Overview

Because this model is only:
  nn.Linear(2,1)
you can even compute the answer by hand:
  y = w1*x1 + w2*x2 + b
Using your weights:
  y =
  0.0598*1 
  +
  0.0642*2 
  +
  (-0.4107)
  ≈  -0.2225
If M04 prints something close to:
  tensor([[-0.2225]])
then you've followed the entire chain:
  .pt file
    ↓
  weights
    ↓ 
  model
    ↓
  input
    ↓
  output
which is the complete inference path of a neural network. 
After that we can decide whether M05 should be "train a model" or "inspect a larger model."

1 Output

python m04_load_use_model.py 

BEFORE LOAD
Parameter containing:
tensor([[ 0.3151, -0.6782]], requires_grad=True)
Parameter containing:
tensor([-0.6311], requires_grad=True)

AFTER LOAD
Parameter containing:
tensor([[0.0598, 0.0642]], requires_grad=True)
Parameter containing:
tensor([-0.4107], requires_grad=True)

INPUT
tensor([[1., 2.]])

OUTPUT
tensor([[-0.2226]], grad_fn=<AddmmBackward0>)
(venv) terry@LAPTOP-HKPDHF7M:/mnt/c/Users/terry/Downloads/607_predictive$

2 PY scripts

# m04_load_use_model.py

import torch
import torch.nn as nn

# -----------------------
# Model definition
# -----------------------

class TinyModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.fc = nn.Linear(2, 1)

    def forward(self, x):
        return self.fc(x)

# -----------------------
# New model object
# -----------------------

model = TinyModel()

print("BEFORE LOAD")
print(model.fc.weight)
print(model.fc.bias)

# -----------------------
# Load weights from disk
# -----------------------

model.load_state_dict(torch.load("m01_tiny_model.pt"))

print()
print("AFTER LOAD")
print(model.fc.weight)
print(model.fc.bias)

# -----------------------
# use model
# -----------------------

x = torch.tensor([[1.0, 2.0]])

y = model(x)

print()
print("INPUT")
print(x)

print()
print("OUTPUT")
print(y)

4 Code with detailed comments

26.0616 (v1 26.0616)