notebook.community

Edit and run



In [1]:

    
import numpy as np
import pandas as pd
import os
import seaborn as sns
import matplotlib.pyplot as plt
from sklearn.metrics import fbeta_score
from sklearn.metrics import roc_curve, auc
from sklearn.metrics import precision_recall_curve
from sklearn.metrics import average_precision_score

%matplotlib inline









    



C:\Users\Lingyu\.conda\envs\tensorflow\lib\site-packages\IPython\html.py:14: ShimWarning: The `IPython.html` package has been deprecated since IPython 4.0. You should import from `notebook` instead. `IPython.html.widgets` has moved to `ipywidgets`.
  "`IPython.html.widgets` has moved to `ipywidgets`.", ShimWarning)



In [2]:

    
WEATHER_LABELS = {'clear': 0,
                  'partly_cloudy': 1,
                  'haze': 2,
                  'cloudy': 3}

COMMON_LABELS = {'primary': 0,
                 'agriculture': 1,
                 'road': 2,
                 'water': 3,
                 'habitation': 4,
                 'cultivation': 5,
                 'bare_ground': 6}

RARE_LABELS = {'selective_logging': 0,
               'artisinal_mine': 1,
               'blooming': 2,
               'slash_burn': 3,
               'blow_down': 4,
               'conventional_mine': 5}
LABELS = [WEATHER_LABELS, COMMON_LABELS, RARE_LABELS]



In [3]:

    
NETWORK_ID = 0



In [4]:

    
EVALUATION_PATH = os.path.join(r'../reports/planet_validation')
TF_RECORD_PATH = os.path.join(r'../data/processed/tfrecord_summary.txt')



In [5]:

    
hashtable = dict([[v,k] for k,v in LABELS[NETWORK_ID].items()])
label_names = list(hashtable.values())



In [6]:

    
label_distribution = np.loadtxt(TF_RECORD_PATH)



In [10]:

    
train_size = label_distribution[-1]
validation_size = int(np.floor(40483 * 0.2))

1. Training set results



In [ ]:

    
labels = np.loadtxt(os.path.join(EVALUATION_PATH, 'out_labels_train.txt'))[0 : train_size, :]
preds = np.loadtxt(os.path.join(EVALUATION_PATH, 'out_predictions_train.txt'))[0 : train_size, :]

2. Validation set results



In [ ]:

    
labels = np.loadtxt(os.path.join(EVALUATION_PATH, 'out_labels_validation.txt'))[0 : train_size, :]
preds = np.loadtxt(os.path.join(EVALUATION_PATH, 'out_predictions_validation.txt'))[0 : train_size, :]