nevikw39/StatisticalThinkingAnalysis

GEC 111503

Statistical Thinking & Analysis

GEC 111503

6/3 23:00

Maximum binary accuracy on validation data set (random sample (20%) from train data set):

• model
  • 0.7380
• model_pearson
  • 0.7166
• model_chi_square
  • 0.7368
• model_anova
  • 0.7726

6/4 02:00

Split data into train, validation, test set.

Maximum binary accuracy during validation & test evaluation:

• No selection

  • {'loss': 0.504382848739624, 'binary_accuracy': 0.7546666860580444, 'precision': 0.7523117661476135, 'recall': 0.7593333125114441}
    

• Pearson

  • {'loss': 0.5367627143859863, 'binary_accuracy': 0.7353333234786987, 'precision_1': 0.7447988986968994, 'recall_1': 0.7160000205039978}
    

• Chi-square

  • {'loss': 0.5034903883934021, 'binary_accuracy': 0.7526666522026062, 'precision_2': 0.7516600489616394, 'recall_2': 0.7546666860580444}
    

• ANOVA

  • {'loss': 0.5044726133346558, 'binary_accuracy': 0.75, 'precision_3': 0.7600554823875427, 'recall_3': 0.7306666374206543}
    

6/4 16:00

Daisy.

Maximum binary accuracy during validation & test evaluation:

• No selection

  • {'loss': 0.5344247817993164, 'binary_accuracy': 0.7273333072662354, 'precision': 0.7246376872062683, 'recall': 0.7333333492279053}
    

• Pearson

  • {'loss': 0.5363144874572754, 'binary_accuracy': 0.734333336353302, 'precision_1': 0.7226092219352722, 'recall_1': 0.7606666684150696}
    

• Chi-square

  • {'loss': 0.5361865758895874, 'binary_accuracy': 0.7293333411216736, 'precision_2': 0.7362637519836426, 'recall_2': 0.7146666646003723}
    

• ANOVA

  • {'loss': 0.5361486077308655, 'binary_accuracy': 0.7336666584014893, 'precision_3': 0.7228226065635681, 'recall_3': 0.7580000162124634}