lotta bug fixes

data analysis/analysis.py

@@ -758,7 +758,7 @@ def optimize_regression(x, y, _range, resolution):#_range in poly regression is
     x_test = []
     y_test = []
 
-    for i in range (0, math.floor(len(x) * 0.4), 1):
+    for i in range (0, math.floor(len(x) * 0.5), 1):
         index = random.randint(0, len(x) - 1)
 
         x_test.append(x[index])

data analysis/data/scores.csv

@@ -0,0 +1 @@
+2022, 21, 23, 39, 50, 89, 97, 191, 213, 233, 236, 272, 289, 308, 310, 314, 317, 329, 355, 428, 436
2011, 1, 25, 34, 106, 112, 132, 146, 167, 184, 223, 256, 267, 304, 337, 339, 370, 411, 460, 485, 494
1101, 7, 11, 55, 81, 92, 144, 151, 157, 165, 174, 177, 220, 241, 312, 320, 341, 343, 347, 419, 473
821374, 6, 9, 11, 28, 31, 194, 200, 201, 246, 298, 340, 351, 355, 364, 388, 395, 414, 422, 429, 442
5, 33, 73, 102, 103, 137, 183, 200, 212, 218, 242, 249, 264, 294, 302, 310, 378, 400, 408, 457, 473

data analysis/superscript.py

@@ -41,6 +41,7 @@ import firebase_admin
 from firebase_admin import credentials
 from firebase_admin import firestore
 import analysis
+import titanlearn
 import visualization
 import os
 import sys
@@ -59,7 +60,7 @@ def titanservice():
     file_list = glob.glob(source_dir + '/*.csv') #supposedly sorts by alphabetical order, skips reading teams.csv because of redundancy
     data = []
     files = [fn for fn in glob.glob('data/*.csv') 
-             if not os.path.basename(fn).startswith('teams')]
+             if not (os.path.basename(fn).startswith('teams'))] #scores will be handled sperately
 
     for i in files:
             data.append(analysis.load_csv(i))
@@ -67,6 +68,7 @@ def titanservice():
     stats = []
     measure_stats = []
     teams = analysis.load_csv("data/teams.csv")
+    scores = analysis.load_csv("data/scores.csv")
 
     end = time.time()
 
@@ -112,18 +114,81 @@ def titanservice():
 
                     #print(r2best_curve)
 
-                    measure_stats.append(teams[i] + ["|"] +  list(analysis.basic_stats(line, 0, 0)) + ["|"] + list(analysis.histo_analysis(line, 1, -3, 3)) + ["|"] + ofbest_curve + ["|"] + r2best_curve)
+                    
+                    measure_stats.append(teams[i] + list(analysis.basic_stats(line, 0, 0)) + list(analysis.histo_analysis(line, 1, -3, 3)) + ofbest_curve + r2best_curve)
 
             stats.append(list(measure_stats))
+            nishant = []
+            for i in range(len(scores)):
+
+                    ofbest_curve = [None]
+                    r2best_curve = [None]
+
+                    line = measure[i]
+
+                    #print(line)
+
+                    x = list(range(len(line)))
+                    eqs, rmss, r2s, overfit = analysis.optimize_regression(x, line, 10, 1)
+
+                    beqs, brmss, br2s, boverfit = analysis.select_best_regression(eqs, rmss, r2s, overfit, "min_overfit")
+
+                    #print(eqs, rmss, r2s, overfit)
+                    
+                    ofbest_curve.append(beqs)
+                    ofbest_curve.append(brmss)
+                    ofbest_curve.append(br2s)
+                    ofbest_curve.append(boverfit)
+                    ofbest_curve.pop(0)
+
+                    #print(ofbest_curve)
+
+                    beqs, brmss, br2s, boverfit = analysis.select_best_regression(eqs, rmss, r2s, overfit, "max_r2s")
+
+                    r2best_curve.append(beqs)
+                    r2best_curve.append(brmss)
+                    r2best_curve.append(br2s)
+                    r2best_curve.append(boverfit)
+                    r2best_curve.pop(0)
+
+                    #print(r2best_curve)
+                    
+                    z = len(scores[0]) + 1
+                    nis_num = []
+
+                    nis_num.append(eval(str(ofbest_curve[0])))
+                    nis_num.append(eval(str(r2best_curve[0])))
+
+                    nis_num.append((eval(ofbest_curve[0]) + eval(r2best_curve[0])) / 2)
+
+                    nishant.append(teams[i] + nis_num)
                 
     json_out = {}
+    score_out = {}
 
-    for i in range(len(stats)):
+    #print(stats)
-            json_out[files[i]]=str(stats[i])
                     
-    #print(json_out)
+    for i in range(len(teams)):
+            json_out[str(teams[i][0])] = (stats[0][i])
 
-    db.collection(u'stats').document(u'stats-noNN').set(json_out)
+    for i in range(len(teams)):
+            score_out[str(teams[i][0])] = (nishant[i])
+
+    print(json_out)
+
+    #print(json_out.get('5'))
+
+    location = db.collection(u'stats').document(u'stats-noNN')
+    for i in range(len(teams)):
+        general_general_stats = location.collection(teams[i][0])
+        for j in range(len(files)):
+            general_general_stats.document(files[j]).set({'stats':json_out.get(teams[i][0])})
+
+    for i in range(len(teams)):
+        nnum = location.collection(teams[i][0]).document(u'nishant_number').set({'nishant':score_out.get(teams[i][0])})
+
+    #general_general_stats.collection().document('stats').set()
+    #db.collection(u'stats').document(u'stats-noNN').set(score_out)
 
 def pulldata():
     #TODO
@@ -173,3 +238,4 @@ firebase_admin.initialize_app(cred)
 db = firestore.client()
 
 service() #finally we write something that isn't a function definition
+#titanservice()

data analysis/titanlearn.py

@@ -198,4 +198,4 @@ def retyuoipufdyu():
     model = linear_nn(8, 100, 1, 20, act_fn = "relu")
     print(model)
     return train_sgd_simple(model,"regression", data, ground, learnrate=1e-4, iters=1000)
-retyuoipufdyu()
+#retyuoipufdyu()