Image Detection with YOLO-v2 (pt.8) Custom Object Detection (Train our Model!)

Share
Embed
  • Published on Jan 13, 2018
  • In this series we will explore the capabilities of YOLO for image detection in python!
    This video will look at
    - how to modify our the tiny-yolo-voc.cfg model file
    - how to modify the labels.txt file
    - how to train our model
    - how to load our trained weights to test our model

    Github example webcam code: github.com/markjay4k/YOLO-series/blob/master/part4_video.py
    Darkflow Repo: github.com/thtrieu/darkflow
    YOLO weights: pjreddie.com/darknet/yolo/
    Discord Server: discordapp.com/invite/Zf8CSB3

Comments • 714

  • Mohammad Zulqurnain
    Mohammad Zulqurnain 13 hours ago

    Hi Mark, nice tutorial. I want to train a YOLO neural network on macbook pro. Can you tell if whether I can train it via my current macbook pro 2017 which has radeon 555 or I need the older macbook with 750M geforce if I want to use GPU for faster training?

  • 楊少溥
    楊少溥 3 days ago

    Thank you three thousand

  • Daniel LaMontagne
    Daniel LaMontagne 12 days ago +2

    Hi Mark Jay

    When I type this flow --model cfg/yolov2-tiny.cfg --load bin/yolov2-tiny.weights --train --annotation new_model_data/annotations_dm --dataset new_model_data/images_dm --gpu 1.0 --epoch 300

    I get this error AssertionError: Over-read bin/yolov2-tiny.weights


    When I type this daniellamontagne$ python flow --model cfg/tiny-yolo-voc-1c.cfg --load bin/yolov2-tiny-voc.weights --train --annotation new_model_data/annotations_dm --dataset new_model_data/images_dm --gpu 1.0 --epoch 300

    I get this error AssertionError: expect 63082060 bytes, found 63471556

    I've been working on this for so long...not sure what I'm doing wrong... =(

    • Brent Labasan
      Brent Labasan 9 days ago

      I've been getting this error too. I'm starting to investigate now.

  • BadCode official
    BadCode official 15 days ago

    How do i check the accuracy of the model using this method? U haven't divided your data into tranning and test set

  • Shivam Singh
    Shivam Singh 18 days ago

    Hello,
    my loss value is not decreasing. It rather fluctuates about around 150. What should I do to make it decrease?

  • Bei Zhou
    Bei Zhou 20 days ago

    Is this custom model only able to detect fidget spinner?

  • Yanzee Nongsiej
    Yanzee Nongsiej 24 days ago

    Greetings Mark! I have a question: What is the purpose of weights file? P.S: I would like the answer to be more theory-oriented.Thank You :)

  • EK MIN WONG
    EK MIN WONG 25 days ago

    Hello Jay,

    Can I train a custom model for face recognition instead of object detection?
    Which mean I want to train my face as the training data, later on, will appear my name with the webcam, other people will display person. Thank you.

  • SATYAM SAREEN
    SATYAM SAREEN 25 days ago

    I am getting an error when I am importing darkflow:

    "from darkflow.net.build import TFnet
    ModuleNotFoundError: No module named 'darkflow' "
    I have downloaded the repository, run the setup file, I have tensorflow , so why is this error.
    Please tell

    • Atharva Choudhary
      Atharva Choudhary 24 days ago

      Did you install it globally or in a virtual environment? Also, which IDE are you using ?

  • Smallker
    Smallker Month ago

    i always got division by zero even my images path was correct

  • oornicha Sinthop
    oornicha Sinthop Month ago

    Thank you very much .
    Would you pleas tell and demo about object detection with point cloud data

  • nitish hs
    nitish hs Month ago

    is training possible in windows ?

  • ABHISHEK CHAUDHURI
    ABHISHEK CHAUDHURI 2 months ago

    Hi i am getting an empty array in the result. can someone please let me know where i am going wrong. @markjay

    • Just Cats
      Just Cats 22 days ago

      ABHISHEK CHAUDHURI did u fimd the solution?

  • apoorve kalot
    apoorve kalot 2 months ago

    hey bro, i was detecting mobile charger by yolo and had used 125 images of it, i don't know why but it always creaing box much farther away from object eant to be detected, and always include (0,0) coordinate while making box, can you suggest where i may be getting error

  • AJAY RAJ SINGH
    AJAY RAJ SINGH 2 months ago

    what is flow here i got an error that there is no such file or directory

  • Aadit Narendar
    Aadit Narendar 2 months ago

    Hi!
    I get an error while loading the weights which says, AssertionError: expect (a 9 digit value) bytes, found (another 9 digit value) bytes.
    I did edit the .cfg as per my detection and also the labels.
    This occurs when I run, after it parses the yolo.cfg file and exactly when it loads the .weights file.

    • Armin Moghadam
      Armin Moghadam Month ago

      Do not use "python" in your command line. Start with ./flow and the rest of the command

  • Archie Jack
    Archie Jack 2 months ago

    I trained on my dataset(840 images of 3 class objects, bottle/ mouse/ pen), modify model tiny-yolo-3c.cfg , batch 8, use the default lr, trainer adam, gpu model(GTX 950M 2GB), about 10k steps(about 300 epoches more than 10 hours), loss about 3 or 2, move ave loss the same, but the bounding box is not correct on train image( threshold is 0.0045), they are so many boxes or no boxes on high threshold, how to solve it?

  • duy trần
    duy trần 2 months ago

    thanks so much, but i have one question. If i trained object and want to add new class to tiny-yolo-voc-1c.cfg. How can i do???

  • Just Cats
    Just Cats 3 months ago

    im trying mine and working but to many boxes while detecting object i change the range from 10 to 1 but the box stuck on the left top conner, anyone help

    • Just Cats
      Just Cats 3 months ago

      @Ananay Gupta thank you for the reply! 😊 I'll give it a try and im running it fully on the cpu as my laptop doesn't have support gpu. I have only train 40 images. It might be the cause why

    • Ananay Gupta
      Ananay Gupta 3 months ago +1

      The range only changes the number of colours that your boxes will have, not the actual boxes themselves. If you want to change the number of boxes, then increase the threshold. If that does not work, then I don't see any reason to be worried, it means that your model can detect many objects with a great accuracy :)

  • A way to Brain
    A way to Brain 3 months ago

    sir can i get your facebook link or any other source of communication >>> i want to discuss my project with you , given to me in the university.

  • saif hamza
    saif hamza 3 months ago

    Hi Mark!

    with this model, I just want it to detect all the labels already present and including that, count the number of persons present in the frame. Can We do it? If yes, how?
    Need help, should present it coming week.

    • saif hamza
      saif hamza 3 months ago

      @Ananay Gupta thanks a Lot :)

    • saif hamza
      saif hamza 3 months ago

      @Ananay Gupta Hi!
      Thanks for your reply.
      But I think with this code, it will count all the persons it detects and keeps incrementing. I only need it to count the person present in the frame.

    • Ananay Gupta
      Ananay Gupta 3 months ago

      Best of luck for that presentation!

    • Ananay Gupta
      Ananay Gupta 3 months ago

      Yes, this is infact very simple. In the code, when you iterate through the for loop, you should check if label == 'person', and if it is, increment a counter variable. Then print this variable at the end of the while loop, and reset it. For more help, you could drop another comment

  • Muhamad Arif Lutfi
    Muhamad Arif Lutfi 3 months ago

    I have followed your instruction, and it is very clear and I rarely stuck anywhere. Except on this training part. I stuck when I test, there is no bounding box, and no results.
    I have train my model with only 2 classes. I also edited the number of classes in CFG. I also changed the filter to 35 since (5x(5+2 classes)) on the last convolutional laye, in CFG file.
    I also change the labels.txt into my own labels(Real, Fake)
    I only have 94 datasets (images, .png), and I set the epoch to 1700. The moving average lost came down from 109 to 0.007 (after 4 or 5 hours of training). But when I tried to do some test, JSON output becomes NULL. Or [ ]
    I also tried on the webcam, instead of showing FPS, I make it print out the results.
    When I use the pretrained model, I got a result of :
    [{'label': 'truck', 'confidence': 0.62592417, 'topleft': {'x': 80, 'y': 32}, 'bottomright': {'x': 998, 'y': 400}}]
    But when I use my model, it became :
    [ ]
    I also test it on my training and testing images, but still no result.
    Did I do something wrong somewhere?

    • Iryna Shevchenko
      Iryna Shevchenko 3 months ago

      @Muhamad Arif Lutfi many thx. im too still searching the solution. Of course if I'll find, I will share it

    • Muhamad Arif Lutfi
      Muhamad Arif Lutfi 3 months ago

      @Iryna Shevchenko I'm still searching for the right solution, and still haven't found yet.
      If somehow I managed to solve the problem in the future, I'll share it in this comment.

    • Iryna Shevchenko
      Iryna Shevchenko 3 months ago +1

      Hi Muhamad Arif Lutfi. I have the same problem: no bounding box. My dataset 1215 images, loss - from 115 to 5.4 (after 3 days training) and 500 epoch. I still have not solved this 😢

  • cinema and games
    cinema and games 3 months ago +1

    i get an error in cmd saying.....'NoneType' object has no attribute 'shape'

    • Ananay Gupta
      Ananay Gupta 3 months ago

      It seems to me that somewhere a part of you code has a variable null. Can you give any more details?

  • Ben M
    Ben M 3 months ago

    Hi Mark Jay, great tutorials, they have helped me a lot! I am interested in performing transfer learning on just the last couple of layers of the CNN. I've tried messing around in the .cfg file but I can't seem to find how to turn training from "Yep!" to "No" in the model.

    Do you know how to do this?

    Thanks in advance, Ben

  • bob BUILDER
    bob BUILDER 3 months ago

    hi mark so i want to train my dataset i have downloaded the weight but there are some errors which u may know how to solve

    C:\Users\eee\Desktop\darkflow-master>python flow --model cfg/yolov2-tiny.cfg --load bin/yolov2-tiny.weights --train --annotation new_model_data/annotations --dataset new_model_data/images --gpu 0.7 --epoch 300

    Parsing ./cfg/yolov2-tiny.cfg
    Parsing cfg/yolov2-tiny.cfg
    Loading bin/yolov2-tiny.weights ...
    Traceback (most recent call last):
    File "flow", line 6, in
    cliHandler(sys.argv)
    File "C:\Users\eee\Desktop\darkflow-master\darkflow\cli.py", line 26, in cliHandler
    tfnet = TFNet(FLAGS)
    File "C:\Users\eee\Desktop\darkflow-master\darkflow\net\build.py", line 58, in __init__
    darknet = Darknet(FLAGS)
    File "C:\Users\eee\Desktop\darkflow-master\darkflow\dark\darknet.py", line 27, in __init__
    self.load_weights()
    File "C:\Users\eee\Desktop\darkflow-master\darkflow\dark\darknet.py", line 82, in load_weights
    wgts_loader = loader.create_loader(*args)
    File "C:\Users\eee\Desktop\darkflow-master\darkflow\utils\loader.py", line 105, in create_loader
    return load_type(path, cfg)
    File "C:\Users\eee\Desktop\darkflow-master\darkflow\utils\loader.py", line 19, in __init__
    self.load(*args)
    File "C:\Users\eee\Desktop\darkflow-master\darkflow\utils\loader.py", line 77, in load
    walker.offset, walker.size)
    AssertionError: expect 44138056 bytes, found 44948600

    • bob BUILDER
      bob BUILDER 3 months ago

      i read online and it stated it might be due to the mismatching cfg and weights files. Can you please advise how can I fix this or get the correct files? thanks :)

  • Siddharth Joshi
    Siddharth Joshi 3 months ago

    Hey Mark Awesome Tutorial. So i trained the model with loss around 3 (It wasnt changing much). Then tried to test the model . I am getting multiple annotations all over the image with the label that i have trained for. Any Advice ? What might have went wrong. (I used Yolo not tiny yolo .) and for trainer i used adam optimizer.

    • Siddharth Joshi
      Siddharth Joshi 3 months ago

      an update. Allowed the code to run for more epochs even when it was hovering at 3. and guess what it worked.

  • Tutor alive
    Tutor alive 3 months ago

    What GPU have you used?

  • Fieuws
    Fieuws 4 months ago

    I have multiple predictions on the same object. Anyway I can drop all these? (putting a higher threshold doesn't help as my bad model doesn't have much detection with bigger percentages)

  • Anoop S
    Anoop S 4 months ago

    thank you so much ❤️
    how to have our weight detect only our trained object and not every objects it was previously detecting. like for example only fidget spinner, not other objects the weight was previously detecting.

  • Matheus Melo
    Matheus Melo 4 months ago

    I've been requested by a professor to implement a project similar to your Yolo tutorial( great btw ), but using C++, and, for that, I am having trouble getting instructios and info on it. Is there a way you could help me out?

  • orange222000
    orange222000 4 months ago

    Thank you for your tutorial
    In training i have big problem i have CUDA8.0/cuDNN5.1/tensorflow1.4.0/ gtx1060/

    2019-04-05 13:04:57.116335: E C:\tf_jenkins\home\workspace\rel-win\M\windows-gpu\PY\35\tensorflow\stream_executor\cuda\cuda_dnn.cc:385]
    could not create cudnn handle: CUDNN_STATUS_NOT_INITIALIZED

    i can't find the solution So, i'm looking forward your advice Thank you

    • 크림Kream
      크림Kream 4 months ago

      for me i have tensorflow 1.13.1, cuda 10.0 cudnn7.3.1 running them on anaconda (make sure u install tensorflow on anaconda using conda install tensorflow-gpu because it will pull a few more dependencies that made it work).

  • Nishad Islam
    Nishad Islam 5 months ago

    UserWarning: ./cfg/yolov2-tiny.cfg not found, use cfg/tiny-yolo-voc-1c.cfg instead
    'Over-read {}'.format(self.path)
    AssertionError: Over-read bin/yolov2-tiny.weights I found this error. How to solve this

  • Jason Chiang
    Jason Chiang 5 months ago

    Hi I am just wondering if you could do similar things here but with yolov3?

  • Don Fung
    Don Fung 5 months ago +1

    Hey Mark I'm curious to know if you know the theory/reason behind why the bounding box jitters when the object in the video is static? 15:50-16:00

  • Omkar Naik
    Omkar Naik 5 months ago

    Can I do this with CPU

    • 크림Kream
      크림Kream 4 months ago

      yes just don't add the -gpu part

  • Dhinza Aizawa
    Dhinza Aizawa 5 months ago

    Hi Mark, How to show yolo object detection output on browser not from from pop-up window? I want to make it can be accessed from different IP.. I thought about using flask but I have no idea to do that.. really need guidance.. thanks

  • D Dopfer
    D Dopfer 5 months ago

    Hi Mark, great tutorials. Can I or how can I transfer your codes to run on Linux or Mac? Thanks

  • Anil kumar
    Anil kumar 5 months ago

    Hi Mark, What if the Image size is big? like 1500*1300. Can we annotate on those ? or Should we resize them first and then annotate? If I resize I'm losing information, what is the right way to do? Thanks.

  • Parichay Didwania
    Parichay Didwania 6 months ago

    bro i am getting OSError: File ./ckpt/tiny-yolo-voc-10c-750.meta does not exist. Even though it exists there, is the file assuming different root folder even though its in main directory?

  • Iryna Shevchenko
    Iryna Shevchenko 6 months ago

    Hi Mark. Thank you very much for such awesome tutorials!!!! maybe you can help me. I trained 1 class but even after 37000 steps, i can't detect any boxes. Loss is about 4.6e-05. Dataset - 1217 images, epoch - 500. where is my mistake??? help me, please :)

  • Naveen Kumar
    Naveen Kumar 6 months ago

    hello mark I have got some error when i run the ,
    sre_constants.error: bad character range
    please help me in solving this issue

  • Samkit Shah
    Samkit Shah 6 months ago

    I have trained modified model tiny-yolo-voc.1c for 1 class but it is not detecting for the object taken which is "bottle".

  • Rizal Hilman
    Rizal Hilman 6 months ago

    So, now the cfg/tiny-yolo-voc-1c.cfg is just contained our custom object?

  • Samir Ahmed
    Samir Ahmed 6 months ago

    Hi Mark, I'm getting the error :AssertionError: Over-read bin\yolov2-tiny.weights" when trying to train with bin\yolov2-tiny.weights and "AssertionError: expect 63082056 bytes, found 180357512" when I try to train with bin/yolo-tiny.weights, I searched in google and found suggestions of downloading google's weights and that didn't work as well.. BTW I'm using windows 10

  • KoraLab
    KoraLab 6 months ago

    Implementing Yolo-V3 in pubg
    ru-clip.net/video/hIV3EJVeiA4/video.html

  • Adi _14
    Adi _14 6 months ago

    Hi Mark,I am getting this error:
    AssertionError: expect 44948596 bytes, found 44948600
    How can I resolve this?

    • Daniel LaMontagne
      Daniel LaMontagne 12 days ago

      Im having the same problem! AssertionError: expect 63082060 bytes, found 63471556

  • Vaibhav Warade
    Vaibhav Warade 7 months ago

    How to give images and get detected images as an output instead of giving webcam input..

  • Vaibhav Warade
    Vaibhav Warade 7 months ago

    ​ Mark Jay I am getting errors while training on tiny yolo v2: Assertion error, As the site which you have suggested has been changed and only contains models trained on COCO any solution for that?
    or can you please give me link to download "weights" file and "cfg" file which you used in this video?

  • Niaz Ahmad
    Niaz Ahmad 7 months ago

    Stuck in the this error plz help any one.
    InvalidArgumentError (see above for traceback): Input to reshape is a tensor with 132496 values, but the requested shape requires a multiple of 7605

    • ankit sharma
      ankit sharma Month ago

      Is your error resolved? I am getting the same error

  • 홍운달
    홍운달 7 months ago

    Hi Mark.Thank you for you tutorial,it's really help.There is a question for me:
    When I trained my model with CPU it worked fine
    But When I tried to train with my GPU (NVIDIA GTX 1060 6GB),the command line showed like this:
    "loss nan - moving ave loss nan" from step 1.
    and I just reduced the learning rate and increased the batch size,but it was not work.
    I really want to know how to fix this : )

    • 홍운달
      홍운달 7 months ago

      and when I tried on TITAN V ,also showed that "loss nan - moving ave loss nan"

  • Lavanya Gupta
    Lavanya Gupta 7 months ago

    Hi Mark, Thank you so much for this wonderful tutorial video lecture series. Just had a quick question. Do all image sizes need to be 608x608 when using Yolov2? How does the model's first layer adjust to different input image sizes?

  • Stefano Di Luzio
    Stefano Di Luzio 7 months ago

    Yea, but how to show some graphs about loss and accuracy with tensorboard? I think this is the most useful thing to know after the things getting trained, isn't it?

  • tahsin siad
    tahsin siad 7 months ago

    Sir, I want to add a new class to the existing pre-trained model. How can I do that?

  • Yunda Hong
    Yunda Hong 7 months ago

    Thank you Mark ,it works perfectly and it's really help

  • Divyesh patel
    Divyesh patel 7 months ago

    Getting error when i'm trying to execute :
    python flow --model cfg/tiny-yolo-voc-1c.cfg --load bin/tiny-yolo-voc.weights --train --annotation new_model_data/annotations --dataset new_model_data/Images --epoch 200

    Traceback (most recent call last):
    File "flow", line 4, in
    from darkflow.cli import cliHandler
    File "YOLO-series-master\darkflow-master\darkflow\cli.py", line 3, in
    from .net.build import TFNet
    File "YOLO-series-master\darkflow-master\darkflow\net\build.py", line 5, in
    from .ops import op_create, identity
    File "YOLO-series-master\darkflow-master\darkflow\net\ops\__init__.py", line 1,
    in
    from .simple import *
    File "YOLO-series-master\darkflow-master\darkflow\net\ops\simple.py", line 1, in

    import tensorflow.contrib.slim as slim
    File "C:\ProgramData\Anaconda3\lib\site-packages\tensorflow\contrib\__init__.py", line 31, in
    from tensorflow.contrib import coder
    File "C:\ProgramData\Anaconda3\lib\site-packages\tensorflow\contrib\coder\__init__.py", line 22, in
    from tensorflow.contrib.coder.python.layers.entropybottleneck import *
    File "C:\ProgramData\Anaconda3\lib\site-packages\tensorflow\contrib\coder\python\layers\entropybottleneck.py", line 24
    , in
    from tensorflow.contrib.coder.python.ops import coder_ops
    File "C:\ProgramData\Anaconda3\lib\site-packages\tensorflow\contrib\coder\python\ops\coder_ops.py", line 30, in
    resource_loader.get_path_to_datafile("_coder_ops.so"))
    File "C:\ProgramData\Anaconda3\lib\site-packages\tensorflow\contrib\util\loader.py", line 56, in load_op_library
    ret = load_library.load_op_library(path)
    File "C:\ProgramData\Anaconda3\lib\site-packages\tensorflow\python\framework\load_library.py", line 56, in load_op_lib
    rary
    lib_handle = py_tf.TF_LoadLibrary(library_filename)
    tensorflow.python.framework.errors_impl.NotFoundError: C:\ProgramData\Anaconda3\lib\site-packages\tensorflow\contrib\cod
    er\python\ops\_coder_ops.so not found

    • Ananay Gupta
      Ananay Gupta 3 months ago

      You seem to be missing some files, and from what it looks like, its from tensorflow. Why not reinstall it?

  • Buzzz
    Buzzz 7 months ago

    I got an error when it comes to train.
    AttributeError: type object 'h5py.h5r.Reference' has no attribute '__reduce_cython__'

  • Nakul Joshi
    Nakul Joshi 7 months ago

    i am getting an invalid syntax error in line 2 of darkflow-master/darkflow/cython_utils/cy_yolo_findboxes.py
    cimport numpy as np

    • Ananay Gupta
      Ananay Gupta 3 months ago

      I think this is because you have not installed numpy. Try running 'pip install numpy' in cmd/terminal, and then do it again

  • Mohammed Al-Alaw
    Mohammed Al-Alaw 7 months ago

    I would train it on smething much more useful like cars and people!

  • Nakul Joshi
    Nakul Joshi 7 months ago

    Hey,
    I wanted to ask that after training yolo with fidget spinner ,will it detect only fidget spinner or person,cell phone and other stuff as well!

  • Aashish Pal
    Aashish Pal 7 months ago

    hello sir, can we load both default yolo weights and self trained weights at the same time ? and if it is possible then how !!

  • jayesh karvir
    jayesh karvir 7 months ago

    Hey Mark is it possible to train this model for classifying multiple small objects like rice grains in an image?

  • Kasun Sandaras
    Kasun Sandaras 7 months ago

    hi mark this is a really awesome video series. I could do it and it detects customized objects.Further can we take the values of the pose,difficulty which we annotate in the xml file when the time we are detecting the object from the video.

  • Ahmet Kalay
    Ahmet Kalay 7 months ago

    Hey mark! You helped me a lot with this series thank you. Can I ask are these weight files under public domain or not? Can we use them commercially ?

  • Subash Kanumuri
    Subash Kanumuri 8 months ago

    ' raise source.error(msg, len(this) + 1 + len(that))
    sre_constants.error: bad character range w-i at position 84' how do i solve this?

  • S 1th
    S 1th 8 months ago

    Can you share your .cfg file and .weights ? it will be too usefull

  • S 1th
    S 1th 8 months ago

    AssertionError: Over-read yolov2-tiny.weights
    i m getting this error! i cant move forward please help me out

  • Noman Yousuf
    Noman Yousuf 8 months ago +1

    how to solve this issue:
    from ...cython_utils.cy_yolo_findboxes import yolo_box_constructor
    ModuleNotFoundError: No module named 'darkflow.cython_utils.cy_yolo_findboxes'

    • Ananay Gupta
      Ananay Gupta 3 months ago

      Were any more details given. This could probably be due to the build folder not being built properly, or due to the fact that the files are not in the right directory

    • Berkay Öntürk
      Berkay Öntürk 3 months ago

      did you fix that?

  • Pekanbaru Animation
    Pekanbaru Animation 8 months ago

    Hello mark,
    i wanna ask a noob question and maybe for anyone that know are welcoming to give the answer, are the confidence score is just the same as the accuracy value?
    thanks for the answer..

  • techgamer
    techgamer 8 months ago

    nice

  • Neil Attewell
    Neil Attewell 8 months ago

    For training a model on something like a fidget spinner, why not record a video of the spinner (or multiple spinners) (where its in a fixed coordinate), for each frame, generate the xml + save the frame as an image to disk. In the video touch it, spin it...etc but don't move it. Then train your model on the thousands of images generated for the spinner/s...
    You could probably be smart and on a plain/white background track what the standard model detects for the spinner (ie: everything other than your hand/person) and use that for the fidget spinners location, then you can move it around and change its orientation/rotation...
    Good idea/Bad idea??? I'm going to give it a try

    • Ananay Gupta
      Ananay Gupta 3 months ago

      It may train the model for different orientations when spinning, but you should not leave it at that. Use different colours, use different orientations(as in, use ones where the camera angle is also shifted), and use the spinner against different backgrounds. The models should also have images of where the spinner is in someone's hands. By giving it may similar images may not necessarily train it for real- life situations

  • Sarfraz Malik
    Sarfraz Malik 8 months ago

    Hi,
    Can you tell me please what should be the possible loss when we train yolo model,
    I am training a model on 2300 images but i don't know at what stage i should stop my training,

    • Ananay Gupta
      Ananay Gupta 3 months ago

      Ideally, you should it depending on the number of classes. If you have only 1, then I recommend a 0.5 or less. Otherwise, for a small number, 1 or less, and for a large number, around 2

  • Sun Shine Kr
    Sun Shine Kr 8 months ago

    Thank you very so much.
    I trained model to detection number plate and its working fine.
    Hope you make more awesome tutorials!

  • Jongkuk Park
    Jongkuk Park 8 months ago

    Please anyone here,
    Is this not possible to train the model yolo-1c.cfg with yolov2.weights using 1070-GPU? Beginning on the video, Mark says there should be at least over 1080-ti to use yolov2. Is this true? If so, Can anyone tell me why it is? I've thought we can still train it whatever GPU we use, it would take just longer.

  • Pankaj Joshi
    Pankaj Joshi 9 months ago

    Hi sir, thanks for nice tutorial with nice presentation, By using this I am able to train my own model, but my model didn't detect anything even with threshold of 0.01 when I change threshold lower than this then I get lots of false detection. Even on same image on which i train i dont get any thing I train Network on one image for one class .
    Do i have to increase number of image datasets?

  • Abel worku
    Abel worku 9 months ago

    Hi Mark Jay, thank you for wonderful tutorial, i have followed your tutorial series from 1 to this and it helped me a lot. but i do have two questions on how to apply the non max suppression on the detected objects so that i can remove multiple detections. i was trying by using the four coordinates of the box, add to box list and apply nms, but i do have problem on this, and the error of training is around 2. and 3. how can i finetune the hyperparameters so that i can achieve the lowest error possible? thank you again for your kindness!