hello,
I have many questions regarding the type of tflite models from tensorflow 2.x that can be imported using TIDL:
1. do you support for pre-quantized models, this means: is it possible to import a full-integer pre-quantized model from tensorflow that used TFlite tools (not yours) and import the model in TIDL?, the documentation is not clear about that https://software-dl.ti.com/jacinto7/esd/processor-sdk-rtos-jacinto7/08_00_00_12/exports/docs/tidl_j7_08_00_00_10/ti_dl/docs/user_guide_html/md_tidl_model_import.html#tidl_model_import_ex6
2. The models in https://software-dl.ti.com/jacinto7/esd/processor-sdk-rtos-jacinto7/08_00_00_12/exports/docs/tidl_j7_08_00_00_10/ti_dl/docs/user_guide_html/md_tidl_models_info.html are pre-quantize (using TOCO convertion -> old model conversion!) how can I do it for tflite models (float/pre-quantized) in the tensorflow > 2.3?
3. Importing a simple CNN exported in tf 2.4.1 with the following structure gave me the next error (even the error contains typos):
```
TFLite Model (Flatbuf) File : small_model_float.tflite
TIDL Network File : tidl_model.bin
TIDL IO Info File : tidl_param.bin
8
Running tidl_optimizeNet
In put of TIDL_InnerProductLayer layer needs to be Faltten. Please add Flatten layer to import this mdoels
```
the model contains a Flatten layer (Reshape node visible in Netron APP)
4.
