BEGIN:VCALENDAR VERSION:2.0 PRODID:-//American Young Coder - ECPv6.10.1.1//NONSGML v1.0//EN CALSCALE:GREGORIAN METHOD:PUBLISH X-ORIGINAL-URL:https://www.ayclogic.com X-WR-CALDESC:Events for American Young Coder REFRESH-INTERVAL;VALUE=DURATION:PT1H X-Robots-Tag:noindex X-PUBLISHED-TTL:PT1H BEGIN:VTIMEZONE TZID:America/Los_Angeles BEGIN:DAYLIGHT TZOFFSETFROM:-0800 TZOFFSETTO:-0700 TZNAME:PDT DTSTART:20260308T100000 END:DAYLIGHT BEGIN:STANDARD TZOFFSETFROM:-0700 TZOFFSETTO:-0800 TZNAME:PST DTSTART:20261101T090000 END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTART;TZID=America/Los_Angeles:20260304T190000 DTEND;TZID=America/Los_Angeles:20260304T200000 DTSTAMP:20260420T153205 CREATED:20260305T062853Z LAST-MODIFIED:20260305T063006Z UID:32754-1772650800-1772654400@www.ayclogic.com SUMMARY:7 PM – AI/ML – Darin DESCRIPTION:Today’s Activities:\n\nIntroduced the topic of ML/AI at a high level.\nIntroduced Kaggle as a platform for running ML code.\n\nHomework:\nCopy the contents from this google colab file to kaggle\, and complete the hw\, then save as .ipynb: https://colab.research.google.com/drive/1qjmmP4LhSIgVD0qmExb_fmuB-Sm_511n \nNotes:\nYou can reach me at ddjapri@ayclogic.com. \nAll class notes can be found here. URL:https://www.ayclogic.com/event/7-pm-ai-ml-darin-10/ CATEGORIES:AI/ML,Python Class END:VEVENT BEGIN:VEVENT DTSTART;TZID=America/Los_Angeles:20260302T190000 DTEND;TZID=America/Los_Angeles:20260302T200000 DTSTAMP:20260420T153205 CREATED:20260303T041846Z LAST-MODIFIED:20260303T041916Z UID:32729-1772478000-1772481600@www.ayclogic.com SUMMARY:7 PM – AI/ML – Darin DESCRIPTION:Today’s Activities:\n\nContinued multi-class classification on the MNIST dataset.\nLearned about data augmentation through the batch_tfms parameter for the dataloader.\n\nHomework:\nFor Reine: \n\nCreate a google doc under your homework google drive \nWrite up your project proposal for training your own AI model detailing:\n– What you will be classifying (has to be multi-class classification!)\n– Where your data comes from (provide a link or methodology of getting the dataset). If you are using an existing dataset\, you have to add more images from online means (gather data from outside of kaggle\, whether it’s online images\, etc)\n– How you will train your AI model (what transformations on the data you will use\, and the flow of training)\nYou can view the list of transformations here: https://docs.fast.ai/vision.augment.html#aug_transforms\n\nEnsure the google doc is public access and that it is in your homework folder. \n  \nFor everyone: \n\nContinue working on your final project. You want to at least ensure you have the data loader setup with the proper transformations (also at this point it should be easy to run training once you have the dataloader). Upload your latest progress as an ipynb to the google drive.\nCreate a new kaggle notebook Mar9_PandasHW.ipynb and do the following:\nYoutube.csv  \n\nUse this csv file – https://drive.google.com/file/d/1kP6A9y0UBssOg3Exunv9Mnmilb0657Sh/view?usp=drive_link \nLoad the data\n\nShow only the Channel and Subscribers columns\n\nFind channels with more than 2000 subscribers\n\nAdd a column Subs_per_Video\n\nWhich channel is the most efficient?\n\n\n\nHW Note: you can display rows with conditions like this:\ndf_math_greater_than_80 = df[df["Math"] > 80]\nprint("\nStudents with Math Score Greater Than 80:")\nprint(df_math_greater_than_80)\nprint(df.head())\nNotes:\nYou can reach me at ddjapri@ayclogic.com. \nAll class notes can be found here. URL:https://www.ayclogic.com/event/7-pm-ai-ml-darin-9/ CATEGORIES:AI/ML,Python Class END:VEVENT BEGIN:VEVENT DTSTART;TZID=America/Los_Angeles:20260223T190000 DTEND;TZID=America/Los_Angeles:20260223T200000 DTSTAMP:20260420T153205 CREATED:20260224T043712Z LAST-MODIFIED:20260224T043712Z UID:32645-1771873200-1771876800@www.ayclogic.com SUMMARY:7 PM – AI/ML – Darin DESCRIPTION:Today’s Activities:\n\nNo class today\, there was a technical issue with zoom links\, we will meet next week in the updated zoom link. Make sure you join the right one!\n\nHomework:\n\nBegin working on your project! You will have exactly 3 weeks to complete it\, and by the end of it you want to deploy a simple streamlit application for your project.\n\nNotes:\nYou can reach me at ddjapri@ayclogic.com. \nAll class notes can be found here. URL:https://www.ayclogic.com/event/7-pm-ai-ml-darin-8/ CATEGORIES:AI/ML,Python Class END:VEVENT BEGIN:VEVENT DTSTART;TZID=America/Los_Angeles:20260216T190000 DTEND;TZID=America/Los_Angeles:20260216T200000 DTSTAMP:20260420T153205 CREATED:20260217T043338Z LAST-MODIFIED:20260217T043554Z UID:32559-1771268400-1771272000@www.ayclogic.com SUMMARY:7 PM – AI/ML – Darin DESCRIPTION:Today’s Activities:\n\nContinued multi-class classification on the MNIST dataset.\nLearned about data augmentation through the batch_tfms parameter for the dataloader.\n\nHomework:\n\nCreate a google doc under your homework google drive \nWrite up your project proposal for training your own AI model detailing:\n– What you will be classifying (has to be multi-class classification!)\n– Where your data comes from (provide a link or methodology of getting the dataset). If you are using an existing dataset\, you have to add more images from online means (gather data from outside of kaggle\, whether it’s online images\, etc)\n– How you will train your AI model (what transformations on the data you will use\, and the flow of training)\nYou can view the list of transformations here: https://docs.fast.ai/vision.augment.html#aug_transforms\n\nEnsure the google doc is public access and that it is in your homework folder. \nNotes:\nYou can reach me at ddjapri@ayclogic.com. \nAll class notes can be found here. URL:https://www.ayclogic.com/event/7-pm-ai-ml-darin-7/ CATEGORIES:AI/ML,Python Class END:VEVENT BEGIN:VEVENT DTSTART;TZID=America/Los_Angeles:20260209T190000 DTEND;TZID=America/Los_Angeles:20260209T200000 DTSTAMP:20260420T153205 CREATED:20260210T043506Z LAST-MODIFIED:20260217T043259Z UID:32426-1770663600-1770667200@www.ayclogic.com SUMMARY:7 PM – AI/ML – Darin DESCRIPTION:Today’s Activities:\n\nContinued multi-class classification on the MNIST dataset.\nLearned about data augmentation through the batch_tfms parameter for the dataloader.\n\nHomework:\n\nCreate a google doc under your homework google drive \nWrite up your project proposal for training your own AI model detailing:\n– What you will be classifying (has to be multi-class classification!)\n– Where your data comes from (provide a link or methodology of getting the dataset). If you are using an existing dataset\, you have to add more images from online means (gather data from outside of kaggle\, whether it’s online images\, etc)\n– How you will train your AI model (what transformations on the data you will use\, and the flow of training)\n\nEnsure the google doc is public access and that it is in your homework folder. \nNotes:\nYou can reach me at ddjapri@ayclogic.com. \nAll class notes can be found here. URL:https://www.ayclogic.com/event/7-pm-ai-ml-darin-5/ CATEGORIES:AI/ML,Python Class END:VEVENT BEGIN:VEVENT DTSTART;TZID=America/Los_Angeles:20260202T190000 DTEND;TZID=America/Los_Angeles:20260202T200000 DTSTAMP:20260420T153205 CREATED:20260203T041351Z LAST-MODIFIED:20260203T041351Z UID:32359-1770058800-1770062400@www.ayclogic.com SUMMARY:7 PM – AI/ML – Darin DESCRIPTION:Today’s Activities:\n\nRevisited homework on using different models for training and analyzed their performance.\nLearnt about virtual environments and dependencies (your “copy” of the computer)\nSetup github and demonstrated web app deployment (now you can ask your friends to check out your ai models haha)\n\nHomework:\n\nCopy the Jan26_CatVSDog.ipynb into Feb2_PetBreed.ipynb\nMODIFY the labeling function such that it labels by the dog BREED.\nYou do not return “dog” vs “cat” but rather you return “Siamese”\, “Abyssinian”\, etc \nHint: You need to only keep the string WITHOUT numbers.\nRun training again after modifying\nThis should result in MULTI-class classification\n\nUpload the file Jan26_CatVSDog_Evaluation.py to the google drive when you are done. \nNotes:\nYou can reach me at ddjapri@ayclogic.com. \nAll class notes can be found here. URL:https://www.ayclogic.com/event/7-pm-ai-ml-darin-4/ CATEGORIES:AI/ML,Python Class END:VEVENT BEGIN:VEVENT DTSTART;TZID=America/Los_Angeles:20260126T190000 DTEND;TZID=America/Los_Angeles:20260126T200000 DTSTAMP:20260420T153205 CREATED:20260127T051814Z LAST-MODIFIED:20260127T051814Z UID:32291-1769454000-1769457600@www.ayclogic.com SUMMARY:7 PM – AI/ML – Darin DESCRIPTION:Today’s Activities:\n\nLearnt a bit about tensors\, pre-trained models\, model sizes\nExported our trained AI model to a .pkl file for use outside of kaggle\nRan a quick streamlit app to visualize model working on the browser\n\nHomework:\n\nRun training again like we did in class\, but this time using 3 different AI backbones (replacing the ‘resnet34’ under the vision_learner function call) like last week.\nReferences for model types: \nhttps://docs.pytorch.org/vision/stable/models.html#initializing-pre-trained-models\nhttps://docs.fast.ai/vision.learner.html#vision_learner\nSAVE these models as .pkl files and put them in your pycharm project directory.\nFinally modify the pycharm file to be able to select between different models you have trained for evaluation.\nYou can add this section beneath the “created by” code to replace the load_learner for multi model selection functionality. \n\nMODELS_DIR = Path("models")\n\n# Find all .pkl files in models/\nmodel_paths = sorted(MODELS_DIR.glob("*.pkl"))\nmodel_names = [p.name for p in model_paths]\n\nif not model_paths:\n st.error(f"No .pkl models found in: {MODELS_DIR.resolve()}")\n st.stop()\n\n# UI: choose which model to use\nselected_name = st.selectbox("Select a model to use:"\, model_names)\n\n@st.cache_resource # cache the loaded learner per selected model\ndef get_model(model_path_str: str):\n return load_learner(model_path_str)\n\nmodel_path = str(MODELS_DIR / selected_name)\ncat_vs_dog_model = get_model(model_path)\n\nst.caption(f"Loaded model: {selected_name}")\n\n\nRUN your app with “streamlit run ” such as “streamlit run src/Jan26_CatVSDog_Evaluation.py”\n\nUpload the file Jan26_CatVSDog_Evaluation.py to the google drive when you are done. \nNotes:\nYou can reach me at ddjapri@ayclogic.com. \nAll class notes can be found here. URL:https://www.ayclogic.com/event/7-pm-ai-ml-darin-3/ CATEGORIES:AI/ML,Python Class END:VEVENT BEGIN:VEVENT DTSTART;TZID=America/Los_Angeles:20260119T190000 DTEND;TZID=America/Los_Angeles:20260119T200000 DTSTAMP:20260420T153205 CREATED:20260120T040741Z LAST-MODIFIED:20260120T040741Z UID:32213-1768849200-1768852800@www.ayclogic.com SUMMARY:7 PM – AI/ML – Darin DESCRIPTION:Today’s Activities:\n\nContinued on the first project: dog and cat classification\nLearnt about the labelling function\, and commenced training with a GPU for classification\n\nHomework:\nRun training again like we did in class\, but this time using 3 different AI backbones (replacing the ‘resnet34’ under the vision_learner function call). \nLook up the different AI backbones from these links (just try around like resnet50\, resnet18\, or even ulearner etc): \nhttps://docs.pytorch.org/vision/stable/models.html#initializing-pre-trained-models\nhttps://docs.fast.ai/vision.learner.html#vision_learner \nThen make your dataset have at least 10 images (I recommend looking for pictures of cats that look like dogs\, and dogs that look like cats)\, and test these images with your newly trained models. \nWrite in the same file Jan12_CatVSDog.ipynb as to which AI backbone you think is better in a text box. Upload this file to google drive when you are done. \nNotes:\nYou can reach me at ddjapri@ayclogic.com. \nAll class notes can be found here. URL:https://www.ayclogic.com/event/7-pm-ai-ml-darin-2/ CATEGORIES:AI/ML,Python Class END:VEVENT BEGIN:VEVENT DTSTART;TZID=America/Los_Angeles:20260112T190000 DTEND;TZID=America/Los_Angeles:20260112T200000 DTSTAMP:20260420T153205 CREATED:20260113T051833Z LAST-MODIFIED:20260113T051833Z UID:32137-1768244400-1768248000@www.ayclogic.com SUMMARY:7 PM – AI/ML – Darin DESCRIPTION:Today’s Activities:\n\nWorked on the first project: dog and cat classification\nGot introduced to various libraries for the training stack\nLearnt about the technicalities of images\n\nHomework:\nSubmit to your respective google drive homework folders when you are finished! \n\nFigure out which two popular dog breeds from list below that DO NOT exists in the Oxford IIIT Pet dataset. YOU HAVE TO WRITE CODE TO FIND THIS. YOU CANNOT JUST GUESS. Hint look at the existing codes where we print the “Abyssinian” cat. You need to modify the code to figure out which of dog breed below that does not exists\n\namerican Pit Bull\nchihuahua\nhusky\nshiba Inu\nsamoyed\ndachshund\npomeranian\nbeagle\nboxer\npug\n\n\nWrite your answers in the latest Jan12_CatVSDog.ipynb file we worked on in a text module and explain how you found the solution.\nDownload the .ipynb file after finishing your code. Note that you can save a copy of the file so your work is persistent.\n\nNotes:\nYou can reach me at ddjapri@ayclogic.com. \nAll class notes can be found here. URL:https://www.ayclogic.com/event/7-pm-ai-ml-darin/ CATEGORIES:AI/ML,Python Class END:VEVENT BEGIN:VEVENT DTSTART;TZID=America/Los_Angeles:20260105T190000 DTEND;TZID=America/Los_Angeles:20260105T200000 DTSTAMP:20260420T153205 CREATED:20260105T233127Z LAST-MODIFIED:20260112T224822Z UID:32059-1767639600-1767643200@www.ayclogic.com SUMMARY:7 PM – AI/ML – Darin DESCRIPTION:Today’s Activities:\n\nIntroduced the motivation of learning ML/AI\nIntroduced various divisions of ML/AI\nIntroduced Kaggle\, an interface for running ML code\n\nHomework:\nSubmit to your respective google drive homework folders when you are finished! \n\nDo this simple assignment: https://colab.research.google.com/drive/1qjmmP4LhSIgVD0qmExb_fmuB-Sm_511n\nAlso upload 2 images and load them with matplotlib at the end of the notebook.\nDownload the .ipynb file after finishing your code. Note that you can save a copy of the file so your work is persistent.\n\nNotes:\nYou can reach me at ddjapri@ayclogic.com. \nAll class notes can be found here. URL:https://www.ayclogic.com/event/5-pm-ai-ml-darin/ CATEGORIES:AI/ML,Python Class END:VEVENT END:VCALENDAR