IBM Coursera Advanced Data Science Capstone – MARCUS WIMALAJEEWA

Note: I incorrectly rounded up on a couple of figures in my audio, where inconsistent, numbers are correct in the visuals. Note: I broke the GridSearch into smaller GridSearches because of session timeout issues on Colab! Notebook links below. Timestamps: 0:00 Introduction and Introduction to Dataset 0:51 Word Frequency Visualization by Class 1:04 Use Case Explanation 1:45 My Solution - Use of Keras 2:00 Multiclass to Binary Classification 2:25 Peer Presentation - Architectural Choices 3:04 Data Quality and Pre-processing 3:44 Stopword removal and Lemmatization (Pre-processing) 4:15 Keras One_Hot and Sequence Padding (Pre-processing) 4:30 Synthesizing new data (Pre-processing) 5:20 Performance Indicators 6:01 ML vs DL 6:17 DL Model Layers 6:45 GridSearch 6:58 Model Evaluation 7:44 Model Deployment Introduction 8:05 Deployment Batch Discussion 8:40 Conclusion Data Exploration: https://gist.github.com/marcusmw/fdcc7af2d9975a2a662421bbd8f6bd6d Extract Transform Load: https://gist.github.com/marcusmw/91b9824c35346d098dea2e763c8f2307 Feature Engineering: https://gist.github.com/marcusmw/607ab3863aa17cc724cdbbb6e2712bd4 Model Definition: https://gist.github.com/marcusmw/3e9775f1a93216b0be05b6f179c67cc9 Model Training: https://gist.github.com/marcusmw/a41902931cf03fe77945cc071bfaca95 Model Evaluation: https://gist.github.com/marcusmw/9aa149292073db66f5613b997f5b46ee Model Deployment: https://gist.github.com/marcusmw/aa3bd73b530f7e576f3d652f1babc4c9

الفئة