How to convert tflite model to quantized model without pb file - python

I'm trying to convert a tflite model to it's quantized version. (I try to convert the pose estimation module multi_person_mobilenet_v1_075_float.tflite hosted here) to it's quantized version.
I therefore installed the tflite_converter command line tool, recommended here. But the examples do not fit my case where I only have a *.tflite file and no corresponding frozen_graph.pb file.
Thus when I just call
tflite_convert --output_file multi_person_quant.tflite --saved_model_dir ./
from within the directory containing multi_person_mobilenet_v1_075_float.tflite, I get an error message:
IOError: SavedModel file does not exist at: .//{saved_model.pbtxt|saved_model.pb}
I guess I need a .pb file for whatever I want to do... Any idea how to generate it from the *.tflite file?
Any other advice would also be helpful.

Related

How to change ML python program to tflite so that we can use it in android studio

I am using nearest neighbour algorithm in the ML program. I am able to change program to .pb file but .pb file is not changing to tflite because of some error :-
RuntimeError: MetaGraphDef associated with tags {'serve'} could not be found in SavedModel, with available tags '[]'. To inspect available tag-sets in the SavedModel, please use the SavedModel CLI: saved_model_cli.
Please tell me how can I change .py to tflite.

Convert .pt to .pth?

I've been trying to compile my custom trained YoloV5 model using SageMaker Neo.
The compilation gives an error :
ClientError: InputConfiguration: No pth file found for PyTorch model.
Please make sure the framework you select is correct.
The weights are a .pt file.
Is there a way to convert the .pt file to .pth?
There is no difference between both file extensions. If SageMaker is only looking for a file with a .pth extension, then you can simply rename your file from filename.pt to filename.pth.

Yolov4 weights conversion to tflite failure

I am using Google Colabs to convert Yolov4 darknet weights to tflite version.
I used this blog to train my own yolo detector and I got to acceptable accuracy for my detections.
Then I tried every repository on github 1,2,3 to covert my custom yolov4 weights to tflite version and I failed in each time. I faced the problem (can not reshape array) here and I solved it by modifying file.names and adjusting the change in the config file. In the end when I executed this code:
!python convert_tflite.py --weights ./data/yolov4.weights --output ./data/yolov4.tflite
I faced this error:
ValueError: Input 0 of node model_1/batch_normalization/AssignNewValue was passed float from model_1/batch_normalization/FusedBatchNormV3/ReadVariableOp/resource:0 incompatible with expected resource.
This problem was discussed here but I couldn't find any solution
I am using google colab which make it easier with using the preinstalled libraries but maybe there could be some incomparability in libraries used in conversion ?
Any help would be appreciated..

Convert .pth Pytorch model to format readable by OpenCv

I'm trying to use this model: https://github.com/zllrunning/face-makeup.PyTorch/blob/master/cp/79999_iter.pth with opencv dnn module but because opencv doesn't read .pth file, I need to convert it to another format.
First I tried to convert to .onnx, I found code to do it but when I call the function "torch.onnx.export", I got "RuntimeError: ONNX symbolic expected a constant value in the trace".
I tried with differrent version of torch but I always got almost the same error message.
I tried to convert to caffe model, there is some available github to do it but none of them worked for me.
I didn't find how to convert directly to tensorflow, they always convert to onnx and then to tensorflow.
If someone can share me a code that works for this model and the version of all packages used. Or if you could share me the model converted to another format, it would be greatly appreciated.

How to generate HDF5/H5 file in Keras with existing Python project

I am trying to get HDF5/H5 file from existing project in keras.
this attention_ocr is related to OCR written in python. I would like to generate HDF5/H5 file so I can convert that with tensorflowjs_converter[ref] and will use in browser.
Reference:
How to import a TensorFlow SavedModel into TensorFlow.js
Importing a Keras model into TensorFlow.js
I am looking for installing keras environment and generating HDF5/H5 file.
Once your model is trained in keras, saving it as an HDF5 is simply one line:
my_model.save('my_filename.h5')

Categories

Resources