Start studying: Saturday, December 7th
Teaching materials, etc .: ・ Miyuki Oshige "Details! Python3 Introductory Note ”(Sotec, 2017): Completed on Thursday, December 19th ・ Progate Python course (5 courses in total): Ends on Saturday, December 21st ・ Andreas C. Müller, Sarah Guido "(Japanese title) Machine learning starting with Python" (O'Reilly Japan, 2017): Completed on Saturday, December 23
Kaggle Participation competition: Real or Not? NLP with Disaster Tweets 12/24 (Tue) ~
・ It seems that I had misunderstood various things, and the problem I had until yesterday (such as not being able to take two values in the argument) was solved.
-However, even if learning with SVC or xgboost, there is a problem that the score is around 0.5. Since the assignments are classified as 0 or 1, it seems that they are not learning correctly after all.
-Since vectorization and division of training data and test data have been completed, we are reviewing whether the model is correct again (currently ongoing).
・ The flow of machine learning that I understand Import various modules → Read data → Pre-processing (data processing, division / x_train, y_train, x_test, y_test, etc.) → Model creation → Learning by model (fit x_train and y_train) → Create predictive model (use fit) X_test is predict) → Evaluation (combine predict and y_test corresponding to the answer part) → Review various processes and tune parameters according to the result. If stacking, implement various learning models required.
Recommended Posts