DependableSystemsLab/TensorFI
TensorFI is a fault injection framework for injecting both hardware and software faults into applications written using the TensorFlow framework. You can find more information about TensorFI in the paper below.
PythonMIT
Issues
- 1
Vgg11 dataset not working on python2
#52 opened by zoythum - 0
'injectFaultMean' doesn't work
#42 opened by ElaineYao - 3
Bit Flips in the weights
#48 opened by abhishektyaagi - 1
ERROR:root:Unable to execute run on Tensor("fi_Mean_1:0", dtype=float32)
#49 opened by abhishektyaagi - 1
- 1
- 1
- 1
py_func Crashes
#47 opened by altostratous - 1
- 3
Unable to execute run on " + str(fiTensor)
#43 opened by xiaozhouqi - 1
Reshape returning NoneType
#41 opened by ManiSadati - 2
OverflowError: cannnot convert float infinity to integer in faultTypes.py Line 95
#39 opened by ElaineYao - 0
Add the support for float16
#38 opened by ElaineYao - 0
error in install the TensorFI
#26 opened by whyseu - 1
- 1
Exclude certain operators from injection
#14 opened by nniranjhana - 0
issues with bitFlip-tensor fault type
#13 opened by nniranjhana - 0
- 2
BatchNorm and data format.
#9 opened by AlessandroToschi - 8
FI Configuration Doubt
#7 opened by AlessandroToschi - 0
error when the model has tf.cond
#6 opened by flyree - 5
TF Keras Layers
#4 opened by AlessandroToschi