#!/usr/bin/env python
# coding: utf-8

# ## Computer Vision Interpret

# [`vision.interpret`](/vision.interpret.html#vision.interpret) is the module that implements custom [`Interpretation`](/train.html#Interpretation) classes for different vision tasks by inheriting from it.

# In[1]:


from fastai.gen_doc.nbdoc import *
from fastai.vision import *
from fastai.vision.interpret import *


# In[2]:


show_doc(SegmentationInterpretation)


# In[3]:


show_doc(SegmentationInterpretation.top_losses)


# In[4]:


show_doc(SegmentationInterpretation._interp_show)


# In[5]:


show_doc(SegmentationInterpretation.show_xyz)


# In[6]:


show_doc(SegmentationInterpretation._generate_confusion)


# In[7]:


show_doc(SegmentationInterpretation._plot_intersect_cm)


# Let's show how [`SegmentationInterpretation`](/vision.interpret.html#SegmentationInterpretation) can be used once we train a segmentation model.

# ### train 

# In[ ]:


camvid = untar_data(URLs.CAMVID_TINY)
path_lbl = camvid/'labels'
path_img = camvid/'images'


# In[ ]:


codes = np.loadtxt(camvid/'codes.txt', dtype=str)
get_y_fn = lambda x: path_lbl/f'{x.stem}_P{x.suffix}'


# In[ ]:


data = (SegmentationItemList.from_folder(path_img)
        .split_by_rand_pct()
        .label_from_func(get_y_fn, classes=codes)
        .transform(get_transforms(), tfm_y=True, size=128)
        .databunch(bs=16, path=camvid)
        .normalize(imagenet_stats))


# In[ ]:


data.show_batch(rows=2, figsize=(7,5))


# In[ ]:


learn = unet_learner(data, models.resnet18)
learn.fit_one_cycle(3,1e-2)
learn.save('mini_train')


# In[ ]:


jekyll_warn("Following results will not make much sense with this underperforming model but functionality will be explained with ease")


# ### interpret

# In[ ]:


interp = SegmentationInterpretation.from_learner(learn)


# Since `FlattenedLoss of CrossEntropyLoss()` is used we reshape and then take the mean of pixel losses per image. In order to do so we need to pass `sizes:tuple` to `top_losses()`

# In[ ]:


top_losses, top_idxs = interp.top_losses(sizes=(128,128))


# In[ ]:


top_losses, top_idxs


# In[ ]:


plt.hist(to_np(top_losses), bins=20);plt.title("Loss Distribution");


# Next, we can generate a confusion matrix similar to what we usually have for classification. Two confusion matrices are generated: `mean_cm` which represents the global label performance and `single_img_cm` which represents the same thing but for each individual image in dataset.
# 
# Values in the matrix are calculated as:
# 
# \begin{align}
# \ CM_{ij} & = IOU(Predicted , True | True) \\
# \end{align}
# 
# Or in plain english: ratio of pixels of predicted label given the true pixels

# In[ ]:


learn.data.classes


# In[ ]:


mean_cm, single_img_cm = interp._generate_confusion()


# In[ ]:


mean_cm.shape, single_img_cm.shape


# `_plot_intersect_cm` first displays a dataframe showing per class score using the IOU definition we made earlier. These are the diagonal values from the confusion matrix which is displayed after. 
# 
# `NaN` indicate that these labels were not present in our dataset, in this case validation set. As you can imagine it also helps you to maybe construct a better representing validation set.

# In[ ]:


df = interp._plot_intersect_cm(mean_cm, "Mean of Ratio of Intersection given True Label")


# Next let's look at the single worst prediction in our dataset. It looks like this dummy model just predicts everything as `Road` :)

# In[ ]:


i = top_idxs[0]
df = interp._plot_intersect_cm(single_img_cm[i], f"Ratio of Intersection given True Label, Image:{i}")


# Finally we will visually inspect this single prediction

# In[ ]:


interp.show_xyz(i, sz=15)


# In[ ]:


jekyll_warn("""With matplotlib colormaps the max number of unique qualitative colors is 20. 
So if len(classes) > 20 then close class indexes may be plotted with the same color.
Let's fix this together :)""")


# In[ ]:


interp.c2i


# In[8]:


show_doc(ObjectDetectionInterpretation)


# In[ ]:


jekyll_warn("ObjectDetectionInterpretation is not implemented yet. Feel free to implement it :)")


# ## Undocumented Methods - Methods moved below this line will intentionally be hidden

# ## New Methods - Please document or move to the undocumented section