|
About the quantization category
|
|
0
|
434
|
October 2, 2019
|
|
Is QAT Inference not support GPU?
|
|
0
|
11
|
January 25, 2021
|
|
How to fuse layers of any convolutional neural network?
|
|
2
|
242
|
January 23, 2021
|
|
Changed quantized weights not reflecting in state_dict()
|
|
0
|
27
|
January 23, 2021
|
|
Changing Quantized Weights
|
|
7
|
79
|
January 23, 2021
|
|
Is there a way to quantize conv_transpose2d layer?
|
|
29
|
694
|
January 20, 2021
|
|
Is there an alternative to do batched matrix multiplication on Quantized Tensors?
|
|
6
|
315
|
January 20, 2021
|
|
Change Linear layer weights in a quantized model
|
|
2
|
34
|
January 20, 2021
|
|
Can't quantize Linear + Relu
|
|
2
|
41
|
January 20, 2021
|
|
Quantizing only a part of the model
|
|
5
|
86
|
January 19, 2021
|
|
Static Quantization: RuntimeError about backend
|
|
1
|
31
|
January 15, 2021
|
|
RuntimeError: Could not run 'aten::thnn_conv2d_forward' with arguments from the 'QuantizedCPUTensorId' backend. 'aten::thnn_conv2d_forward' is only available for these backends: [CPUTensorId, VariableTensorId]
|
|
11
|
1352
|
January 15, 2021
|
|
The difference of `torch.nn.quantized.FloatFunctional.add` and `torch.nn.quantized.FloatFunctional.add_scalar`?
|
|
1
|
32
|
January 15, 2021
|
|
RuntimeError: Module 'NnapiModule' has no attribute 'weights' (This attribute exists on the Python module, but we failed to convert Python type: 'list' to a TorchScript type.):
|
|
1
|
51
|
January 12, 2021
|
|
`Expected self.scalar_type() == ScalarType::Float to be true, but got false.` when doing quantization aware training?
|
|
2
|
60
|
January 14, 2021
|
|
Loading state to a quantized Inception_v3 model
|
|
11
|
131
|
January 11, 2021
|
|
TypeError: avg_pool2d(): argument 'kernel_size' must be tuple of ints, not Proxy
|
|
1
|
31
|
January 11, 2021
|
|
Quantization-aware training with DataParallel crashes randomly
|
|
5
|
142
|
January 11, 2021
|
|
Can we prune a quantized model?
|
|
0
|
47
|
January 10, 2021
|
|
Should I do the following if I implement quantization aware training?
|
|
3
|
61
|
January 8, 2021
|
|
How could I save state dict in int8 format and load back to float32?
|
|
3
|
52
|
January 7, 2021
|
|
nn.Linear does not have `weight` any more after quantized
|
|
1
|
44
|
January 6, 2021
|
|
I quantized input tensor and all layers except last layer, but got RuntimeError: Could not run 'quantized::conv2d.new' with arguments from the 'CPU' backend
|
|
2
|
67
|
January 6, 2021
|
|
Size of exported onnx model
|
|
0
|
44
|
December 30, 2020
|
|
Broken disable_fake_quant
|
|
7
|
67
|
December 28, 2020
|
|
Quantized model takes more time than fp32 model on CPU
|
|
3
|
72
|
December 28, 2020
|
|
Getting original module from RecursiveScriptModule
|
|
2
|
76
|
December 26, 2020
|
|
Torch Jit Modules without parameters
|
|
6
|
106
|
December 26, 2020
|
|
Replace parameters of model
|
|
1
|
61
|
December 22, 2020
|
|
QAT: RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu
|
|
5
|
147
|
December 22, 2020
|
