a0|>=3.25.1,<4.0a0|>=3.25.2,<4.0a0|>=3.25.3,<4.0a0|>=3.26.0,<4.0a0|>=3.27.2,<4.0a0|>=3.28.0,<4.0a0|>=3.29.0,<4.0a0|>=3.30.0,<4.0a0|>=3.30.1,<4.0a0']
python==3.6 -> sqlite=3.13
Package python-dateutil conflicts for:
jupyter -> jupyter_console -> prompt_toolkit[version='>=1.0.0,<2.0.0'] -> python-dateutil[version='>=2.1']
matplotlib -> python-dateutil
`

regarding scale and aspect ratio in SSD

In SSD ,the author mentions usage of aspect ration and scale. Would it be possible to add some notes that clarify the role these parameters play in determining the actual bounding box

TensorFlow Object Detection Model Zoo not available anymore

Good day -

The previous link to TensorFlow object detection model zoo is not available anymore.

It seems that at the beginning of July TensorFlow released the model garden, where can be found two object detection zoos, one for TF V1 and one for V2. Links should be updated.

Thanks, and regards,

M. Russo

AttributeError: 'NoneType' object has no attribute 'shape'

AttributeError Traceback (most recent call last)
in
5 detection_classes = sess.graph.get_tensor_by_name('detection_classes:0')
6
----> 7 new_clip = clip.fl_image(pipeline)
8
9 # write to file

~/anaconda3/lib/python3.6/site-packages/moviepy/video/VideoClip.py in fl_image(self, image_func, apply_to)
512 if apply_to is None:
513 apply_to = []
--> 514 return self.fl(lambda gf, t: image_func(gf(t)), apply_to)
515
516 # --------------------------------------------------------------

~/anaconda3/lib/python3.6/site-packages/moviepy/Clip.py in fl(self, fun, apply_to, keep_duration)
135
136 #mf = copy(self.make_frame)
--> 137 newclip = self.set_make_frame(lambda t: fun(self.get_frame, t))
138
139 if not keep_duration:

in set_make_frame(self, mf)

~/anaconda3/lib/python3.6/site-packages/moviepy/decorators.py in outplace(f, clip, *a, **k)
12 """ Applies f(clip.copy(), *a, **k) and returns clip.copy()"""
13 newclip = clip.copy()
---> 14 f(newclip, *a, **k)
15 return newclip
16

~/anaconda3/lib/python3.6/site-packages/moviepy/video/VideoClip.py in set_make_frame(self, mf)
667 """
668 self.make_frame = mf
--> 669 self.size = self.get_frame(0).shape[:2][::-1]
670
671 @outplace

AttributeError: 'NoneType' object has no attribute 'shape'

The .mp4 video seems to be corrupted

It's not even playable

Recommend Projects

React

A declarative, efficient, and flexible JavaScript library for building user interfaces.
Vue.js

🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript

TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
TensorFlow

An Open Source Machine Learning Framework for Everyone
Django

The Web framework for perfectionists with deadlines.
Laravel

A PHP framework for web artisans
D3

Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

javascript

JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
web

Some thing interesting about web. New door for the world.
server

A server is a program made to process requests and deliver data to clients.
Machine learning

Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Visualization

Some thing interesting about visualization, use data art
Game

Some thing interesting about game, make everyone happy.

Recommend Org

Facebook

We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft

Open source projects and samples from Microsoft.
Google

Google ❤️ Open Source for everyone.
Alibaba

Alibaba Open Source for everyone
D3

Data-Driven Documents codes.
Tencent

China tencent open source team.