TensorFlow Object Detection API trains only one class from several
Im running training process with TF Object Detection API on my own dataset with several classes, however after pretty big amount of steps (about 20k) I can see at PerformanceByCategory menu in TensorBoard that mAP is increasing only for one category, other are not even started.
My dataset is in PascalVOC format and I was following this answer to create it. After data preparation I have generated Pascal records, edited label map and pipeline config, downloaded all the staff with ssd model checkpoints to Google Cloud Platform and started training and evaluation jobs accordingly to Object Detection API Documentation.
But it seems from TensorBoard that something is wrong with data or something other because nothing happens with my another classes.
Are there any hints or features when training Object Detection API with several classes?
Thanks for any help in advance!
[EDITED]:
So having tried to train model with PascalVOC dataset they provide by default in tutorial I noticed that their Pascal records are much larger than mine. I opened them in text editor and there were references for every class from ImageSets, however my records contain references for only that class I pointed in create_pascal_tf_records.py.
What am I missing guys? Please help!
OK, I think I have figured out what was wrong.
The main problem was I made ImageSets uncorrectly. For each class I have been adding their corresponding lists of images, so each train and val files were different in my case.
Due to this answer:
The punchline is that we currently only look at the first column of the aeroplane_train.txt file (which is the same as the other xxx_train.txt files) and lets us figure out the path for each image in the training set.
Because of every file is different the create_pascal_tf_records.py was picking paths to only my first class.
Hope this will help someone else.
- Python method chaining in functional programming style
- flask-jwt-extended: Fake Authorization Header during testing (pytest)
- For loop through the list unless empty?
- Polars make all groups the same size
- Is there a way to specify a default base-template for all templates in django?
- How to tackle time limit exceeded error in leetcode
- Is pd.get_dummies() updated in newer versions of Pandas making it default to Booleans (True/False) instead of (0/1)?
- What's the function like sum() but for multiplication? product()?
- How to type hint a dynamically-created dataclass
- Issue with pulling the data with EIA API with Python
- 403 Forbidden Error when scraping a site, user-agents already used and updated. Any ideas?
- Fullstack web-hosting services
- How to handle an AnalysisException on Spark SQL?
- Python requests is slow and takes very long to complete HTTP or HTTPS request
- Is there a way to modify an element in a Numpy array based on the value of other elements?
- Tkinter grid manager height/width nonconsistent
- Sql Alchemy Insert Statement failing to insert, but no error
- How can I create a Polars struct while eval-ing a list?
- Excel using win32com and python
- django: on pypy, psyco, unladen swallow or cpython, which one is the fastest?
- How convert a list into multiple columns and a dataframe?
- Name not defined in type annotation
- Static type checkers and Language Servers not recognizing attributes of objects that are subclasses
- How do I get multiple OID values in PySNMP?
- how to log a file in Django
- Is there a simple and efficient way to evaluate Elementary Symmetric Polynomials in Python?
- Iterating over two lists one after another
- Python -i flag for production
- What is PyCompilerFlags in Python C API?
- How to make Python check whether a variable is a number or letter