I already have an object trained files called detect.tflite, labelmap.txt and tflite_graph.pbtxt and tflite_graph.pb. but i don’t have the dataset’s used for training this model. I only have above mentioned files.
Now i need to add or append some more dataset with this existing files called detect.tflite and labelmap.txt for the new dataset i labelled the images and uploaded for tensorflow training, and i conveted the grpah and downloaded the files called detect.tflite, labelmap.txt and labelmap.pbtxt
I followed the below link for training the new data set.
Now i want to merge both files.
how to append new data or how to merge both files??? Please guide me.
Thanks for your response.
In previous data set, 5 objects classified and trained. Now one of the object shape is changed slightly. So with my previous dataset AI is not recognize the image.
So I trained the new object (only changed object) and downloaded 1 .tflite file. for this training i took 400 images. Now the AI detect this object alone. But from previous data set it needs to detect balance 4 object.
But the problem is we missed 4 object images was used for training.
Hi @Kaviyarasi1234, As far as i know, it’s not suggestible to combine models trained on 2 different datasets. You have to train the model on the whole dataset for better results. Also, you can try stacking which involves training multiple models on the same or different data sets and then using their predictions as inputs for another model. Thank You.