[StaticQuant] add a linear observer class and test

[drisspg]

Stacked PRs:

• ->[StaticQuant] add a linear observer class and test #807

---

[StaticQuant] add a linear observer class and test

This adds a new flow for calculating static activation scales. Rough example of flow:

import torch
import torch.nn as nn
from torchao.quantization.linear_observer_tensor import insert_observers_
from torchao.quantization.observer import (
    AffineQuantizedMinMaxObserver,
    PerTensor,
    MappingType
)

# Create observers
input_observer = AffineQuantizedMinMaxObserver(
    MappingType.SYMMETRIC,
    torch.float8_e4m3fn,
    granularity_type=PerTensor(),
    eps=torch.finfo(torch.float32).eps,
    scale_dtype=torch.float,
    zero_point_dtype=torch.int,
    zero_point_domain=None,
)

# Create a linear module
linear_module = nn.Linear(10, 20)

# Convert the linear module's weight to an observed tensor
insert_observers_(linear_module, input_observer, weight_observer=None)

# The linear_module can now be used as usual, with observers calculating statistics
output = linear_module(torch.randn(10, 10))

# Get the scale and zero point of the input observer
scale, zero_point = linear_module.weight.input_observer.calculate_qparams()

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/807

• 📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit 96fe56a with merge base a1b3e67 ():
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

[jerryzh168]

LG, see some comments inline, please fix before landing