![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F764901%2F0aea46971f757488640bab63f92648b9%2FPetals%20to%20the%20Metal.png?generation=1610139021106399&alt=media\")
Hello Kagglers π,
\nI hope you are doing great and are excited about the possibilities 2021 could bring us π
\nI wanted to share a live session hosted today January 8th by my community, Deep Learning Adventures.
I will be taking a deep dive into my approach for this challenge and show you how transfer learning, hyperparameter tuning and learning from each other helped me perform relatively well in this competition.
\nHere is my notebook I will be covering in detail: https://www.kaggle.com/georgezoto/computer-vision-petals-to-the-metal
\nI appreciate your comments and feedback here or on my notebook.
\nIf you would like to learn more as well as join a larger data science community π, feel free to join us at: https://www.meetup.com/Deep-Learning-Adventures/events/275438349/
\nAll our sessions are recorded (including this one) π and available on YouTube at: http://bit.ly/dla-kaggle-courses
\n
You can find our recording at the bottom of our playlist.
\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F764901%2F3cdc2140332b24f2499ea1a6911d155e%2FPetal.png?generation=1610203641021480&alt=media\")
Best,
\nGeorge
Hello Kagglers π,
\nI hope you are doing great and are excited about the possibilities 2021 could bring us π
\nI wanted to share a live session hosted today January 8th by my community, Deep Learning Adventures.
I will be taking a deep dive into my approach for this challenge and show you how transfer learning, hyperparameter tuning and learning from each other helped me perform relatively well in this competition.
\nHere is my notebook I will be covering in detail: https://www.kaggle.com/georgezoto/computer-vision-petals-to-the-metal
\nI appreciate your comments and feedback here or on my notebook.
\nIf you would like to learn more as well as join a larger data science community π, feel free to join us at: https://www.meetup.com/Deep-Learning-Adventures/events/275438349/
\nAll our sessions are recorded (including this one) π and available on YouTube at: http://bit.ly/dla-kaggle-courses
\n
You can find our recording at the bottom of our playlist.
\n
Best,
\nGeorge
PS: In my current experience and setup, I could get a best Public Score=0.69 using VGG16, no augmentation with pretrained_model.trainable = False
\nPlease take a look at:
\nhttps://www.kaggle.com/code/rbhambri/petals-classifier-prediction?scriptVersionId=116895380
Next thing i wish to attempt is set of other models, inspired by your logs. Thanks!
", "rawMarkdown": "PS: In my current experience and setup, I could get a best Public Score=0.69 using VGG16, no augmentation with pretrained_model.trainable = False\n\nPlease take a look at:\nhttps://www.kaggle.com/code/rbhambri/petals-classifier-prediction?scriptVersionId=116895380\n\nNext thing i wish to attempt is set of other models, inspired by your logs. Thanks!\n\n" }, { "id": 2203086, "postDate": "2023-03-30T13:57:02.330Z", "content": "This is great work, nice plot with different models. That was something I was really looking for.
", "rawMarkdown": "This is great work, nice plot with different models. That was something I was really looking for.\n\n\n" } ], "comments": [ { "id": 2203090, "author_name": "RahulBhambri", "author_url": "", "post_date": "2023-03-30T14:01:11.960000", "content": "PS: In my current experience and setup, I could get a best Public Score=0.69 using VGG16, no augmentation with pretrained_model.trainable = False
\nPlease take a look at:
\nhttps://www.kaggle.com/code/rbhambri/petals-classifier-prediction?scriptVersionId=116895380
Next thing i wish to attempt is set of other models, inspired by your logs. Thanks!
", "votes": 0, "replies": [] }, { "id": 2203086, "author_name": "RahulBhambri", "author_url": "", "post_date": "2023-03-30T13:57:02.330000", "content": "This is great work, nice plot with different models. That was something I was really looking for.
", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "1145050": "Hello Kagglers π,\n\nI hope you are doing great and are excited about the possibilities 2021 could bring us π\nI wanted to share a live session hosted today January 8th by my community, Deep Learning Adventures. \n\nI will be taking a deep dive into my approach for this challenge and show you how transfer learning, hyperparameter tuning and learning from each other helped me perform relatively well in this competition.\n\nHere is my notebook I will be covering in detail: https://www.kaggle.com/georgezoto/computer-vision-petals-to-the-metal\n\nI appreciate your comments and feedback here or on my notebook.\n\nIf you would like to learn more as well as join a larger data science community π, feel free to join us at: https://www.meetup.com/Deep-Learning-Adventures/events/275438349/\n\nAll our sessions are recorded (including this one) π and available on YouTube at: http://bit.ly/dla-kaggle-courses\n\n\n\nYou can find our recording at the bottom of our playlist.\n\n\nBest,\nGeorge", "2203090": "PS: In my current experience and setup, I could get a best Public Score=0.69 using VGG16, no augmentation with pretrained_model.trainable = False\n\nPlease take a look at:\nhttps://www.kaggle.com/code/rbhambri/petals-classifier-prediction?scriptVersionId=116895380\n\nNext thing i wish to attempt is set of other models, inspired by your logs. Thanks!\n\n", "2203086": "This is great work, nice plot with different models. That was something I was really looking for.\n\n\n" } }