Train your own model

[Delta144]

I have been trying out the train your own data tutorial on the yolov5 git (https://colab.research.google.com/github/roboflow-ai/yolov5-custom-training-tutorial/blob/main/yolov5-custom-training.ipynb) and I got it working there with roboflow but when I try to put it on Jupyter it adds an \ instead of a / to the dataset path which is what I think is making it come up with an error

code

`# -- coding: utf-8 --
"""Prawn YOLOv5-Roboflow.ipynb

Automatically generated by Colaboratory.

Original file is located at
https://colab.research.google.com/drive/18xuKkU7DeiyidTe29h7Ujs0PMvID-e-B

Custom Training with YOLOv5

In this tutorial, we assemble a dataset and train a custom YOLOv5 model to recognize the objects in our dataset. To do so we will take the following steps:

• Gather a dataset of images and label our dataset
• Export our dataset to YOLOv5
• Train YOLOv5 to recognize the objects in our dataset
• Evaluate our YOLOv5 model's performance
• Run test inference to view our model at work

Step 1: Install Requirements

"""

Commented out IPython magic to ensure Python compatibility.

#clone YOLOv5 and /content/sample_data 39, 35 after bath, 37.8, 37.6, 38.4
#from git import Repo

#Repo.clone_from("https://github.com/ultralytics/yolov5", "C:/Users/Helix/Documents/woodham/yolov5reO")
#!git clone https://github.com/ultralytics/yolov5 # clone repo

%cd C:\Users\Helix\Documents\woodham\yolov5repo

%pip install -qr requirements.txt # install dependencies

%pip install -q roboflow

import torch
import os
from IPython.display import Image, clear_output # to display images
import cv2

#print(f"Setup complete. Using torch {torch.version} ({torch.cuda.get_device_properties(0).name if torch.cuda.is_available() else 'CPU'})")

"""# Step 2: Assemble Our Dataset

In order to train our custom model, we need to assemble a dataset of representative images with bounding box annotations around the objects that we want to detect. And we need our dataset to be in YOLOv5 format.

In Roboflow, you can choose between two paths:

• Convert an existing dataset to YOLOv5 format. Roboflow supports over 30 formats object detection formats for conversion.
• Upload raw images and annotate them in Roboflow with Roboflow Annotate.

Annotate

Version

"""

#from roboflow import Roboflow
#rf = Roboflow(model_format="yolov5", notebook="ultralytics")

Commented out IPython magic to ensure Python compatibility.

set up environment

os.environ["DATASET_DIRECTORY"] = "C:/Users/Helix/Documents/woodham/yolov5repo"# "/content/sample_data"
#%cd C:/Users/Helix/Documents/woodham/yolov5repo/

Commented out IPython magic to ensure Python compatibility.

#after following the link above, recieve python code with these fields filled in
#!pip install roboflow

from roboflow import Roboflow
rf = Roboflow(api_key="eeeeeeeeeeeeeeee")
project = rf.workspace(myworkspace).project("projectname")
dataset = project.version(2).download("yolov5")
%pycat {dataset.location}/data.yaml

"""# Step 3: Train Our Custom YOLOv5 model

Here, we are able to pass a number of arguments:

• img: define input image size
• batch: determine batch size
• epochs: define the number of training epochs. (Note: often, 3000+ are common here!)
• data: Our dataset locaiton is saved in the dataset.location
• weights: specify a path to weights to start transfer learning from. Here we choose the generic COCO pretrained checkpoint.
• cache: cache images for faster training
  """

!python C:/Users/Helix/Documents/woodham/yolov5repo/train.py --img 416 --batch 16 --epochs 150 --data C:/Users/Helix/Documents/woodham/yolov5repo/projectname-2/data.yaml --weights yolov5s.pt --cache

"""# Evaluate Custom YOLOv5 Detector Performance
Training losses and performance metrics are saved to Tensorboard and also to a logfile.

If you are new to these metrics, the one you want to focus on is mAP_0.5 - learn more about mean average precision here.
"""

Commented out IPython magic to ensure Python compatibility.

Start tensorboard

Launch after you have started training

logs save in the folder "runs"

%load_ext tensorboard

%tensorboard --logdir runs

"""#Run Inference With Trained Weights
Run inference with a pretrained checkpoint on contents of test/images folder downloaded from Roboflow.
"""

!python C:/Users/Helix/Documents/woodham/yolov5repo/detect.py --weights C:/Users/Helix/Documents/woodham/yolov5repo/runs/train/exp5/weights/best.pt --img 416 --conf 0.1 --source C:/Users/Helix/Documents/woodham/yolov5repo/test/images

#display inference on ALL test images

import glob
from IPython.display import Image, display

for imageName in glob.glob('C:/Users/Helix/Documents/woodham/yolov5repo/runs/detect/exp2/*.jpg'): #assuming JPG
display(Image(filename=imageName))
print("\n")

"""# Conclusion and Next Steps

Congratulations! You've trained a custom YOLOv5 model to recognize your custom objects.

To improve you model's performance, we recommend first interating on your datasets coverage and quality. See this guide for model performance improvement.

To deploy your model to an application, see this guide on exporting your model to deployment destinations.

Once your model is in production, you will want to continually iterate and improve on your dataset and model via active learning.
"""

#export your model's weights for future use
#from google.colab import files
#files.download('./runs/train/exp/weights/best.pt')
error logloading Roboflow workspace...
loading Roboflow project...
Downloading Dataset Version Zip in C:/Users/Helix/Documents/woodham/yolov5repo/Company-access-2 to yolov5pytorch: 100% [24409923 / 24409923] bytes
Extracting Dataset Version Zip to C:/Users/Helix/Documents/woodham/yolov5repo/Company-access-2 in yolov5pytorch:: 100%|██████████| 940/940 [00:00<00:00, 1023.28it/s]
usage: train.py [-h] [--weights WEIGHTS] [--cfg CFG] [--data DATA] [--hyp HYP]
[--epochs EPOCHS] [--batch-size BATCH_SIZE] [--imgsz IMGSZ]
[--rect] [--resume [RESUME]] [--nosave] [--noval]
[--noautoanchor] [--noplots] [--evolve [EVOLVE]]
[--bucket BUCKET] [--cache [CACHE]] [--image-weights]
[--device DEVICE] [--multi-scale] [--single-cls]
[--optimizer {SGD,Adam,AdamW}] [--sync-bn] [--workers WORKERS]
[--project PROJECT] [--name NAME] [--exist-ok] [--quad]
[--cos-lr] [--label-smoothing LABEL_SMOOTHING]
[--patience PATIENCE] [--freeze FREEZE [FREEZE ...]]
[--save-period SAVE_PERIOD] [--seed SEED]
[--local_rank LOCAL_RANK] [--entity ENTITY]
[--upload_dataset [UPLOAD_DATASET]]
[--bbox_interval BBOX_INTERVAL]
[--artifact_alias ARTIFACT_ALIAS]
train.py: error: unrecognized arguments: C:\Users\Helix\Documents\woodham\yolov5repo\projectname-2/data.yaml
detect: weights=['C:/Users/Helix/Documents/woodham/yolov5repo/runs/train/exp5/weights/best.pt'], source=C:/Users/Helix/Documents/woodham/yolov5repo/test/images, data=yolov5repo\data\coco128.yaml, imgsz=[416, 416], conf_thres=0.1, iou_thres=0.45, max_det=1000, device=, view_img=False, save_txt=False, save_conf=False, save_crop=False, nosave=False, classes=None, agnostic_nms=False, augment=False, visualize=False, update=False, project=yolov5repo\runs\detect, name=exp, exist_ok=False, line_thickness=3, hide_labels=False, hide_conf=False, half=False, dnn=False, vid_stride=1
YOLOv5 2022-11-9 Python-3.9.15 torch-1.13.0+cpu CPU

Traceback (most recent call last):
File "C:\Users\Helix\Documents\woodham\yolov5repo\detect.py", line 258, in
main(opt)
File "C:\Users\Helix\Documents\woodham\yolov5repo\detect.py", line 253, in main
run(**vars(opt))
File "Z:\anaconda3\envs\AUTOINSTALyolo\lib\site-packages\torch\autograd\grad_mode.py", line 27, in decorate_context
return func(*args, **kwargs)
File "C:\Users\Helix\Documents\woodham\yolov5repo\detect.py", line 95, in run
model = DetectMultiBackend(weights, device=device, dnn=dnn, data=data, fp16=half)
File "C:\Users\Helix\Documents\woodham\yolov5repo\models\common.py", line 345, in init
model = attempt_load(weights if isinstance(weights, list) else w, device=device, inplace=True, fuse=fuse)
File "C:\Users\Helix\Documents\woodham\yolov5repo\models\experimental.py", line 79, in attempt_load
ckpt = torch.load(attempt_download(w), map_location='cpu') # load
File "Z:\anaconda3\envs\AUTOINSTALyolo\lib\site-packages\torch\serialization.py", line 771, in load
with _open_file_like(f, 'rb') as opened_file:
File "Z:\anaconda3\envs\AUTOINSTALyolo\lib\site-packages\torch\serialization.py", line 270, in _open_file_like
return _open_file(name_or_buffer, mode)
File "Z:\anaconda3\envs\AUTOINSTALyolo\lib\site-packages\torch\serialization.py", line 251, in init
super(_open_file, self).init(open(name, mode))
FileNotFoundError: [Errno 2] No such file or directory: 'C:\Users\Helix\Documents\woodham\yolov5repo\runs\train\exp5\weights\best.pt'`

[capjamesg]

It looks like the weights were not found. Can you confirm you put them in runs\train\exp5\weights\?