{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/0c909764d6794fadb759b8a58c715323\" frameborder=\"0\" width=\"1280\" height=\"960\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":960,"width":1280,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":960,"thumbnail_width":1280,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/0c909764d6794fadb759b8a58c715323-00001.gif","duration":211,"title":"Roboflow Train - Train from Scratch vs. Train from Checkpoint","description":"How to employ \"Train from Scratch\" vs. \"Train from Checkpoint\" to create a more robust model with Roboflow's one-click training solution, Roboflow Train. Training from a checkpoint leverages \"transfer learning.\" At its core, transfer learning is using what a given model has learned about one domain and applying those learnings to attempt to learn a related problem. For more on transfer learning, read the blog post from Roboflow CEO, Joseph Nelson, \"A Primer on Transfer Learning\": https://blog.roboflow.com/a-primer-on-transfer-learning/."}