|
Quantize the CRNN model
|
|
8
|
756
|
March 24, 2022
|
|
Input(activation) quantization
|
|
4
|
203
|
March 20, 2022
|
|
After using dynamic quantization and static quantization, the accuracy of the model decreases a lot
|
|
0
|
140
|
March 19, 2022
|
|
For 4bit quantization
|
|
0
|
135
|
March 19, 2022
|
|
Is bias quantized while doing pytorch static quantization?
|
|
12
|
325
|
March 17, 2022
|
|
Do i need to quantize pretrained vgg16 model before converting to onnx for Tflite?
|
|
3
|
165
|
March 16, 2022
|
|
Issue on quantization aware training of MobileNet
|
|
4
|
330
|
March 16, 2022
|
|
Quantization of CNN with nn.functional.normalize
|
|
5
|
308
|
March 16, 2022
|
|
Qat_swap_modules should consider the non_traceable_module_class
|
|
1
|
130
|
March 15, 2022
|
|
Manual quantization vs prequantized model
|
|
1
|
116
|
March 14, 2022
|
|
Why does the output Tensor shares the same quantization parameter with input
|
|
2
|
154
|
March 14, 2022
|
|
Quantization backend(fbgemm) is installed automatically during pip install?
|
|
1
|
256
|
February 16, 2022
|
|
How is quantization of activations handled in pytorch after QAT?
|
|
8
|
258
|
March 11, 2022
|
|
Dynamic Quantization accuracy loss
|
|
4
|
148
|
March 11, 2022
|
|
Satic Quantization of Inception Resnet Model
|
|
6
|
272
|
March 11, 2022
|
|
Question on skipping quantization on unsupported modules
|
|
7
|
268
|
March 7, 2022
|
|
Error while post training quantizing the model
|
|
4
|
258
|
March 5, 2022
|
|
Why quantized_conv2d layer is much slower than fp32
|
|
1
|
192
|
February 23, 2022
|
|
Fuse_modules more sequence support
|
|
3
|
317
|
February 22, 2022
|
|
Post training quantization using eager api in one part of the model
|
|
4
|
185
|
February 22, 2022
|
|
NNAPI Serializer: dealing with unsupported operation transpose
|
|
1
|
196
|
February 17, 2022
|
|
Calibration of model in post training static quantization using fx api
|
|
5
|
343
|
February 11, 2022
|
|
Extract quantized parameters for use in custom compute unit
|
|
2
|
179
|
February 10, 2022
|
|
Bottleneck on data loading
|
|
2
|
791
|
February 8, 2022
|
|
AssertionError: The only supported dtype for nnq.Embedding is torch.quint8
|
|
3
|
314
|
February 5, 2022
|
|
Is it planned to support nn.Embeddings quantization?
|
|
16
|
1759
|
February 5, 2022
|
|
Quantization on avx
|
|
4
|
316
|
February 4, 2022
|
|
Static quantized int8 inference slower than fp32 original model
|
|
3
|
255
|
February 3, 2022
|
|
How to quantize model have more than one output?
|
|
1
|
252
|
February 3, 2022
|
|
RuntimeError: stride should contain 2 elements for 2D convolution
|
|
1
|
139
|
February 1, 2022
|