YOLOv3 in the CLOUD : Install and Train Custom Object Detector (FREE GPU)

[@TheAIGuy]

Let me know what classes you are making your custom detector detect! Also comment if you have any issues!

[@mubinmodi7529]

thanks man great video . But I have issues in installing the gpu version in my laptop . While using cmake configuration, cmake won't detect my cuda drivers . I have installed cuda 10.2 Also i didn't install cudnn

[@TheAIGuy]

@@mubinmodi7529 for this video you don't need cmake and cuda will come predownloaded on the cloud VM. Did you mean to comment on one of my different videos?

[@sahiluppal3093]

it's saving weights after every 100 epoch. how can we override this default behavior and save weights after let's say 1000 epochs.??

[@manutyagi8019]

Good afternon sir, I have one more suggestion kindly make a video on how to install and use detectron

[@TheAIGuy]

Manu Tyagi awesome will do!

[@jyothsnap3138]

Your in depth explanation solved so many problems! Thanks AI Guy! I wish this video had more views!

[@TheAIGuy]

Appreciate the kind words and glad you liked the video! Thanks so much!

[@Dheeraj-ed6rr]

Exactly

[@tejasnareshsuvarna7948]

Thank you so much for this beautiful written Google Colab notebook. It is very helpful. This channel should get a lot of subscribers. Keep it up :)

[@TheAIGuy]

Thanks so much!

[@zakariaabderrahmanesadelao3048]

you're a fucking legend. I spent a week compiling darknet on my laptop, built my own labelmaker with python and when i got to train my model on cpu it estimated 1600 hours. nearly had a mental breakdown. my first reflex was to look into AWS EC2. spent a day with that headache. then youtube suggests your video and god man you have saved my life. subbed and forever your follower. thank you a thousand

[@mlnjsh1]

This kinds of abusive words if you praise someone are not good....praise in good words

[@aaditkolar5521]

10:57 was so creative that you earned my subscription

[@TheAIGuy]

Appreciate it. Thanks a lot! :)

[@aravindabilash151]

I followed all your tutorial on YOLOv3 and within 2 days i was able to setup YOLOv3, detect object with pre-trained model, create datasets, train them and i was able to detect custom object. Thank you for you tutorials.

[@leafteaneko]

For all those encountering the following error in STEP 3 line 44 " Traceback (most recent call last): File "generate_train.py", line 4, in os.chdir(os.path.join("data", "obj")) FileNotFoundError: [Errno 2] No such file or directory: 'data/obj' " go to the generate_train.py and change the line4 from os.chdir(os.path.join("data", "obj")) to os.chdir("data") and image_files.append("data/obj/" + filename) to image_files.append("data/" + filename) . It seems the images are saved in a data folder directly instead of data/obj even when you follow the steps correctly

[@N-e0N]

I'm getting this error even though my images are inside darknet/data/obj

[@leafteaneko]

@@N-e0N Post your errors here and maybe we can help

[@N-e0N]

@@leafteaneko Its okay, I tinkered around and made do. For some reason, the code really misbehaves at times for me. I added "darkflow" to os.path.join and also made it darknet/data/obj+filename in the image_files.append. That worked for a bit but then I got an error in the training phase when it said "Can't load images from darknet/data/obj" . At this point I went back and changed everything to original and loaded it...then it worked fine!

[@N-e0N]

@@leafteaneko Ok I'll take you up on your offer because the minor tweaking no longer works. Here's the error message : Traceback (most recent call last): File "generate_train.py", line 4, in os.chdir(os.path.join("data","obj")) FileNotFoundError: [Errno 2] No such file or directory: 'data/obj' But there is a data folder inside darknet, and it does contain obj inside of which are all the images and annotations.

[@leafteaneko]

@@N-e0N that is really weird, the same thing happened to me once before but I did a factory reset runtime and started fresh and it worked. I know it works because I successfully trained my model and I have the config files too. Drop me a email so I can send over my code, maybe that'll help your case(?)

[@kshitijnair7758]

Hey, great tutorial dude! I spent hours figuring all of this out and got it running back in Nov 2019, this video is a lifesaver for anyone who's new to YOLO! Appreciate the efforts man, cheers!

[@woosal-kc7tr]

Done everything from scratch with the help of you, really informative and god tier tutorial :D. Thanks!

[@thatoyaonebogopa9483]

god tier indeed.

[@peregrinussolutionsllc6010]

Wow! I have my fingers crossed, but so far this is much easier than running YOLOv3 with Darknet and OpenCV on Windows 10 for me :-) Thank you for the great tutorials!

[@TheAIGuy]

Peregrinus Solutions LLC Thanks for the support! Glad you enjoy.

[@arunselvabio]

Thank you for your hard work to frame format the content beautifully and to make it work :)

[@argoz1636]

Thanks a lot for the video, I was having a lot of trouble installing DarkNet and cuda in my computer and all that stuff, so this was really usefull. Great video, dude, keep it up!

[@TheAIGuy]

Glad it worked for you! Thanks!

[@satyapratik]

Traceback (most recent call last): File "generate_train.py", line 4, in os.chdir(os.path.join("data", "obj")) FileNotFoundError: [Errno 2] No such file or directory: 'data/obj' I have also done %cd content darknet as just %cd darknet gives no directory named darknet. I have obj.names and obj.data in my data folder. I have tried everything. Can someone help me out? Do I need to change something in the code?

[@이진우]

I have a same issue here

[@isaactnt]

I found the problem, you have uploaded the obj.zip and extracted it outside darknet/data and in another folder named data. Make sure that you extract it in darknet/data otherwise it will not work

[@kalleheinz758]

# unzip the zip file and its contents should now be in /darknet/data/obj !unzip ../obj.zip d darknet/data/obj

[@argishtib]

Hey man, love this tutorial on the cloud. I struggled so much getting it work on the local GPU, so glad to see a much better way!

[@TheAIGuy]

Glad you enjoyed it! I really appreciate the kind words.

[@argishtib]

@@TheAIGuy Dude, you're killing it with great content. It's truly a great time to be alive and seeing how you can custom label objects, build a detector, and run it in real time. All for free! Would love to know you're background in tech and aspirations besides growing the channel, and openness to possible collaboration down the line.

[@TheAIGuy]

@@argishtib Feel free to reach out to me at my email as I'd love to chat more. You can find it in my About tab of this channel.

[@argishtib]

@@TheAIGuy Sorry just saw this now. Just sent you an email.

[@makeoutloud]

Bro make video on predicting yolov3 on video and exporting the video in windows

[@zakariaabderrahmanesadelao3048]

my task is to detect a paper being help up. it's a bit tricky because most of the test pictures have a white wall background resembling the paper but the network is doing a pretty good job so far, thank you

[@hamzalatif115]

I followed every steps of your tutorial but at the when i start custom training it gives me the error "/bin/bash: ./darknet: Is a directory" .Please help me in this

[@keshavkantsharma1946]

Were u able to solve this?

[@andrekestler8256]

I have the same issue

[@hamzalatif115]

some how after trying many solution i get one go to darknet directory by "cd darknet" and after that run this command "!chmod +x darknet" this will give you permissions to execute the darknet files after that run your custom detector hopefully it works

[@iantairea859]

I'm having same problem. "!chmod +x darknet" doesn't solve for me. AI Guy, can you help us?

[@iantairea859]

hey guys i solved this problem. at some point i was in the wrong directory so my data/obj was is /content/ instead of content/darknet/data/obj. my yolov3_custom.cfg was also not appearing in the correct location. i ended up factory reseting the runtime (in the runtime menu) and then started again taking more care to make sure everything was popping up in the corrent folders and checking up current directory with %pwd if i wasnt sure. AI Guy thanks heaps for these tutorials. keep them coming!

[@hansgabrieldaduya2565]

I always wanted to do deep learning but the resources on the internet were really difficult to understand. Thank you since you really explained it well!

[@TheAIGuy]

Appreciate it! Thanks for the support.

[@akhilgkrishnan4469]

After running the JavaScript code in console it's showing error...

[@TheAIGuy]

Akhil G Krishnan whats the error

[@GKMUSICINDIAVIDEOS]

@@TheAIGuy Uncaught TypeError: Cannot read property 'click' of null at ClickConnect (:3:55) Uncaught error: ErrorEvent {isTrusted: true, message: "Script error.", filename: "", lineno: 0, colno: 0, …}

[@akhilgkrishnan4469]

@@TheAIGuy Uncaught TypeError: Cannot read property 'click' of null at ClickConnect (:3:55) Uncaught error: ErrorEvent {isTrusted: true, message: "Script error.", filename: "", lineno: 0, colno: 0, …}

[@akhilgkrishnan4469]

@@TheAIGuy Uncaught TypeError: Cannot read property 'click' of null at ClickConnect (:3:55) Uncaught error: ErrorEvent {isTrusted: true, message: "Script error.", filename: "", lineno: 0, colno: 0, …}

[@TheAIGuy]

GK MUSIC INDIA try changing the “colab-toolbar-button#connect” to “paper-icon-button” inside the queryselector

[@satyabratsharma486]

i am getting this error...what should i do?? File "generate_train.py", line 4, in os.chdir(os.path.join("data", "obj")) FileNotFoundError: [Errno 2] No such file or directory: 'data/obj'

[@ravirajsinhdabhi9255]

when i run this cell # train your custom detector !./darknet detector train data/obj.data cfg/yolov3_custom.cfg darknet53.conv.74 -dont_show i am getting this type of error /bin/bash: ./darknet: Is a directory please help

[@TheAIGuy]

Ravirajsinh Dabhi you need to run “%cd darknet” command before, you must be in the root directory and you need to be in the darknet folder to run the detection.

[@ravirajsinhdabhi9255]

@@TheAIGuy sorry for to trouble you after i am getting Couldn't open file: data/obj.data

[@ravirajsinhdabhi9255]

@The AI Guy ok it's solved thank you for this lovely content.If i want to contribute something and support this content for you what can i do?where can i do?

[@TheAIGuy]

Ravirajsinh Dabhi just liking this video and subscribing to the channel means a lot to me. Thanks!

[@ravirajsinhdabhi9255]

@The AI Guy this is my dataset for custom model,this is whole dataset collected by me and my friend from casting foundry www.kaggle.com/ravirajsinh45/real-life-industrial-dataset-of-casting-product This is we doing for identify defective product and ok product!! For reducing tedious process of manual inspection of all product. We already made binary classifier from this dataset now we working for detection of it. We already made working model of that with help of arduino and stepper motor which can push defective from inspection line.

[@rob26r]

Hi Mr AI Guy. Just wanted to say a million thanks for your help working through issues I had and for creating such a great tutorial. I have my object detection via jpeg and now I can move on to getting object detection working on a live webcam. Like and subscribed. Thanks again.

[@TheAIGuy]

Glad you got it all working!

[@alokranjansanga6759]

I was hoping you will say "welcome back aliens"

[@magelauditore333]

Thats reserved for telusko

[@justinpatel8751]

I am making my custom detector on particular traffic sign classes such as stop sign, right-hand side curve etc. As of now to check that how does it work out for only 4 classes, if it works well i will try to train it for more signs around 40 no of classes😅 I hope that won't be an issue in case of that much classes. This tutorial helped a lot. Thanks.🙌

[@ismail_..]

Did you do that does this work

[@snehitvaddi]

May I know how download trained model and implement that model on realtime camera and further image processing stuff...Any reference will be appreciated 👏👏

[@TheAIGuy]

check out the readme for my "Object Detection API" video or my "How to Build an Object Detection Classifier" video, both will do the trick

[@johanbekker2]

This is awesome, thanks a lot! Couldn't get darknet to work on my own PC (cuda and opencv troubles) for 2 days, and this solved my suffering. Took a bit of puzzling to get it to work for Yolov4, but totally worth it.

[@hristinabiserinska2407]

The video was really helpful! Thank you so much for your effort! Could you please give me an idea how to apply the model on multiple test images (I have test.txt file analogue to the train.txt file, containing the path to the image) and create a csv file (or pandas dataframe) with the bounding box coordinates, the center (x,y) and the class probability?

[@TheAIGuy]

You can check out the readme for alexeyab/darknet github repository, he has amazing documentation and shows how to test on multiple images at once!

[@hristinabiserinska2407]

@@TheAIGuy I tried multiple things that alexeyab suggests. The following one is running without an error !./darknet detector test data/obj.data cfg/yolo-obj.cfg yolo-obj_final.weights -dont_show -ext_output /mydrive/yolo/images/imgname.jpg result.txt but the result.txt is not saved in darknet repository and I cannot find it. Did you had the same issue?

[@TheAIGuy]

@@hristinabiserinska2407 That will mostly likely save the results.txt into the colab VM. Try making it /mydrive/yolo/results.txt or something similar!

[@arslansiddique32]

You are the legend. I used to hate colab because of disconnection with internet. Thanks for your hack and all other tutorial. Liked and Subscribed !

[@TheAIGuy]

Thanks so much! Glad you found it useful.

[@channaperera2978]

We need video about yolo detection in webcam realtime feed.. Thanks in advance :)

[@TheAIGuy]

Will do!

[@iamayeshanoor]

@@TheAIGuy yeah i really need to do this on webcam realtime

[@TheAIGuy]

@@iamayeshanoor check out my latest video. Its all about running yolov3 on webcam or video in real-time! It will walk you through how to do it with GPU or CPU.

[@TheAIGuy]

@abdessamad makdad yes I will try and figure that out! In the meantime you can train your model in the cloud and then use my latest video to see how to run it on video and webcam!

[@TheAIGuy]

@abdessamad makdad you can still run video detections on CPU! You wont get as many frames per second but it still works!

[@nept4ne]

Hi, I just discovered your channel and loved the content about computer vision. I'll study your videos and many thanks for sharing.

[@TheAIGuy]

Harry glad you enjoy! Thanks for the kind words. All the best

[@dannyboynoynay6170]

Thank you so much! you saved my life! hehhehehe

[@gabrielegranello3239]

Awesome job!! I was struggling so much to just install TensorFlow with GPU on my local machine, and this solves everything! Thank you so much, keep iy up!

[@구라니은]

when i want to test with a video, instead of this cord: !./darknet detector test data/obj.data cfg/yolov3_custom.cfg /mydrive/yolov3/backup/yolov3_custom_last.weights /mydrive/images/safari.jpg -thresh 0.3 which command do i have to use?? and do i need another weight file download??

[@jadalee4278]

wondering this too

[@sharlenelee-jendili8408]

me too!

[@jayaharshang5454]

Yes, please help.

[@faqihhusaen8186]

I try to train my dataset but i always get this problem Saving weights to /backup//yolov3_custom_last.weights Couldn't open file: /backup//yolov3_custom_last.weights How can i fix this problem?

[@joancvcd]

I had the same problem. Now it's working for me. My mistake was to write the animals class name in lowercase in obj.names file.

[@gaurangpatel9968]

I have one question here. How can we use preprinted model and add our newly trained model with that. For example i have 80 classes in coco names but now i trained model with 2 new classes and i want that 80 and also 2 new classes. So in result my model is able to detect total 82 object. Please Help me for this...

[@saishriram2064]

For those facing issues with the output causing crashes - use the following code in the console: function ClearOutput(){ console.log("Cleared Output"); document.querySelector("iron-icon[command = 'clear-focused-or-selected-outputs']").click() } setInterval(ClearOutput,60000) The updated version for the ClickConnect function is: function ClickConnect(){ console.log("Clicked on connect button"); document.querySelector("colab-connect-button").click() } setInterval(ClickConnect,300000)

[@jayshah9697]

One of the best channel for computer vision beginners!!!!

[@arindammondal9364]

nice video. please make related projects like counting custom objects, vehicle speed predictions etc. your way of teaching and code explanation is absolutely understandable. that is why i'm asking for. thank you.

[@student-op1jt]

i am training for hand detection. My dataset is folder obj. which has all the images and texts. But during training it is showing Wrong annotation: x = 0, y = 0, < 0 or > 1" Is it because there is no class.txt file in the obj folder.

[@sourdurian2839]

you da man! i literally struggled to do this but your tutorial helped me a shit lot. Liked and subbed!

[@hans317]

Hello bro. I got into final step and tried to use my trained module. but it said like CUDA-version: 10010 (10010), cuDNN: 7.6.5, GPU count: 1 OpenCV version: 3.2.0 names: Using default 'data/names.list' Couldn't open file: data/names.list What Should I in this case? I clearly declared the path of the file

[@reyden008]

help!!! i get the error CUDA-version: 11000 (11020), cuDNN: 7.6.5, GPU count: 1 OpenCV version: 3.2.0 names: Using default 'data/names.list' Couldn't open file: data/names.list when i try to run the last code run your custom detector with this command (upload an image to your google drive to test, thresh flag sets accuracy that detection must be in order to show it) !./darknet detector test data/obj.data cfg/yolov3_custom.cfg /mydrive/yolov3/backup/yolov3_custom_last.weights /mydrive/images/safari.jpg -thresh 0.3 imShow('predictions.jpg') }

[@gsshsvvdjdjd4236]

Saving weights to /mydrive/yolo3/backup//yolov3_custom_last.weights Couldn't open file: /mydrive/yolo3/backup//yolov3_custom_last.weights It is not saving and not working... Any help?

[@TheAIGuy]

make sure your path is proper in obj.data file for backup, try removing the / after backup at the end of the line and see if that works!

[@josedavidmonterourena3378]

What can I say? You're the best! All your videos are freaking awesome!

[@amarnathg4537]

I have an doubt sir! Is it possible to yolo pretrained weight to detect only specific object! For example yolo tiny weight consist of 80 class but I want the detector to detect only bicycle car motor bike ! Not to detect others... If it is possible please tell me how...!

[@TheAIGuy]

Yes this is certainly possible. I would recommend using the code repository from my most recent video called "Realtime Yolov3 Object Detection..." You will want to customize the detect.py or detect_video.py file (depends if you want to run on images or video). You will just want to add a line or two somewhere saying something along the lines of "if class == car, motorbike..." then run detections.

[@vandanashekhar4474]

hey I needed a help when I am trying to run this cell " python generate_train.py" it gives me an error: Traceback (most recent call last): File "generate_train.py", line 4, in os.chdir(os.path.join("data", "obj")) NotADirectoryError: [Errno 20] Not a directory: 'data/obj' Can you please help me out with this?

[@aymanahmed9631]

Hello, is it possible to use 360° videos and photos or panoramic images as an input-dataset for Yolo? would the object detection be distorted?

[@ilse2040]

I am following your toturials for days now and find them extremely clear, thank you for making rgeat content! Question: When running a custom training model for one class on my own laptop, for a batch of 4000 it will take 67 hours. How much time do you reckon it will take on this cloud VM? Might give it a try this weekend, thanks!

[@michaelroditis1952]

You sir are one amazing man. This video deserves more views.

[@이한규-r5t]

Hi sir! How can i test multiple images? I searched for this on google. But it didn’t work. The error is : cannot load image:a.jpg It works only one file.. Help me ..

[@ash24042]

Hi, sorry I got a problem when I try to run the generate_train.py Got an error say NotADirectoryError: [Errno 20] Not a directory: 'data/obj/'. It was line 4 that gives this error. Edit: solve it by %cd content/darknet

[@noirRozelle]

where did you put %cd content/darknet?

[@manojrajsr6001]

Traceback (most recent call last): File "generate_train.py", line 4, in os.chdir(os.path.join("data", "obj")) FileNotFoundError: [Errno 2] No such file or directory: 'data/obj' I get this error when i run generate_train.py at 20th cell, pls help me to fix this out...

[@howtodrawit2978]

Thank you very much...can you continue making a video about test and validate process image?

[@blackunicorn_tm]

hi, CUDA-version: 11000 (11020), cuDNN: 7.6.5, GPU count: 1 OpenCV version: 3.2.0 Couldn't open file: cfg/yolov3_custom_drowning.cfg/mydrive/yolov3/backup/yolov3_custom_last.weights this messages appears, What went wrong?

[@vinayakmikkal]

I'm getting a problem. After I run !./darknet detector train data/obj.data cfg/yolov3-custom.cfg darknet53.conv.74 -dont_show I'm getting CUDA-version: 10010 (10010), cuDNN: 7.6.5, GPU count: 1 OpenCV version: 3.2.0 valid: Using default 'data/train.txt' yolov3-custom compute_capability = 750, cudnn_half = 0 How do I solve this issue? I have downloaded the darknet directly into my drive, then i navigated into the darknet file then ran the above command with !chmod u+x darknet because the permission was getting denied.

[@TheAIGuy]

hmm is anything outputting other than those lines? That is the beginning of what should be outputted.

[@GuzelYadigarova]

I got an error that says Error: You set incorrect value batch=1 for Training! You should set batch=64 subdivision=64 ... How can I solve it? [net] # Testing #batch=1 #subdivisions=1 # Training batch=64 subdivisions=64 width=416 height=416 channels=3 momentum=0.9 decay=0.0005 angle=0 saturation = 1.5 exposure = 1.5 hue=.1

[@prashanthmangipudi7160]

Hi, I m not getting any predictions even after training for 4000 steps.... I have done everything as you said. I m not understanding why is it happening ? Can you help me?

[@aqiff123]

I have this error, what's wrong? Saving weights to /mydrive/yolov3/backup/yolov3_custom_last.weights Couldn't open file: /mydrive/yolov3/backup/yolov3_custom_last.weights

[@jayaharshang5454]

I have the following error: Even though it says, Saving weights to /content/drive/My\Drive/yolov3/backup//yolov3_custom_last.weights It neither saves it that location and can't open the saved weights after 100th iteration and shows this: Couldn't open file: /content/drive/My\Drive/yolov3/backup//yolov3_custom_last.weights Please help

[@TheAIGuy]

In your google drive did you make sure to create a yolov3 and backup folder? The folder structure has to be in your google drive already in order for it to work.

[@bhavyaahir9833]

@@TheAIGuy I am facing the same issue even the folder is in the right place

[@jayaharshang5454]

@@TheAIGuy I have created both yolov3 and backup folder

[@TheAIGuy]

@@jayaharshang5454 are you using the mydrive symbolic link?

[@rblansxx3444]

@@TheAIGuy I have exactly the same problem. I have done every step and this is the only error

[@fabianvieri1907]

Great videos and tutorial, I'm curious about the max_batches, batch, and sub_division is it always to be 64,16 and classes * 2000 ?

[@ilkeraykut7064]

Could you find any solutions? Yes,I tihnk there is a formula max_batches= classes * 2000 ,but I am not sure other two.

[@preetharavindran4174]

Thankyou very much for the wonderful video ! It just helped me to sort out all kinda doubts I had related to this topic.

[@sureshramanujam]

Hello, I could start the training but weights file is not creating in backup folder. What could be the reason? It just shows this after the training is completed and I am unable to find these files in the root of darknet. /yolov3-custom.backup /yolov3-custom.backup

[@TheAIGuy]

Are you using google drive?

[@sureshramanujam]

@@TheAIGuy Yes, I am using Google Drive. However, I found the issue. The issue was with the pjreddie repository that I cloned. Instead, I cloned the Darknet from AlexyAB repository and all worked fine. Not sure why the custom weights file wasn't created when I used Darknet repository that I cloned from pjreddie

[@pavancheyutha3228]

I am getting this error while training yolov3 model in colab... Error: l.outputs == params.inputs filters= in the [convolutional]-layer doesn't correspond to classes= or mask= in [yolo]-layer Could you help me out?

[@daanvanheusden9641]

Hi! I've done your tutorial and five others but my output is always the same. Google colab stops immediately. It used to work but now that i'm trying to train tiny weights nothing works anymore. I tried to add -clear 1 at the train command but that doesn't work. Do you maybe have any idea on what to do?

[@damirmukametkarim8835]

Thank you man for great tutorial! I have a question though. I was training model for vehicle detection and classification. Recently I found out that vehicles from particular location are not being detected (cuz of some obstacle). I decided to collect more data from that particular location. Thing is that I have already trained on old dataset and got final weights. Do I have to retrain model on whole dataset again (old dataset + new data) or I can just make dataset on new data and begin training from last weights?

[@khawlasghir6586]

Interesting question. I think that the answer for this is transfer learning. You freeze training weights of the old dataset and you retrain with the newer one to update the model.

[@damirmukametkarim8835]

@@khawlasghir6586 thank you for reply. Yes I've already came up with solution and did what you've just wrtitten. 😁👌

[@sagarsinha6561]

I am trying to run the convert_annotations.py file on my custom dataset. It's showing an error as: File "convert_annotations.py", line 64, in coords = convert(filename_str, coords) File "convert_annotations.py", line 18, in convert coords[0] /= int(image.shape[1]) AttributeError: 'NoneType' object has no attribute 'shape' How to solve this issue? Also in what way can I reach out to you for insights and project ideas? I have recently initiated learning deep learning.

[@kapkap29]

i get that error :( Traceback (most recent call last): File "generate_train.py", line 4, in os.chdir(os.path.join("data", "obj")) FileNotFoundError: [Errno 2] No such file or directory: 'data/obj'

[@tanishqtambe7408]

I had the same issue.... so before extracting all the images from obj.zip , change your directory to /darknet/... this will solve the problem... you run this following code to change the directory import os os.chdir('/content/darknet') and then extract the zip... this will solve your problem

[@wenshuolin5165]

@@tanishqtambe7408 run this command before unzip the obj.zip?

[@logesh9141]

Should need use the images with .jpg format only??? Because I used .jpeg format image i got 'Error: load_data_detection() - Opencv' .Please help me..

[@khainguyen5663]

Hello admin, Can i ask you a question about dataset ? Now i have 4 folder which consist of image and file txt to descript every image in folder but i am very confused in every txt file of 4 folders the class object is 0, Can i fix class object in every folder from 0 to 3 or not ? For example: in txt file of folder 1 is 0, next folder is 1 and etc.

[@saikrishnachalavadi]

hey. i tried this code and everything worked fine but the training part. i checked what is the problem and found that the train.txt was empty. i had like 100 labelled images but its empty. can you help me with this issue.

[@sametyilmaz8106]

Hi. While training the model, I get the error "Can't open label file.(This can be normal only if you use MSCOCO)", how can I solve it?

[@m.mustafayildiz614]

Hello, I did the training as you said and it completed without any problems. I want to try it with a video I posted on google cloud during the testing phase, but I'm getting an error, can you help me?

[@ilkeraykut7064]

Thanks for the video.How many photos I should upload for 1 class? And for more classes,What it should be the values for batch,subdivision and step?Please help.

[@samflynn6004]

My custom detector is currently runnning! Just a question- I only tried uploading 50 images. Where is this number coming from? 132: 680.273926, 495.005737 avg loss, 0.000000 rate, 20.118069 seconds, 8448 images, 79.817513 hours left (8448 images)

[@TheAIGuy]

That number is the total amount of images it has trained over. In your cfg you have a batch value and a max iterations value. Your batch is how many images of your training set you train on per iteration. So if your batch is 32 and your iterations is 4000, it will look at 32 of your images at once and train, then grab a different combo of 32 images and train some more. It does this 4000 times so by the end it has looked at 32 * 4000 = 128,000 images. It will train on each of your 50 images a large amount of times.

[@yavuzgulesen9012]

You said it will take few hours but I have 5 classes and 50 images per class but it takes 9-10 hours to train. Is there a problem I cannot see?

[@akankshyabhattacharyaa7008]

While training I am constantly getting an error which says "1.outputs == params.inputs , filters =in the [convolution]- layer doesn't correspond to classes = or mask =in [yolo] layer Please help

[@kaushikjadhav1367]

Awesome video sir! I am new to your channel and I am already impressed by the videos you make. Thanks for sharing everything!!! 💛💛💛

[@nottp3090]

I have some problem when 13:45 That show output ln: failed to create symbolic link '/mydrive/My Drive': Operation not supported Classroom 'Colab Notebooks' image and 15:19 my out put show me that cp: cannot create regular file '/mydrive/images/detection1.jpg': No such file or directory what should I do?

[@chaitrakulkarni7704]

I am working on trash detection, my academic project plastic bottle, plastic cover, paper, and trash. I am soo confused not getting how to upload my dataset I downloaded from GitHub to yolo. I don't even know if my dataset is compatible with yolo.

[@elijahturatsinze1576]

Thank you so much for this informative tutorial but I got error Traceback (most recent call last): File "generate_train.py", line 4, in os.chdir(os.path.join("data", "obj")) NotADirectoryError: [Errno 20] Not a directory: 'data/obj' May you please help me solve this issue?

[@abrahammathew9783]

Hi, I am getting following error while connecting to Google drive. It would be kind of you if you can help. error : ln: failed to create symbolic link '/mydrive/My\ Drive': Operation not supported

[@angelortiz3564]

This is an absoulutly awesome video!! Do you think it the process has changed that much since 2020?

[@W0lfbaneShikaisc00l]

I find the hacks no longer work... looks like you'll need to fix this so it actually makes the model as now whenever I run the commands it says: uncaught SyntaxError: Unexpected token '.' and whenever I run your notebook it then says: /bin/bash: ./darknet: Is a directory I think your code is broken or Google grew wise to your hacks.

[@achyutmorbekar189]

Will it save the weight files after training is over? Or after every 100 iterations. My model is training and it completed about 260 steps but I can't see weight files in the backup folder.

[@TheAIGuy]

achyut morbekar should save weights every 100 iterations, make sure you created a backup folder on your google drive

[@Lucas-hobbies]

Very Nice Tutorial! Thanks for sharing!

[@jhonjamiltonmajinerazo7894]

hello again, i have a question, i am using custom images, and i am using the "labellmg" tool version windows_v1.8 to label the images. what is the difference of using the .txt (yolo) and xml (pascalVOC) tag files. In this video, which one was used, or which one do you recommend is better?

[@TheAIGuy]

JHON JAMILTON MAJIN ERAZO i use the yolo (.txt) format. It was made to use with yolo as is so you dont need to convert anything. I definitely recommend using yolo format.

[@musicmagic1013]

man I really appreciate it ,this really make trainnig custom object model simple and it helps me a lot.

[@TheAIGuy]

Appreciate it! Thanks for the kind words. All the best.

[@anadimondal2632]

I have tried to get the loss graph after training but got this error. Would you tell me how can I solve this? NameError Traceback (most recent call last) in () ----> 1 imShow('chart.png') NameError: name 'imShow' is not defined

[@TheAIGuy]

you have to run the cell above that defines imShow before you can call it!

[@kushagrak4903]

I'm using Google Colab. Saving weights to /content/drive/MyDrive/yolov3/backup/yolov3-custom_last.weights Couldn't open file: /content/drive/MyDrive/yolov3/backup/yolov3-custom_last.weights In my darknet/data/obj.data I have saved; backup = /content/drive/My Drive/yolov3/backup Please help. P.S it is saving to MyDrive whereas I'm saving it to My Drive. Does this space affects?

[@sheikhwajeeh9753]

Hello can I have some help how to edit this path cause my path have space between 'My Drive' while ( Saving weights to /content/drive/'MyDrive'/yolov3/backup/yolov3_custom_last.weights Couldn't open file: /content/drive/'MyDrive'/yolov3/backup/yolov3_custom_last.weights) does not I hope this is the issue. Thanks

[@newdreamza1-z038]

If we have trained the model on grayscale images with channels = 1 in yolov3custom.cfg file, During inference it is not working.Any other parameters we have to change during inference on test dataset?

[@imanhassan1628]

Hello! I have an error trying to run generate_train.py that says Traceback (most recent call last): File "generate_train.py", line 4, in os.chdir(os.path.join("data", "obj")) FileNotFoundError: [Errno 2] No such file or directory: 'data/obj' generate_train.py is already in my yolov3 file. Does anyone know what could be the problem?

[@WeAllLoveMarlene]

Hey AI Guy! Thanks for this super clear tutorial! I am just running the training and I’m wondering, is it supposed to take THIS long? I’m using 1000 training images from 1 class, I did leave “random=1” in the yolo config file, and it is currently on the iteration 132 with 17.5 hours left still.... is that normal or did I do sth wrong?

[@WeAllLoveMarlene]

And my avg loss is already 139 (went down from 1300 approx)

[@saifullahbhutta]

i am always getting this error, i have trained three times can you, please help me. RuntimeError: shape '[256, 128, 3, 3]' is invalid for input of size 264248

[@tensor7252]

A great tutorial. This really helped me to clear all the doubts regarding YOLOV3 implementation. I have another question.Do you know how to get the coordinates of the bounding box?

[@TheAIGuy]

ten sor to get bounding boxes just add the flag -ext_output to your detection command

[@hibazafar1815]

hey your tutorial is super good, I really learned a lot. Can you please tell me how I can store output of each detection in a file along with the coordinates. Thanks

[@TheAIGuy]

@@hibazafar1815 Hi there and thanks for the kind words! Refer to the official readme of alexeyab/darknet on github as he shows how to do it with detail. A way that works for me is adding this to the end of your darknet detection command '-ext_output >> results.txt'. This will save the detections in a text file. Make sure to then save the results file to your machine or google drive.

[@strangershani]

Hi , I am getting following error please help me out: CUDA Status Error: file: ./src/dark_cuda.c : () : line : 39 : build time : Aug 7 2020 -04:59:39 CUDA Error : no CUDA- capable device is detected CUDA Error: no CUDA-capable device is detected : success darknet: ./src/utils.c: 326 : error : Assertion '0' failed.

[@schneizel935]

Hey! When I tried to copy the configuration file, I got the following error: cp: cannot stat '/cfg/yolov3.cfg': No such file or directory Even though I've cloned the entire github repo perfectly. Appreciate it if you could help

[@rebeccacook40]

Hi, I would like to test the model with my 20% test data after training. I noticed in the video that when creating the obj.data file you mentioned that we wouldn't be using the test.txt. If I would like to test the model I assume I need to generate the test.txt file. Is there anything else I need to change or do? Will it then provide me with the accuracy of the model after training?

[@林冠智-r3s]

Saving weights to /mydrive/yolov3/backup/yolov3_custom_last.weights Couldn't open file: /mydrive/yolov3/backup/yolov3_custom_last.weights I still get this error after changing my obj.data last line "backup = mydrive/yolov3/backup" and I'm sure I creat a bakcup folder in yolov3 , how can I fix it?

[@syedumaidahmed6295]

Can you please guide. How many iterations I have to do for three classes. I have taken 3000 steps and got the loss to 0.59678

[@TheAIGuy]

There is no right answer to determine how many iterations is enough. Totally depends on your model and your training data. However, the recommended iterations is 2000 per class. So your case would be 6000. This could be too many for your case but if you train longer there will be a weights file saved every 1000 iterations so you can test them all.