|
Index out of bounds Error with PerChannel Quantization
|
|
6
|
561
|
March 19, 2024
|
|
How is quantization of activations handled in pytorch after QAT?
|
|
15
|
2286
|
March 18, 2024
|
|
Example of using quantized.Embedding?
|
|
1
|
103
|
March 18, 2024
|
|
Where is torch.ops.quantized.conv2d defined?
|
|
1
|
112
|
March 18, 2024
|
|
Could not run 'quantized::conv2d.new' with arguments from the 'QuantizedCUDA' backend
|
|
11
|
10966
|
October 5, 2022
|
|
How to initialize a quantized.Embedding?
|
|
1
|
109
|
March 3, 2024
|
|
What is the correct way to qat a conv layer with weight norm
|
|
2
|
536
|
March 2, 2024
|
|
Input image with int?
|
|
8
|
2037
|
February 29, 2024
|
|
Post Training Static Quantization API still uses float weights instead of int?
|
|
4
|
1248
|
February 28, 2024
|
|
How to apply per_tensor_symmetric activation quantization?
|
|
7
|
634
|
February 28, 2024
|
|
LSTM quant to get the scale and zp of hidden input
|
|
0
|
94
|
February 28, 2024
|
|
Quantized Resnet has no parameters()
|
|
2
|
129
|
February 27, 2024
|
|
TypeError: quantized_add() missing 2 required positional arguments: 'op_scale' and 'op_zero_point'
|
|
6
|
626
|
February 27, 2024
|
|
How does gradient calculation in quantization-aware training? Straight through estimatior?
|
|
0
|
141
|
February 27, 2024
|
|
Does "De-Quantization" happens/executed after every Conv/Linear layer when using "torch.quantization.quantize_dynamic"
|
|
4
|
148
|
February 22, 2024
|
|
QAT - how to handle nn.Parameter
|
|
5
|
235
|
February 21, 2024
|
|
What is the recommended way to use (PyTorch naïve quantization) when deploying to int8 TensorRT?
|
|
0
|
141
|
February 21, 2024
|
|
Correctly changing precision in DLRM
|
|
1
|
96
|
February 20, 2024
|
|
Wav2vec2 quantization dimention error
|
|
4
|
169
|
February 20, 2024
|
|
How to export a correct quantized model to onnx format
|
|
5
|
1275
|
February 16, 2024
|
|
Decrease in model parameters in dynamic quantization
|
|
3
|
186
|
February 13, 2024
|
|
Dynamic Quantization produces inconsistent outputs
|
|
1
|
187
|
February 13, 2024
|
|
Could not run 'quantized::conv2d.new'
|
|
2
|
213
|
February 13, 2024
|
|
Quantizing to int8 without stubs for input and output?
|
|
6
|
287
|
February 13, 2024
|
|
Why does modules fusion replace fused modules by nn.Identity?
|
|
1
|
151
|
February 13, 2024
|
|
How can I save a convert_pt2e model?
|
|
1
|
168
|
February 1, 2024
|
|
RuntimeError: promoteTypes with quantized numbers is not handled yet; figure out what the correct rules should be, offending types: QUInt8 Float
|
|
13
|
342
|
February 1, 2024
|
|
RuntimeError: Quantized cudnn conv2d is currenty limited to groups = 1; received groups =16 , during QAT
|
|
2
|
504
|
January 31, 2024
|
|
Inference with own scaling factors
|
|
0
|
143
|
January 21, 2024
|
|
Additional layer in the conv weight after quantization
|
|
0
|
144
|
January 20, 2024
|