Open SagarSharma4244 opened 6 years ago
Have a look at this line in the retrain.py
script:
https://github.com/googlecodelabs/tensorflow-for-poets-2/blob/6be494e0300555fd48c095abd6b2764ba4324592/scripts/retrain.py#L899
You need to specify the suffix as quantized
not quant
.
I'm using Ubuntu16 and Tensorflow1.9 and python3.5 I have used all of these commands to retrain quant model and I'm getting the same error:
python scripts/retrain.py --output_graph=tf_files/retrained_graph.pb --output_labels=tf_files/retrained_labels.txt --image_dir=flower_photos --architecture MobileNet_1.0_224_quant
orpython scripts/retrain.py --output_graph=tf_files/retrained_graph.pb --output_labels=tf_files/retrained_labels.txt --image_dir=flower_photos --architecture mobilenet_1.0_224_quant
orpython scripts/retrain.py --output_graph=tf_files/retrained_graph.pb -- output_labels=tf_files/retrained_labels.txt --image_dir=flower_photos --architecture MobileNet_v1_1.0_224_quant
Error:ERROR:tensorflow:Couldn't understand architecture suffix 'quant' for 'mobilenet_1.0_224_quant'
ERROR:tensorflow:Did not recognize architecture flag
Can someone please tell me the correct command