quantize | Simple color palette quantization using MMCQ | Compression library

I've trained a quantized model (with help of quantized-aware-training method in pytorch). I want to create the calibration cache to do inference in INT8 mode by TensorRT. When create calib cache, I get the following warning and the cache is not created:

I want to use a generator to quantize a LSTM model.

Questions

I start with the question as this is quite a long post. I actually want to know if you have manged to quantize (int8) a LSTM model with post training quantization.

I tried it different TF versions but always bumped into an error. Below are some of my tries. Maybe you see an error I made or have a suggestion. Thanks

Working Part

The input is expected as (batch,1,45). Running inference with the un-quantized model runs fine. The model and csv can be found here:
csv file: https://mega.nz/file/5FciFDaR#Ev33Ij124vUmOF02jWLu0azxZs-Yahyp6PPGOqr8tok
modelfile: https://mega.nz/file/UAMgUBQA#oK-E0LjZ2YfShPlhHN3uKg8t7bALc2VAONpFirwbmys

I am able to convert the new_target column into numerical form. But as the factor form is already numerical, I am left with a bunch of numbers. I want them ordered and reassigned to their equivalent from 0 to the length of the factor. I have a numerical target at first, then I quantize it to 20 bins. As a result, I obtain new_target column which consists of the unique values (0,1,3,14,16,18,19). Instead of these unique values I need values ordered from 0 to length of the unique values in new_target. Which are c(0,1,2,3,4,5,6). The expected output is given in new_target_expected column. How can I create new_target_expected column without manually creating it? I have a bigger dataframe I am dealing with and it is not possible to do this manually.

I have a raspberry pi 4 with a inkyWHAT display, I've managed to get the display showing my own images.

What I need help with is to run the following commands one after the other, at present I paste each line in one at a time:

I am working with quantized neural networks (need input image with pixels [0, 255]) for a while. For the ssd_mobilenet_v1.tflite model the following standartization parameter are given though https://tfhub.dev/tensorflow/lite-model/ssd_mobilenet_v1/1/metadata/2 :

Goal: Amend this Notebook to work with Albert and Distilbert models

Kernel: conda_pytorch_p36. I did Restart & Run All, and refreshed file view in working directory.

Error occurs in Section 1.2, only for these 2 new models.

For filenames etc., I've created a variable used everywhere:

When working with the built in decimal module in python I can round decimals as follows.

I came across two methods to get precision in floating numbers - using round or using the Decimal package.

What i observed (with few examples tried in REPL) that both produce the same results:

I create a simple plot using GLMakie: