import fastai
import fastai.vision
import PIL
import gradio
import matplotlib
import numpy
import pandas
from fastai.vision.all import *
#
# create class
class ADA_SKIN(object):
  #
  # initialize the object
  def __init__(self, name="Wallaby",verbose=True,*args, **kwargs):
    super(ADA_SKIN, self).__init__(*args, **kwargs)
    self.author = "Duc Haba"
    self.name = name
    if (verbose):
      self._ph()
      self._pp("Hello from class", str(self.__class__) + " Class: " + str(self.__class__.__name__))
      self._pp("Code name", self.name)
      self._pp("Author is", self.author)
      self._ph()
    #
    self.article = '<div><h3>Warning:</h3>Do NOT use this for any medical diagnosis.<br>' 
    self.article += 'I am not a dermatologist, and NO dermatologist has endorsed it. '
    self.article += 'This DL model is for my independent research. <br>Please refer to the GPL 3.0 for usage and license.'
    self.article += '<h3>Citation:</h3><ul><li>'
    self.article += 'Author/Dev: Duc Haba, 2022.</li>'
    self.article += '<li><a target="_blank" href="https://linkedin.com/in/duchaba">https://linkedin.com/in/duchaba</a></li>'
    self.article += '<li>The training dataset are from the International Skin Imaging Collaboration (ISIC)</li>'
    self.article += '<li>The Skin Cancer Identification are from 3 seperate datasets.</li>'
    self.article += '<ol>'
    self.article += '<li>https://www.kaggle.com/datasets/surajghuwalewala/ham1000-segmentation-and-classification</li>'
    self.article += '<li>https://www.kaggle.com/datasets/andrewmvd/isic-2019</li>'
    self.article += '<li>https://www.kaggle.com/datasets/jnegrini/skin-lesions-act-keratosis-and-melanoma</li>'        
    self.article += '<ul><li>'
    self.article += 'The Malignant versus Benign dataset</li>'
    self.article += '<ol><li>https://www.kaggle.com/datasets/fanconic/skin-cancer-malignant-vs-benign</li>'
    self.article += '</ol></ul>'
    self.article += '<h3>Articles:</h3><ul>'
    self.article += '<li><a target="_blank" href="https://www.linkedin.com/pulse/skin-cancer-diagnose-using-deep-learning-duc-haba/">'
    self.article += '"Skin Cancer Diagnose"</a> on LinkedIn, on <a target="_blank" href='
    self.article += '"https://www.linkedin.com/pulse/skin-cancer-diagnose-using-deep-learning-duc-haba/">Medium.</a></li>'
    self.article += '</ul>'  
    self.article += '<h3>Example Images: (left to right)</h3><ol>'
    self.article += '<li>Bowen Disease (AKIEC)</li>'
    self.article += '<li>Basal Cell Carcinoma</li>'
    self.article += '<li>Benign Keratosis-like Lesions</li>'
    self.article += '<li>Dermatofibroma</li>'
    self.article += '<li>Melanoma</li>'
    self.article += '<li>Melanocytic Nevi</li>'
    self.article += '<li>Squamous Cell Carcinoma</li>'   
    self.article += '<li>Vascular Lesions</li>' 
    self.article += '<li>Benign</li>' 
    self.article += '<li>Benign 2</li></ol>'         
    self.article += '<h3>Train Result:</h3><ul>'
    self.article += '<li>Skin Cancer Classificaiton: F1-Score, Precision, and Recall Graph</li>'
    self.article += '<li><img src="file/ada_f1_skin.png" alt="F1-Score, Precision, and Recall Graph" width="640"</li>'
    self.article += '<li>Skin Cancer Malignant or Benign: F1-Score, Precision, and Recall Graph</li>'
    self.article += '<li><img src="file/ada_f1_skin_be.png" alt="F1-Score, Precision, and Recall Graph" width="640"</li>'    
    self.article += '</ul>'
    self.article += '<h3>Dev Stack:</h3><ul>'
    self.article += '<li>Jupyter Notebook, Python, Pandas, Matplotlib, Sklearn</li>'
    self.article += '<li>Fast.ai, PyTorch</li>'
    self.article += '</ul>'
    self.article += '<h3>Licenses:</h3><ul>'
    self.article += '<li><a target="_blank" href="https://www.gnu.org/licenses/gpl-3.0.txt">GNU GPL 3.0</a></li>'        
    self.article += '</ul></div>'
    self.examples = ['akiec1.jpg','bcc1.jpg','bkl1.jpg','df1.jpg','mel1.jpg',
      'nevi1.jpg','scc1.jpg','vl1.jpg','benign1.jpg','benign3.jpg']
    self.title = "Skin Cancer Diagnose"
    return
  # 
  # pretty print output name-value line
  def _pp(self, a, b):
    print("%34s : %s" % (str(a), str(b)))
    return
  #
  # pretty print the header or footer lines
  def _ph(self):
    print("-" * 34, ":", "-" * 34)
    return
  # 
  def _predict_image(self,img,cat):
    pred,idx,probs = learn.predict(img)
    return dict(zip(cat, map(float,probs)))
  #
  def _predict_image2(self,img,cat):
    pred,idx,probs = learn2.predict(img)
    return dict(zip(cat, map(float,probs)))
  #  
  def _draw_pred(self,df_pred, df2):
    canvas, pic = matplotlib.pyplot.subplots(1,2, figsize=(12,6))
    ti = df_pred["vocab"].head(3).values
    ti2 = df2["vocab"].head(2).values
    # special case
    #if (matplotlib.__version__) >= "3.5.2":
    try:
      df_pred["pred"].head(3).plot(ax=pic[0],kind="pie",
        cmap="Set2",labels=ti, explode=(0.02,0,0),
        wedgeprops=dict(width=.4),
        normalize=False)
      df2["pred"].head(2).plot(ax=pic[1],kind="pie",
        colors=["cornflowerblue","darkorange"],labels=ti2, explode=(0.02,0),
        wedgeprops=dict(width=.4),
        normalize=False)      
    except:
      df_pred["pred"].head(3).plot(ax=pic[0],kind="pie",
        cmap="Set2",labels=ti, explode=(0.02,0,0),
        wedgeprops=dict(width=.4))
      df2["pred"].head(2).plot(ax=pic[1],kind="pie",
        colors=["cornflowerblue","darkorange"],labels=ti2, explode=(0.02,0),
        wedgeprops=dict(width=.4))
    t = str(ti[0]) + ": " + str(numpy.round(df_pred.head(1).pred.values[0]*100, 2)) + "% Certainty"
    pic[0].set_title(t,fontsize=14.0, fontweight="bold")
    pic[0].axis('off')
    pic[0].legend(ti, loc="lower right",title="Skin Cancers: Top 3")
    #
    k0 = numpy.round(df2.head(1).pred.values[0]*100, 2)
    k1 = numpy.round(df2.tail(1).pred.values[0]*100, 2)
    if (k0 > k1):
      t2 = str(ti2[0]) + ": " + str(k0) + "% Certainty"
    else:
      t2 = str(ti2[1]) + ": " + str(k1) + "% Certainty"
    pic[1].set_title(t2,fontsize=14.0, fontweight="bold")
    pic[1].axis('off')
    pic[1].legend(ti2, loc="lower right",title="Skin Cancers:")
    #
    # # draw circle
    # centre_circle = matplotlib.pyplot.Circle((0, 0), 0.6, fc='white')
    # p = matplotlib.pyplot.gcf()
    # # Adding Circle in Pie chart
    # p.gca().add_artist(centre_circle)
    #
    #p=plt.gcf()
    #p.gca().add_artist(my_circle)
    #
    canvas.tight_layout()
    return canvas
  #
  def predict_donut(self,img):
    d = self._predict_image(img,self.categories)
    df = pandas.DataFrame(d, index=[0])
    df = df.transpose().reset_index()
    df.columns = ["vocab", "pred"]
    df.sort_values("pred", inplace=True,ascending=False, ignore_index=True)
    #
    d2 = self._predict_image2(img,self.categories2)
    df2 = pandas.DataFrame(d2, index=[0])
    df2 = df2.transpose().reset_index()
    df2.columns = ["vocab", "pred"]
    #
    canvas = self._draw_pred(df,df2)
    return canvas
#
maxi = ADA_SKIN(verbose=False)
#
learn = fastai.learner.load_learner('ada_learn_skin_norm2000.pkl')
learn2 = fastai.learner.load_learner('ada_learn_malben.pkl')
maxi.categories = learn.dls.vocab
maxi.categories2 = learn2.dls.vocab
hf_image = gradio.inputs.Image(shape=(192, 192))
hf_label = gradio.outputs.Label()
intf = gradio.Interface(fn=maxi.predict_donut, 
  inputs=hf_image, 
  outputs=["plot"], 
  examples=maxi.examples,
  title=maxi.title,
  live=True,
  article=maxi.article)
intf.launch(inline=False,share=True)