Examples.torch.quantize ERROR

[bqFirst]

Issue Description:

When I run python -m Examples.torch.quantize, I get the following error:

RuntimeError: Failed to run QuantizeDequantize since quantization parameters are not initialized. Please initialize the quantization parameters using compute_encodings().

After I call compute_encodings() before _prefill_inputs, the problem disappears.

Is this a bug?

[ananyamukh6]

Hi, Can you please share the command you used to run and what quantization technique you used

[bqFirst]

I used "pcq_spinquant_adascale" and the command is:

python -m Examples.torch.quantize \
 --model-id "Qwen/Qwen3-1.7B" \
 --recipe "pcq_spinquant_adascale" \
 --export-path "./torch_pcq" \
 --adascale-num-batches 128 --adascale-num-iterations 2048

[bqFirst]

I used "pcq_spinquant_adascale" and the command is:

python -m Examples.torch.quantize \
 --model-id "Qwen/Qwen3-1.7B" \
 --recipe "pcq_spinquant_adascale" \
 --export-path "./torch_pcq" \
 --adascale-num-batches 128 --adascale-num-iterations 2048