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

img = open_image('imgs/cat_example.jpg')
img

img.show()

_,axs = plt.subplots(1,4,figsize=(12,4))
for i,ax in enumerate(axs): img.show(ax=ax, title=f'Copy {i+1}')

img.data.shape

show_doc(Image, title_level=3)

show_doc(open_image)

img = open_image('imgs/cat_example.jpg')
img

show_doc(Image.show, arg_comments ={
    'ax': 'matplotlib.pyplot axes on which show the image',
    'figsize': 'Size of the figure',
    'title': 'Title to display on top of the graph',
    'hide_axis': 'If True, the axis of the graph are hidden',
    'cmap': 'Color map to use',
    'y': 'Potential target to be superposed on the same graph (mask, bounding box, points)'
}, full_name='Image.show')

img.show(figsize=(2, 1), title='Little kitten')

img.show(figsize=(10,5), title='Big kitten')

img.data, img.shape, img.size

show_doc(ImageSegment, title_level=3)

show_doc(open_mask)

open_mask('imgs/mask_example.png')

df = pd.read_csv('imgs/mask_rle_sample.csv')
encoded_str = df.iloc[1]['rle_mask']; 
df[:2]

mask = open_mask_rle(df.iloc[0]['rle_mask'], shape=(1918, 1280)).resize((1,128,128))
mask

show_doc(open_mask_rle)

rle_decode(encoded_str, (1912, 1280)).shape

show_doc(rle_decode)

type(mask)

rle_encode(mask.data)

show_doc(rle_encode)

img = open_image('imgs/car_example.jpg')
mask = open_mask('imgs/mask_example.png')
_,axs = plt.subplots(1,3, figsize=(8,4))
img.show(ax=axs[0], title='no mask')
img.show(ax=axs[1], y=mask, title='masked')
mask.show(ax=axs[2], title='mask only', alpha=1.)

show_doc(ImagePoints, doc_string=False, title_level=3)

img = open_image('imgs/face_example.jpg')
pnts = torch.load('points.pth')
pnts = ImagePoints(FlowField(img.size, pnts))
img.show(y=pnts)

pnts.data[:10]

show_doc(ImageBBox, doc_string=False, title_level=3)

show_doc(ImageBBox.create, arg_comments={
    'bboxes': 'list of bboxes (each of those being four integers with the top, left, bottom, right convention)',
    'h': 'height of the input image',
    'w': 'width of the input image',
    'labels': 'labels of the images',
    'pad_idx': 'padding index that will be used to group the ImageBBox in a batch'
})

img = open_image('imgs/car_bbox.jpg')
bbox = ImageBBox.create(*img.size, [[96, 155, 270, 351]])
img.show(y=bbox)

show_doc(show_image)

show_doc(pil2tensor)

pil2tensor(PIL.Image.open('imgs/cat_example.jpg').convert("RGB"), np.float32).div_(255).size() 

pil2tensor(PIL.Image.open('imgs/cat_example.jpg').convert("I"), np.float32).div_(255).size()

pil2tensor(PIL.Image.open('imgs/mask_example.png').convert("L"), np.float32).div_(255).size()

pil2tensor(np.random.rand(224,224,3).astype(np.float32), np.float32).size()

pil2tensor(PIL.Image.open('imgs/cat_example.jpg'), np.float32).div_(255).size()

pil2tensor(PIL.Image.open('imgs/mask_example.png'), np.float32).div_(255).size()

show_doc(image2np)

show_doc(scale_flow)

show_doc(bb2hw)

show_doc(Image.apply_tfms, arg_comments={
    'tfms': '`Transform` or list of `Transform`',
    'do_resolve': 'if False, the values of random parameters are kept from the last draw',
    'xtra': 'extra arguments to pass to the transforms',
    'size': 'desired target size',
    'mult': 'makes sure the final size is a multiple of mult',
    'resize_method': 'how to get to the final size (crop, pad, squish)',
    'padding_mode': "how to pad the image ('zeros', 'border', 'reflection')"
})

def get_class_ex(): return open_image('imgs/cat_example.jpg')
def get_seg_ex(): return open_image('imgs/car_example.jpg'), open_mask('imgs/mask_example.png')
def get_pnt_ex():
    img = open_image('imgs/face_example.jpg')
    pnts = torch.load('points.pth')
    return img, ImagePoints(FlowField(img.size, pnts))
def get_bb_ex():
    img = open_image('imgs/car_bbox.jpg')
    return img, ImageBBox.create(*img.size, [[96, 155, 270, 351]])

tfms = get_transforms()
_, axs = plt.subplots(2,4,figsize=(12,6))
for ax in axs.flatten():
    img = get_class_ex().apply_tfms(tfms[0], get_class_ex(), size=224)
    img.show(ax=ax)

tfms = get_transforms()
_, axs = plt.subplots(2,4,figsize=(12,6))
for ax in axs.flatten():
    img,mask = get_seg_ex()
    img.apply_tfms(tfms[0], size=224)
    mask.apply_tfms(tfms[0], do_resolve=False, size=224)
    img.show(ax=ax, y=mask)

tfms[0][4]

tfms = get_transforms()
_, axs = plt.subplots(2,4,figsize=(12,6))
for ax in axs.flatten():
    img,pnts = get_pnt_ex()
    img.apply_tfms(tfms[0], size=224)
    pnts.apply_tfms(tfms[0], do_resolve=False, size=224)
    img.show(ax=ax, y=pnts)

tfms = get_transforms()
_, axs = plt.subplots(2,4,figsize=(12,6))
for ax in axs.flatten():
    img,bbox = get_bb_ex()
    img.apply_tfms(tfms[0], size=224)
    bbox.apply_tfms(tfms[0], do_resolve=False, size=224)
    img.show(ax=ax, y=bbox)

show_doc(rand_bool)

rand_bool(0.5, 8)

show_doc(uniform)

uniform(0,1,(8,))

show_doc(uniform_int)

uniform_int(0,2,(8,))

show_doc(log_uniform, doc_string=False)

log_uniform(0.5,2,(8,))

show_doc(ResizeMethod, doc_string=False)

show_doc(Transform, title_level=3, 
         alt_doc_string="Create a `Transform` for `func` and assign it a priority `order`.")

show_doc(RandTransform, title_level=3, doc_string=False)

show_doc(RandTransform.resolve)

show_doc(TfmAffine, title_level=3, doc_string=False)

show_doc(TfmCoord, title_level=3, doc_string=False)

show_doc(TfmLighting, title_level=3, doc_string=False)

show_doc(TfmPixel, title_level=3, doc_string=False)

show_doc(TfmCrop, title_level=3, doc_string=False)

show_doc(logit)

show_doc(logit_)

show_doc(Image.affine, doc_string=False)

show_doc(Image.clone)

show_doc(Image.coord, doc_string=False)

show_doc(Image.lighting, doc_string=False)

show_doc(Image.pixel, doc_string=False)

show_doc(Image.refresh)

show_doc(Image.resize)

show_doc(Image.save)

show_doc(Image.show, full_name='show')

show_doc(Image.show_batch)

show_doc(Image.show_results)

show_doc(FlowField, title_level=3)

show_doc(Image.crop_pad)

show_doc(Image.contrast)

show_doc(Image.brightness)

show_doc(Image.flip_lr)

show_doc(Image.pad)

show_doc(Image.pixel)

show_doc(Image.zoom)

show_doc(Image.dihedral)

show_doc(ImageSegment.refresh)

show_doc(Image.jitter)

show_doc(Image.squish)

show_doc(Image.skew)

show_doc(Image.perspective_warp)

show_doc(Image.zoom_squish)

show_doc(Image.crop)

show_doc(Image.tilt)

show_doc(Image.rotate)

show_doc(ImageSegment.lighting)

show_doc(Image.symmetric_warp)

show_doc(Image.dihedral_affine)

show_doc(ImagePoints.pixel)

show_doc(ImageBBox.clone)

show_doc(ImagePoints.refresh)

show_doc(ImagePoints.coord)

show_doc(Image.set_sample)

show_doc(ImageSegment.show)

show_doc(ImagePoints.show)

show_doc(ImagePoints.clone)

show_doc(ImagePoints.lighting)

show_doc(Transform.calc)

show_doc(Image.flip_affine)

show_doc(ImageBBox.show)

show_doc(ImagePoints.resize)

show_doc(ImagePoints.reconstruct_output)

show_doc(ImageSegment.reconstruct_output)