2020/02 Python 3.7 + TensorFlow 2.1 + Keras 2.3.1 + YOLOv3 Object detection with the latest version

Machine learning is a transcendental beginner who has touched php-ml for a while, but I got into image recognition.

The first goal is to detect an object from a given miscellaneous photo.

I don't understand anything

When I went through it properly, there were a lot of stories that did not work with the combination of library versions, so I do not understand very much.

So I stopped thinking and tried to put in the latest version without thinking for the time being

Most of the articles are more than half a year old, there are many ones that are one year, two years or more old, it's a trendy neighborhood, and even the latest ones are working properly (a rough idea) I'm collecting old versions one by one And that alone will break my heart

I made various trials and errors without knowing anything, but as a final form, I generally follow the article below. At first I tried ssd_keras, but it became completely unclear and I repartitioned Ta

[OK in 5 minutes] Move YOLOv3 with Keras and challenge object detection![Keras-yolov3]

anaconda

As of February 2020, the latest version is python3.7 By the way, since I'm studying, I'm doing it roughly on my Windows, but when I go around properly, there are articles that I do not understand, such as "TensorFlow does not work with python 3.6" and "I put 3.6 because it does not work with 3.7". It comes out, for the time being, there is only 3.7 in the installer dropdown, I don't know, 3.7 Irel

Create a test environment

This is no problem

Put various things

↑ Refer to the article and put it in

Data conversion

python convert.py yolov3.cfg yolov3.weights model_data/yolo.h5

I got an error Error on TensorBoard

pip uninstall h5py
pip install h5py

I still get an error

pip uninstall keras
pip install keras

I got it with this

tftest01.png

I did it

So, I'm not sure about the version, but for the time being, I put in the latest one and was able to detect the object.

By the way, the pip list is below

Package              Version
-------------------- -------------------
absl-py              0.9.0
asn1crypto           1.3.0
astor                0.8.1
blinker              1.4
cachetools           4.0.0
certifi              2019.11.28
cffi                 1.14.0
chardet              3.0.4
Click                7.0
cryptography         2.8
cycler               0.10.0
gast                 0.2.2
google-auth          1.11.2
google-auth-oauthlib 0.4.1
google-pasta         0.1.8
grpcio               1.27.2
h5py                 2.10.0
idna                 2.9
Keras                2.3.1
Keras-Applications   1.0.8
Keras-Preprocessing  1.1.0
kiwisolver           1.1.0
Markdown             3.2.1
matplotlib           3.1.3
mkl-fft              1.0.15
mkl-random           1.1.0
mkl-service          2.3.0
numpy                1.18.1
oauthlib             3.1.0
olefile              0.46
opt-einsum           3.1.0
Pillow               7.0.0
pip                  20.0.2
protobuf             3.11.4
pyasn1               0.4.8
pyasn1-modules       0.2.8
pycparser            2.19
PyJWT                1.7.1
pyOpenSSL            19.1.0
pyparsing            2.4.6
pyreadline           2.1
PySocks              1.7.1
python-dateutil      2.8.1
PyYAML               5.3
requests             2.23.0
requests-oauthlib    1.3.0
rsa                  4.0
scipy                1.4.1
setuptools           45.2.0.post20200210
six                  1.14.0
tensorboard          2.1.0
tensorflow           2.1.0
tensorflow-estimator 2.1.0
termcolor            1.1.0
tornado              6.0.3
urllib3              1.25.8
Werkzeug             1.0.0
wheel                0.34.2
win-inet-pton        1.1.0
wincertstore         0.2
wrapt                1.12.0

Recommended Posts

2020/02 Python 3.7 + TensorFlow 2.1 + Keras 2.3.1 + YOLOv3 Object detection with the latest version
[Python] Real-time object detection with iPad camera
Get the latest Linux kernel version with Arch Linux
I tried object detection with YOLO v3 (TensorFlow 2.1) on the GPU of windows!
Check version with python
Identify the name from the flower image with keras (tensorflow)
Install the latest stable Python with pyenv (both 2 and 3)
Install the latest Python from pyenv installed with homebrew
Face detection with Python + dlib
Face detection with Python + OpenCV
Call the API with python3.
Blockchain tampering detection with Python
System trade starting with Python3: Get the latest program code
Specify python version with virtualenv
Real-time object detection Android app with TensorFlow and Camera X
[For beginners] I tried using the Tensorflow Object Detection API
[Cloudian # 6] Try deleting the object stored in the bucket with Python (boto3)
[September 2020 version] Explains the procedure to use Gmail API with Python
TensorFlow 2.2 can't be installed with Python 3.8!
MNIST (DCNN) with Keras (TensorFlow backend)
Extract the xz file with python
Face detection with YOLO Face (Windows10, Python3.6)
Logo detection using TensorFlow Object Detection API
Face detection with Lambda (Python) + Rekognition
[Python] Using OpenCV with Python (Edge Detection)
Get the weather with Python requests
Get the weather with Python requests 2
pyenv-change the python version of virtualenv
How to get the Python version
Find the Levenshtein Distance with python
Change the Python version of Homebrew
[TensorFlow] [Keras] Neural network construction with Keras
Hit the Etherpad-lite API with Python
Install the Python plugin with Netbeans 8.0.2
I liked the tweet with python. ..
Manage each Python version with Homebrew
Master the type with Python [Python 3.9 compatible]
[Python Windows] pip install with Python version
I tried object detection with YOLO v3 (TensorFlow 2.0) on a windows CPU!
Challenge image classification with TensorFlow2 + Keras CNN 1 ~ Move for the time being ~
If you want to include awsebcli with CircleCI, specify the python version
Download the file by specifying the download destination with Python & Selemiun & Chrome (Windows version)
[TensorFlow / Keras] The road to assembling an RNN with your favorite structure
Make the Python console covered with UNKO
Face detection with Python + OpenCV (rotation invariant)
[Python] Set the graph range with matplotlib
Behind the flyer: Using Docker with Python
Check the existence of the file with python
About the virtual environment of python version 3.7
Compare raw TensorFlow with tf.contrib.learn and Keras
[Python] Get the variable name with str
[Python] Round up with just the operator
Display Python 3 in the browser with MAMP
Search the maze with the python A * algorithm
[Python3] Rewrite the code object of the function
Let's read the RINEX file with Python ①
Install by specifying the version with pip
Working with OpenStack using the Python SDK
Nowadays, implement DQN (complete version) with Tensorflow
Object detection using Jetson Nano (YOLOv3)-(1) Jetson Nano settings-
[Python] Try pydash of the Python version of lodash