I'm using Tensorflow 2 Model Maker to perform transfer training of EfficientDet-Lite (ultimately to run on a Coral EdgeTPU) and I care much more about the classification output and much less about the precision of the bounding boxes. Is there a way to modify some training parameters to improve the accuracy of the classes at the expense of the accuracy of the bounding boxes? Or does this not make sense?
1
There are 1 best solutions below
Related Questions in TENSORFLOW
- A deterministic GPU implementation of fused batch-norm backprop, when training is disabled, is not currently available
- Keras similarity calculation. Enumerating distance between two tensors, which indicates as lists
- Does tensorflow have a way of calculating input importance for simple neural networks
- How to predict input parameters from target parameter in a machine learning model?
- Windows 10 TensorFlow cannot detect Nvidia GPU
- unable to use ignore_class in SparseCategoricalCrossentropy
- Why is this code not working? I've tried everything and everything seems to be fine, but no
- Why convert jpeg into tfrecords?
- ValueError: The shape of the target variable and the shape of the target value in `variable.assign(value)` must match
- The kernel appears to have died. It will restart automatically. whenever i try to run the plt.imshow() and plt.show() function in jupyter notebook
- Pneumonia detection, using transfer learning
- Cannot install tensorflow ver 2.3.0 (distribution not found)
- AttributeError: module 'keras._tf_keras.keras.layers' has no attribute 'experimental'
- Error while loading .keras model: Layer node index out of bounds
- prediction model with python tensorflow and keras, gives error when predicting
Related Questions in TENSORFLOW2.0
- What is the alternative to module: tf.keras.preprocessing?
- Tensorflow Graph Execution Permission Denied Error
- No module named 'keras.layers.core
- TypeError: 'NoneType' object is not subscriptable when training with tf.dataset
- Cannot install tensorflow ver 2.3.0 (distribution not found)
- Keras predict/predict_on_batch giving different answers than predict_step/__call__()
- Problems with using the most recent version if Tensorflow
- I have been trying to convert a TensorFlow code to pytorch and main problem is ,what to use in place tf.keras.layers.Layer in pytorch
- How to run tflite interpreter with custom ops of tensorflow_text ops in java android
- Tensorflow is unable to find GPU
- tensorflow bert 'tuple' object has no attribute problem
- TensorFlow : how to use Model.Predict for multiple predictions at once
- Understanding output of model.get_weights() in Keras
- DeepXDE error - NotImplementedError: Unsupported backend: torch
- unable to import model model_main (from object_detection import model_main)
Related Questions in TENSORFLOW-LITE
- How to add a class to an existing model, reduce the images or classes and limit the objects it can detect at a time
- Comparing analog signal from Electret mic with samples
- How to run tflite interpreter with custom ops of tensorflow_text ops in java android
- Tensorflow Lite error Unity : NullReferenceException: Object reference not set to an instance of an object
- TensorFlowLite Error : ‘Interpreter.GetOutputTensor(int)’ is inaccessible due to its protection level
- Measuring TensorFlow Lite Model Speed
- RangeError when converting Uint8List to ByteBuffer for image processing in Flutter
- TensorflowLite output tensorsor data extraction
- Tensorflow Lite: ImportError: libusb-1.0.so.0: cannot open shared object file: No such file or directory
- How to Verify TensorFlow Lite Model is Using XNNPACK?
- Why does python produce a more accurate result than kotlin when using tensorflow?
- How to compile a Gemma 7b TFLite model for MediaPipe?
- Internal error: Cannot create interpreter: Didn't find op for builtin opcode 'FULLY_CONNECTED' version '9' with firebase ml kit and tensorflow-lite
- ML Kit object detection using custom model how to get bounding box coordinate?
- Can't convert saved keras model to TFLite
Related Questions in EDGE-TPU
- Can running the same model on CPU and Edge TPU yield different results?
- How to include Coral into Yocto Build
- MiDaS model quantization on coral edge TPU
- How to deal with Google-Coral source code, especially libedgetpu library?
- Why is my custom model not running on the coral micro edge tpu?
- Segmentation fault when initializing a TFLite interpreter on one edge TPU, but no issue initializing it on another device, using the same model
- Limit for Model Size on Google Coral Micro Dev Board?
- Compile a TF Javascript model for Google Coral Edge TPU module
- How do I resolve the error: Attempting to use a delegate that only supports static-sized tensors with a graph that has dynamic-sized tensors?
- Converting a TF model to TFLite and then to EdgeTPU
- how to link delegate with edgetpu compiler to compile
- Attempting to use a delegate that only supports static-sized tensors with a graph that has dynamic-sized tensors
- Facing such problem like no model name edge tpu found and no module name edgeTPU_cpp_wrapper
- import edgetpu.detection.engine ModuleNotFoundError: No module named 'edg
- "IndexError: index 10 is out of bounds for axis 0 with size 10" when running inference on Coral Dev Board
Trending Questions
- UIImageView Frame Doesn't Reflect Constraints
- Is it possible to use adb commands to click on a view by finding its ID?
- How to create a new web character symbol recognizable by html/javascript?
- Why isn't my CSS3 animation smooth in Google Chrome (but very smooth on other browsers)?
- Heap Gives Page Fault
- Connect ffmpeg to Visual Studio 2008
- Both Object- and ValueAnimator jumps when Duration is set above API LvL 24
- How to avoid default initialization of objects in std::vector?
- second argument of the command line arguments in a format other than char** argv or char* argv[]
- How to improve efficiency of algorithm which generates next lexicographic permutation?
- Navigating to the another actvity app getting crash in android
- How to read the particular message format in android and store in sqlite database?
- Resetting inventory status after order is cancelled
- Efficiently compute powers of X in SSE/AVX
- Insert into an external database using ajax and php : POST 500 (Internal Server Error)
Popular # Hahtags
Popular Questions
- How do I undo the most recent local commits in Git?
- How can I remove a specific item from an array in JavaScript?
- How do I delete a Git branch locally and remotely?
- Find all files containing a specific text (string) on Linux?
- How do I revert a Git repository to a previous commit?
- How do I create an HTML button that acts like a link?
- How do I check out a remote Git branch?
- How do I force "git pull" to overwrite local files?
- How do I list all files of a directory?
- How to check whether a string contains a substring in JavaScript?
- How do I redirect to another webpage?
- How can I iterate over rows in a Pandas DataFrame?
- How do I convert a String to an int in Java?
- Does Python have a string 'contains' substring method?
- How do I check if a string contains a specific word?
Unfortunately, TensorFlow 2 Model Maker doesn't support such customization at this moment.
If you want to do so, you can bypass Model Maker and directly use AutoML repo. The technical detail is to adjust weights for different losses by adding
loss_weightsincompile()function.