import re, os, sys, shelve, time, dill, io
from pickle import PicklingError
from dill import Pickler, Unpickler
shelve.Pickler = Pickler
shelve.Unpickler = Unpickler
from IPython.display import display, Markdown, Math, clear_output, Image
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from plotnine import *
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import MinMaxScaler
base_url = ""
# File Loaders
try:
base_url = "drive/MyDrive/Colab Notebooks/proyecto-final"
uploaded = os.path.join(base_url, "data/sinaica-imputated.pickle.gz")
if(not os.path.isfile(uploaded)):
from google.colab import files
uploaded = files.upload()
except:
base_url = ""
uploaded = "data/sinaica/sinaica-imputated.pickle.gz"
def render_mpl_table(data, col_width=3.0, row_height=0.625, font_size=14,
header_color='#40466e', row_colors=['#f1f1f2', 'w'], edge_color='w',
bbox=[0, 0, 1, 1], header_columns=0,
ax=None, **kwargs):
"""
Taken from https://stackoverflow.com/a/39358722/7323086
"""
if ax is None:
size = (np.array(data.shape[::-1]) + np.array([0, 1])) * np.array([col_width, row_height])
fig, ax = plt.subplots(figsize=size)
ax.axis('off')
mpl_table = ax.table(cellText=data.values, bbox=bbox, colLabels=data.columns, **kwargs)
mpl_table.auto_set_font_size(False)
mpl_table.set_fontsize(font_size)
for k, cell in six.iteritems(mpl_table._cells):
cell.set_edgecolor(edge_color)
if k[0] == 0 or k[1] < header_columns:
cell.set_text_props(weight='bold', color='w', size=font_size*1.05)
cell.set_facecolor(header_color)
else:
cell.set_facecolor(row_colors[k[0]%len(row_colors) ])
plt.show()
#df.dropna(inplace=True)
clear_output()
def performance_plot(history, a=None, b=None,
metrics=["accuracy", "val_accuracy"],
plot_validation=True,
title="Gráficas de Desempeño."):
"""
Prints performance plot from a, to b on a history dict.
Inputs:
history: dict containing "loss" and "accuracy" keys
a: epoch start
b. last epoch
metrics: plot these metrics (train and validation). Always 2.
plot_validation: boolean indicating if validation data should be plotted.
a: from this epoch
b: to this epoch
"""
if a is None:
a = 0
if b is None:
b = len(history['loss'])
a = np.min((a,b))
b = np.max((a,b))
imgrows = (len(metrics) + 1) / 2
imgrows = np.round(imgrows, 0)
imgrows = int(imgrows)
#print(imgrows)
# Plot loss
plt.figure(figsize=(14, 5
*imgrows))
plt.suptitle(title)
plt.subplot(imgrows, 2, 1)
plt.title('Loss')
plt.plot(history['loss'][a:b], label='Training', linewidth=2)
if plot_validation:
plt.plot(history['val_loss'][a:b], label='Validation', linewidth=2)
plt.legend()
plt.xlabel('Epoch')
plt.ylabel(f'Loss')
quantiles = np.quantile(range(a, b),
[.2, .4, .6, .8]).round(0).astype(int)
quantiles = np.insert(quantiles, 0, [a])
quantiles += 1
quantiles = np.append(quantiles, [b-1])
plt.xticks(ticks=quantiles-a,
labels=quantiles)
plt.grid(True)
# Plot accuracy
for i, metric in enumerate(metrics):
#print(f"metric: {metric}, i: {i}")
#print(f"mean metric: {np.mean(history[metric])}")
plt.subplot(imgrows, 2, i+2)
plt.title(metric)
plt.plot(history[metric][a:b], label='Training',
linewidth=2)
if plot_validation:
plt.plot(history["val_" + metric][a:b],
label='Validation', linewidth=2)
plt.legend()
plt.xlabel('Epoch')
plt.ylabel(metric)
#plt.xlim(a, b)
#print(range(0, b-a))
plt.xticks(ticks=quantiles-a,
labels=quantiles)
plt.grid(True)
plt.show()
#render_mpl_table(df.head().applymap(shorten), col_width=5)
sinaica = pd.read_pickle(uploaded)
airdata = pd.read_pickle(os.path.join(base_url, "data/airdata/air-imputated.pickle.gz"))
#sinaica.head()
