#!/usr/bin/env python
# coding: utf-8
#
# Author: Kazuto Nakashima
# URL: http://kazuto1011.github.io
# Created: 2017-05-26
from collections import Sequence
import numpy as np
import torch
import torch.nn as nn
from torch.nn import functional as F
from tqdm import tqdm
[docs]class BaseWrapper(object):
def __init__(self, model):
super(BaseWrapper, self).__init__()
self.device = next(model.parameters()).device
self.model_with_norm = model
self.model = model.model
self.handlers = [] # a set of hook function handlers
def _encode_one_hot(self, ids):
one_hot = torch.zeros_like(self.logits).to(self.device)
one_hot.scatter_(1, ids, 1.0)
return one_hot
[docs] def forward(self, image):
self.image_shape = image.shape[2:]
self.logits = self.model_with_norm(image)
self.probs = torch.sigmoid(self.logits)
return self.probs.sort(dim=1, descending=True) # ordered results
[docs] def backward(self, ids):
"""
Class-specific backpropagation
"""
one_hot = self._encode_one_hot(ids)
self.model_with_norm.zero_grad()
self.logits.backward(gradient=one_hot, retain_graph=True)
[docs] def generate(self):
raise NotImplementedError
[docs] def remove_hook(self):
"""
Remove all the forward/backward hook functions
"""
for handle in self.handlers:
handle.remove()
[docs]class GradCAM(BaseWrapper):
"""
"Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization"
https://arxiv.org/pdf/1610.02391.pdf
Look at Figure 2 on page 4
"""
def __init__(self, model, candidate_layers=None):
super(GradCAM, self).__init__(model)
self.fmap_pool = {}
self.grad_pool = {}
self.candidate_layers = candidate_layers # list
def save_fmaps(key):
def forward_hook(module, input, output):
self.fmap_pool[key] = output.detach()
return forward_hook
def save_grads(key):
def backward_hook(module, grad_in, grad_out):
self.grad_pool[key] = grad_out[0].detach()
return backward_hook
# If any candidates are not specified, the hook is registered to all the layers.
for name, module in self.model.named_modules():
if self.candidate_layers is None or name in self.candidate_layers:
self.handlers.append(module.register_forward_hook(save_fmaps(name)))
self.handlers.append(module.register_backward_hook(save_grads(name)))
def _find(self, pool, target_layer):
if target_layer in pool.keys():
return pool[target_layer]
else:
raise ValueError("Invalid layer name: {}".format(target_layer))
[docs] def generate(self, target_layer):
fmaps = self._find(self.fmap_pool, target_layer)
grads = self._find(self.grad_pool, target_layer)
weights = F.adaptive_avg_pool2d(grads, 1)
gcam = torch.mul(fmaps, weights).sum(dim=1, keepdim=True)
gcam = F.relu(gcam)
gcam = F.interpolate(
gcam, self.image_shape, mode="bilinear", align_corners=False
)
B, C, H, W = gcam.shape
gcam = gcam.view(B, -1)
gcam -= gcam.min(dim=1, keepdim=True)[0]
gcam /= gcam.max(dim=1, keepdim=True)[0]
gcam = gcam.view(B, C, H, W)
return gcam