pyctb 0.0.3

Python Customized Tracebacks

Python Customized Tracebacks

Installation

pip install pyctb

Synopsis

pyctb is tool for customizing Python traceback information. For each traceback frame, pyctb adds extra lines of the form var = render(value) for each variable in the currently executing function of that frame. By default, render simply calls str(value). However, the user may register sub-functions for rendering specific types of their choice.

You may render a new function to render instances of a class with:

pyctb.add(cls, func)

Objects are rendered using the most specific (in method-resolution order) rendering function registered. For example:

pyctb.add(Tabby, render_tabby)
pyctb.add(Cat, render_cat)
pyctb.add(Animal, render_animal)

Then, a Tabby will be rendered with render_tabby. A Persian will use render_cat, and Dog will use render_animal, and so forth. By default, object is registered with str.

import torch
import pyctb

# show custom render functions for types
pyctb.inventory()
tf:   Render a tf.Tensor, tf.Variable, or EagerTensor
torch:   Render a torch.Tensor or torch.nn.parameter.Parameter

# add the torch render group
pyctb.add_group('torch')

conv = torch.nn.Conv1d(5, 5, 10)
inp = torch.zeros([5, 100])

# This will produce an exception and the default traceback (see below)
conv(inp)

# Turn on the custom traceback
pyctb.on()

# This will produce the same exception but display a custom traceback
# showing argument values
conv(inp)

pyctb.off()

The tracebacks are as follows:

# The default traceback:
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/henry/miniconda3/envs/pytorch181/lib/python3.9/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/home/henry/miniconda3/envs/pytorch181/lib/python3.9/site-packages/torch/nn/modules/conv.py", line 263, in forward
    return self._conv_forward(input, self.weight, self.bias)
  File "/home/henry/miniconda3/envs/pytorch181/lib/python3.9/site-packages/torch/nn/modules/conv.py", line 259, in _conv_forward
    return F.conv1d(input, weight, bias, self.stride,
RuntimeError: Expected 3-dimensional input for 3-dimensional weight [5, 5, 10], but got 2-dimensional input of size [5, 100] instead

# The custom traceback:
Custom Traceback (most recent call last):
  <module>()
  File "<stdin>", line 1, in <module>
  torch.nn.modules.module._call_impl(self=Conv1d(5, 5, kernel_size=(10,), stride=(1,)), input[0]=[5,100]:float32:cpu)
  File "/home/henry/miniconda3/envs/pytorch181/lib/python3.9/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  torch.nn.modules.conv.forward(self=Conv1d(5, 5, kernel_size=(10,), stride=(1,)), input=[5,100]:float32:cpu)
  File "/home/henry/miniconda3/envs/pytorch181/lib/python3.9/site-packages/torch/nn/modules/conv.py", line 263, in forward
    return self._conv_forward(input, self.weight, self.bias)
  torch.nn.modules.conv._conv_forward(self=Conv1d(5, 5, kernel_size=(10,), stride=(1,)), input=[5,100]:float32:cpu, weight=[5,5,10]:float32:cpu, bias=[5]:float32:cpu)
  File "/home/henry/miniconda3/envs/pytorch181/lib/python3.9/site-packages/torch/nn/modules/conv.py", line 259, in _conv_forward
    return F.conv1d(input, weight, bias, self.stride,
RuntimeError: Expected 3-dimensional input for 3-dimensional weight [5, 5, 10], but got 2-dimensional input of size [5, 100] instead