Transformed resnet20atcifar10 tflite model output is nan

[sunzhe09]

INFO:tensorflow:./models/model_transformed.pb generated
INFO:tensorflow:input: import/net_input:0 / output: import/net_output:0
INFO:tensorflow:outputs from the *.pb model: [[0.0504508 0.1560941 0.13346776 0.3841547 0.01690246 0.04628923
0.0204697 0.08250584 0.0209629 0.08870252]]
INFO:tensorflow:time consumption of *.pb model: 5.40 ms
INFO:tensorflow:./models/model_transformed.pb -> ./models/model_transformed.tflite
INFO:tensorflow:./models/model_transformed.tflite generated
INFO:tensorflow:input details: [{'index': 175, 'quantization': (0.0, 0), 'shape': array([ 1, 32, 32, 3], dtype=int32), 'dtype': <class 'numpy.float32'>, 'name': 'net_input'}]
INFO:tensorflow:output details: [{'index': 176, 'quantization': (0.0, 0), 'shape': array([ 1, 10], dtype=int32), 'dtype': <class 'numpy.float32'>, 'name': 'net_output'}]
INFO:tensorflow:outputs from the *.tflite model: [[nan nan nan nan nan nan nan nan nan nan]]
INFO:tensorflow:time consumption of *.tflite model: 3.37 ms

[jiaxiang-wu]

How did you obtain the compressed model? Trained with ChannelPrunedLearner?

[sunzhe09]

yes，besides，with enbl_dst

[jiaxiang-wu]

Can you provide the *.ckpt models?

[sunzhe09]

sorry，I am afraid I don‘t have the right

[jiaxiang-wu]

That's fine. Let's try to reproduce your issue, but this may take some time.

[Augnine]

I have the same problem today, have you solved it? @sunzhe09 @jiaxiang-wu