Merge pull request #7 from Unilasalle-SmartCare/contrucao-modelo

Construção e representação do Modelo

.gitattributes

@@ -0,0 +1 @@
+*.npy filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,2 @@
+logs/
+modelo/modelo.old.py

README.md

@@ -1,11 +1,9 @@
 # Smartcare Machine Learning
-## Modelo
-Local onde se encontrará a parte mais técnica referente a deep learning, a construção e treinamento das redes neurais artificiais recorrentes (RNN). Confira [clicando aqui](https://github.com/Unilasalle-SmartCare/smartcare-machinelearning/tree/master/modelo)
-
-## Servidor
-Local onde são carregadas as informações do modelo já treinado, será como uma api que ao receber requisições com dados novos retornará a previsão se o idoso está estressado ou estará estressado. Confira [clicando aqui](https://github.com/Unilasalle-SmartCare/smartcare-machinelearning/tree/master/server)
-
-## Lançamentos
-Visando uma maior facilidade em juntar as features e fixes desenvolvidas no projeto, você pode ver os lançamentos (releases)  [clicando aqui](https://github.com/Unilasalle-SmartCare/smartcare-machinelearning/releases).
+## Model
+Where the technical part of the deep learning is, the build and train of the convolutional neural network (CNN). Check it out by [clicking here](https://github.com/Unilasalle-SmartCare/smartcare-machinelearning/tree/master/modelo)
 
+## Server
+It is planned in the future to have an api server with the model running (Flask). Check it out by [clicking here](https://github.com/Unilasalle-SmartCare/smartcare-machinelearning/tree/master/server)
 
+## Releases
+In order to make it easier to join the features and fixes developed in the project, it was grouped with releases, you can see them by [clicking here](https://github.com/Unilasalle-SmartCare/smartcare-machinelearning/releases).

modelo/DataProcessing.py

@@ -0,0 +1,98 @@
+#!/usr/bin/env python
+# coding: utf-8
+
+# Abrir python 3.5
+# In[2]:
+
+
+import pandas as pd
+import json
+import joblib
+from sklearn.preprocessing import MinMaxScaler
+import numpy as np
+import matplotlib.pyplot as plt
+
+
+# In[4]:
+
+
+path = "dados/eval.json" # Caminho do dataset vindo do gerador
+# Carregamento Dataset Pandas
+with open(path , 'r') as f:
+    data = json.load(f)
+    data = json.loads(data)
+
+
+# In[5]:
+
+
+df = pd.DataFrame.from_dict(pd.io.json.json_normalize(data), orient='columns')
+##
+df["stress"] = df["stress"].astype(int) # True e False para 1 e 0
+
+
+# In[7]:
+
+
+
+# https://stackoverflow.com/questions/31638651/how-can-i-draw-lines-into-numpy-arrays
+def naive_line(r0, c0, r1, c1):
+    # The algorithm below works fine if c1 >= c0 and c1-c0 >= abs(r1-r0).
+    # If either of these cases are violated, do some switches.
+    if abs(c1-c0) < abs(r1-r0):
+        # Switch x and y, and switch again when returning.
+        xx, yy, val = naive_line(c0, r0, c1, r1)
+        return (yy, xx, val)
+
+    # At this point we know that the distance in columns (x) is greater
+    # than that in rows (y). Possibly one more switch if c0 > c1.
+    if c0 > c1:
+        return naive_line(r1, c1, r0, c0)
+
+    # We write y as a function of x, because the slope is always <= 1
+    # (in absolute value)
+    x = np.arange(c0, c1+1, dtype=float)
+    y = x * (r1-r0) / (c1-c0) + (c1*r0-c0*r1) / (c1-c0)
+
+    valbot = np.floor(y)-y+1
+    valtop = y-np.floor(y)
+
+    return (np.concatenate((np.floor(y), np.floor(y)+1)).astype(int), np.concatenate((x,x)).astype(int),
+            np.concatenate((valbot, valtop)))
+
+
+# In[50]:
+
+
+# split the array into daily paths
+x = np.split(np.array(df)[:, 1:4].astype("int"), np.cumsum(np.unique(np.array(df)[:, 0], return_counts=True)[1])[:-1])
+# create a list of 2D arrays of 602x369
+pathImgs = np.zeros((len(x),602,369))
+
+y = df.groupby(by=["date"]).first()["stress"].values
+
+# In[51]:
+
+
+# For each 2D array, paint the path that was made in that day.
+for pathIdx,currPath in enumerate(x):
+    prevRow =  currPath[0]
+    for i,row in enumerate(currPath):
+        pathImgs[pathIdx,row[1],row[2]] = 1
+        if i > 0:
+            #draw a line between the previous xy point and the current xy point
+            rr, cc, _ = naive_line(prevRow[1],prevRow[2], row[1], row[2])
+            ## rr and cc are a lists of indexes that defines where in the x,y axis we our line is drawn
+            pathImgs[pathIdx,rr, cc] = 1
+        prevRow = row
+
+
+# In[59]:
+np.save("dados/x-validacaoo.npy", pathImgs)
+np.save("dados/y-validacaoo.npy", y)
+
+for i,img in enumerate(pathImgs):
+    plt.title("") 
+    plt.imshow(img.T, cmap="gray")
+    plt.show()
+