Hass.io + tensorflow object_detection

Hello, I searched all over the internet but I can not seem to find any help related to this. I´m new to this and perhaps this has a easy fix. My Hass.io in installed in a docker, and the ubuntu server as a VM in proxmox, the hypervisor is installed on a Intel NUC. 

The logs says: 
**Invalid config for [image_processing.tensorflow]: not a file for dictionary value @ data['model']['graph']. Got '/configuration/tensorflow/ssd_mobilenet_v2_coco_2018_03_29/frozen_inference_graph.pb'. (See /config/configuration.yaml, line 29). Please check the docs at https://www.home-assistant.io/integrations/tensorflow**

My configuration for tensorflow: 
**image_processing:**
**  - platform: tensorflow**
**    source:**
**      - entity_id: camera.camera**
**    model:**
**      graph: /configuration/tensorflow/ssd_mobilenet_v2_coco_2018_03_29/frozen_inference_graph.pb**

I really hope someone has a solution for this, I sat for hours and hours trying to solve this. 

Not sure if it matters but I have the full path for mine…

/home/homeassistant/.homeassistant/tensorflow/faster_rcnn_inception_v2_coco_2018_01_28/frozen_inference_graph.pb
/usr/share/hassio/.homeassistant/configuration/tensorflow/ssd_mobilenet_v2_coco_2018_03_29/frozen_inference_graph.pb

Did put full path from root, same result :confused: 

check your file permissions if you are sure that it is definitely located where it is supposed to be, you (home assistant user) may not have the proper file permissions

wow, i missed ./, I only wrote /
Thank you for your time :slight_smile:

1 Like