I am yolov4 on raspberry piThis repositoryI'm thinking of detecting an object using (referred to).
However, when I start object detection, I get an error saying that there is not enough memory in just 10 seconds.
The error content is:
RuntimeError: [enforce fail at CPUAllocator.cpp: 64]. DefaultCPUAllocator: can't allocate memory: you tried to allocate 2957312 bytes. Error code 12 (Cannot allocate memory)
I interpreted this error as "I tried to use 2957312 bytes of memory, but it wasn't enough."
However, when I typed the free command while processing the object detection, I found that 4.7GB was still free.
Then you can use enough memory, right?
I'm still wondering why the error occurs.
Answer # 1
You're probably using memory to read the learned weight.
I made the input width and height of obj.cfg smaller (multiple of 32) and it worked.
- i tried to collect by google_images_download to collect machine learning data, but i can not download even one
- Bayesian classification for python machine learning
- Meta linear regression for machine learning in Python
- Multiple linear regression for machine learning in Python
- KNN classification algorithm for python machine learning
- KNN algorithm principle of machine learning and Python implementation method
- Detailed machine learning k-nearest neighbor algorithm
- Python Machine Learning Library NumPy Getting Started Tutorial
- Random machine learning in python (7)
- C # machine entry learning to determine whether the daily report is qualified
- Example of kNN algorithm for machine learning in pure python
- K-means clustering for Python machine learning algorithms
- Python machine learning theory and actual combat (four) logistic regression
- Python machine learning theory and actual combat (two) decision tree
- Python machine learning theory and practice (six) support vector machine
- Python machine learning theory and actual combat (1) K nearest neighbor method
- Python machine learning theory and practice (5) support vector machine
- Java virtual machine deep learning memory management mechanism
- python - [mnist] pytorch size mismatch error
- machine learning - basic questions about batch normalization
- python - the bias part of the expression conv-bias-relu?
- machine learning - i do not understand how to read the output shape of torchsummary
- python - backpropagation and differentiability problem
- python - does ssd (single shot multibox detector) not have "roi pooling" or "roi align" processing?
- python - principle of inputting and using feature maps of different sizes in parallel in ssd and m2det?
- python - in yolov3 original data learning log iou wants to eliminate nan state
- python - what is the difference between ssd's multibox and faster r-cnn's sliding window?