'''

TRABAJO FINALIZADO GENTEEEEEEEEEEEEEEEEEEEEE

'''

Aprendizaje_Automtico_Proyecto

data_analysis -> En este archvio tenemos varias gráficas sobre las relaciones de algunas características. Así como cierta información relativa a los datos, como el número de datos nulos, los percentiles...

data_preprocessing -> Hecho

train_students_prep_minmax -> Hecho

train_students_prep_standard -> Hecho

decisive_model -> Hecho

APS_Solver -> Hecho

characteristics_selection.py -> Acabado

kfold -> Acabado

dt_metrics.csv -> Acabado

knn_metrics.csv -> Acabado

Logistic_Regresion -> Acabado

LR_minmax_scaler.csv -> Acabado

LR_standard_scaler.csv -> Acabado

Prueba red neuronal

• 3 capas con 11 nodos

  • relu: Neuronal Network accuracy: [np.float64(88.95399201116511)] Neuronal Network f1: [np.float64(88.90245642505784)] Neuronal Network precision: [np.float64(88.99894942961589)] Neuronal Network recall: [np.float64(88.95399201116511)] Neuronal Network error: [np.float64(11.046007988834884)] Neuronal Network train: 0.891042909370791 Neuronal Network test: 0.8895399201116512

  • tanh: Neuronal Network accuracy: [88.98647673131525] Neuronal Network f1: [88.94581358793175] Neuronal Network precision: [89.00462180058902] Neuronal Network recall: [88.98647673131525] Neuronal Network error: [11.013523268684732] Neuronal Network train: 0.8951799114873966 Neuronal Network test: 0.8898647673131526

  • logistic: Training loss did not improve more than tol=0.000010 for 10 consecutive epochs. Stopping. Neuronal Network accuracy: [np.float64(88.36445449732902)] Neuronal Network f1: [np.float64(88.28923195439617)] Neuronal Network precision: [np.float64(88.46387476767313)] Neuronal Network recall: [np.float64(88.36445449732902)] Neuronal Network error: [np.float64(11.635545502670963)] Neuronal Network train: 0.8802193573215316 Neuronal Network test: 0.8836445449732903

• 3 capas con 10 nodos

  • relu: Neuronal Network accuracy: [88.90466336204823] Neuronal Network f1: [88.85123940949902] Neuronal Network precision: [88.95332697871112] Neuronal Network recall: [88.90466336204823] Neuronal Network error: [11.09533663795178] Neuronal Network train: 0.8930633057533193 Neuronal Network test: 0.8890466336204822

  • tanh: Neuronal Network accuracy: [89.00332066028203] Neuronal Network f1: [88.96703204059675] Neuronal Network precision: [89.0119708378856] Neuronal Network recall: [89.00332066028203] Neuronal Network error: [10.996679339717986] Neuronal Network train: 0.8955166442178178 Neuronal Network test: 0.8900332066028203

  • logistic: Neuronal Network accuracy: [86.05683622888493] Neuronal Network f1: [85.9173603108546] Neuronal Network precision: [86.25693997650237] Neuronal Network recall: [86.05683622888493] Neuronal Network error: [13.943163771115064] Neuronal Network train: 0.859341928035405 Neuronal Network test: 0.8605683622888494

• 3 capas con 9 nodos

  • relu: Neuronal Network accuracy: [88.96000770008182] Neuronal Network f1: [88.91988801266486] Neuronal Network precision: [88.97495656823577] Neuronal Network recall: [88.96000770008182] Neuronal Network error: [11.03999229991819] Neuronal Network train: 0.8876274773908024 Neuronal Network test: 0.8896000770008181

  • tanh: Neuronal Network accuracy: [np.float64(88.94917946003176)] Neuronal Network f1: [np.float64(88.90208757640549)] Neuronal Network precision: [np.float64(88.98298288661618)] Neuronal Network recall: [np.float64(88.94917946003176)] Neuronal Network error: [np.float64(11.050820539968242)] Neuronal Network train: 0.8932076197806428 Neuronal Network test: 0.8894917946003176

  • logistic: Neuronal Network accuracy: [86.15669666490206] Neuronal Network f1: [86.00004830633029] Neuronal Network precision: [86.44103378681633] Neuronal Network recall: [86.15669666490206] Neuronal Network error: [13.843303335097934] Neuronal Network train: 0.859245718683856 Neuronal Network test: 0.8615669666490205

• Con 3 capas, 10, 11, 9: Neuronal Network accuracy: [89.11039992299919] Neuronal Network f1: [89.07484092854784] Neuronal Network precision: [89.11634996830695] Neuronal Network recall: [89.11039992299919] Neuronal Network error: [10.889600077000816] Neuronal Network train: 0.891956898210506 Neuronal Network test: 0.891103999229992

• Con 2 capas, 10, 10 Neuronal Network accuracy: [89.03700851821552] Neuronal Network f1: [89.00813942181888] Neuronal Network precision: [89.03287110840171] Neuronal Network recall: [89.03700851821552] Neuronal Network error: [10.96299148178449] Neuronal Network train: 0.8897440831248797 Neuronal Network test: 0.8903700851821551

• Con 3 capas, 100, 100, 50: Training loss did not improve more than tol=0.000010 for 10 consecutive epochs. Stopping. Neuronal Network accuracy: [np.float64(87.24072380769046)] Neuronal Network f1: [np.float64(87.23797224333165)] Neuronal Network precision: [np.float64(87.2356732288661)] Neuronal Network recall: [np.float64(87.24072380769046)] Neuronal Network error: [np.float64(12.759276192309542)] Neuronal Network train: 0.935876467192611 Neuronal Network test: 0.8724072380769046

• Con 3 capas de 20: Neuronal Network accuracy: [np.float64(89.21386977236632)] Neuronal Network f1: [np.float64(89.20176854133676)] Neuronal Network precision: [np.float64(89.2011430785718)] Neuronal Network recall: [np.float64(89.21386977236632)] Neuronal Network error: [np.float64(10.786130227633672)] Neuronal Network train: 0.8969116798152781 Neuronal Network test: 0.8921386977236633

• Con 2 capas 20, 20: Neuronal Network accuracy: [89.10869649990032] Neuronal Network f1: [89.06915790921431] Neuronal Network precision: [89.1258106982782] Neuronal Network recall: [89.10869649990032] Neuronal Network error: [10.891303500099689] Neuronal Network train: 0.8941042174107061 Neuronal Network test: 0.8910869649990032

• Con 3 capas de 15: Neuronal Network accuracy: [np.float64(88.9840704557486)] Neuronal Network f1: [np.float64(88.94496550921535)] Neuronal Network precision: [np.float64(88.99804956058291)] Neuronal Network recall: [np.float64(88.9840704557486)] Neuronal Network error: [np.float64(11.015929544251414)] Neuronal Network train: 0.8949393881085241 Neuronal Network test: 0.8898407045574859

• Con 4 capas de 20 Neuronal Network accuracy: [89.12914429294759] Neuronal Network f1: [89.09833278602679] Neuronal Network precision: [89.13749573461098] Neuronal Network recall: [89.12914429294759] Neuronal Network error: [10.870855707052414] Neuronal Network train: 0.8978573280554979 Neuronal Network test: 0.8912914429294758

• Con 3 capas de 30: Neuronal Network accuracy: [np.float64(89.1079936474325)] Neuronal Network f1: [np.float64(89.081414404818)] Neuronal Network precision: [np.float64(89.10420317685673)] Neuronal Network recall: [np.float64(89.1079936474325)] Neuronal Network error: [np.float64(10.892006352567495)] Neuronal Network train: 0.9050894746969405 Neuronal Network test: 0.8910799364743249

• Con 3 capas de 40: Neuronal Network accuracy: [89.01341879941684] Neuronal Network f1: [88.98139205284865] Neuronal Network precision: [89.01847528367902] Neuronal Network recall: [89.01341879941684] Neuronal Network error: [10.98658120058315] Neuronal Network train: 0.9062306461385601 Neuronal Network test: 0.8901341879941684

• Con 3 capas de 25: Neuronal Network accuracy: [0.8922322242489615] Neuronal Network f1: [0.8919248909661142] Neuronal Network precision: [0.8923072696375229] Neuronal Network recall: [0.8922322242489615] Neuronal Network error: [0.10776777575103843] Neuronal Network train: 0.9011393545808686 Neuronal Network test: 0.8922322242489615

• Con 3 capas de 25,25,24: Neuronal Network accuracy: [89.30262425590904] Neuronal Network f1: [89.27336750390413] Neuronal Network precision: [89.30388279532174] Neuronal Network recall: [89.30262425590904] Neuronal Network error: [10.697375744090968] Neuronal Network train: 0.9002153239277051 Neuronal Network test: 0.8930262425590904