SVM Classifier - Githubissues

ijyliu / computer-vision-project

Using classical and neural image embeddings and finetuned end-to-end networks to achieve top-tier performance on a vehicle type classification task. Containerized and deployed model as a web app

https://cv-web-app-3m4f2rmfzq-uc.a.run.app/

0 stars 0 forks source link

SVM Classifier #39

Closed ijyliu closed 5 months ago

ijyliu commented 6 months ago

Input: all features train and test files. to construct X in the training run you should drop all string variables, and have 'Class', as the y variable. to construct X in the prediction run you should drop all string variables and there should not be a y

Output:

[x] joblib file of model.
[x] Excel file containing time to fit/train overall, train and validation accuracies.
[x] Pickle file of dictionaries of all hyperparameters searched and optimal hyperparameters
[x] Excel file of predictions - this should be all string columns in all features test + one column called classifier_name + '_Prediction'. Excel file of inference time.

CalBlitz commented 6 months ago

WIP. Will be done by March 30.

ijyliu commented 6 months ago

Definitely suggest using gridsearchCV in skleanr since they require grid search

Also I have things set up so we can save predictions from the classifier and then have an evaluation file that merges all the predictions

ijyliu commented 5 months ago

still running for All_Features version

might not be feasible for it to finish in time

CalBlitz commented 5 months ago

No worries! I'll come up with a reason for it's substandard performance.

ijyliu commented 5 months ago

Too many features pretty much

On Sun, Apr 14, 2024, 3:04 AM Gaurav Sharan Srivastava < @.***> wrote:

Closed #39 https://github.com/ijyliu/computer-vision-project/issues/39 as completed.

— Reply to this email directly, view it on GitHub https://github.com/ijyliu/computer-vision-project/issues/39#event-12460041227, or unsubscribe https://github.com/notifications/unsubscribe-auth/AQCGE4JZUNA5PVLUUJXGOH3Y5JICVAVCNFSM6AAAAABFE3OG3SVHI2DSMVQWIX3LMV45UABCJFZXG5LFIV3GK3TUJZXXI2LGNFRWC5DJN5XDWMJSGQ3DAMBUGEZDENY . You are receiving this because you authored the thread.Message ID: @.*** com>