This is a joint post with Inbar Naor. Originally published at engineering.taboola.com Understanding what a model doesn’t know is important both from the practitioner’s perspective
Read more
Aggregated news around AI and co.
This is a joint post with Inbar Naor. Originally published at engineering.taboola.com Understanding what a model doesn’t know is important both from the practitioner’s perspective
Read moreHow Raven Protocol is closing the last gap At the beginning of my professional career, I’ve been working as a Data Scientist, and one of my
Read moreUnderstanding what a model doesn’t know is important both from the practitioner’s perspective and for the end users of many different machine learning applications. In
Read moreQue es autokeras y por que es un paso en la dirección correcta Si uno ve la descripción de la librería Autokeras puede pensar que
Read moreConversational interfaces and natural language processing(NLP) are, arguably, the most widely adopted segment of modern artificial intelligence(AI). Despite the continuous progress in NLP research, most
Read moreThe automated curation of music playlists has become a significant problem in the last decade with the rise of colossal streaming platforms. Current state-of-the-art recommender
Read moreI started exploring the different ways to visualize the training process while working on the Dog breed identification dataset from Kaggle. I created a basic
Read moreSummit’s 12th flagship, LA18 will gather leaders across all disciplines. For three days, the festival will host a wide array of talks, performances, wellness classes,
Read moreWe bring out the ‘genius’ in you. There is a mandatory circular relationship between the thought and the thinker. It looks like this: Thought. Thinker.
Read more要能訓練出全面的模型, 模式1 : 精確的開在車道線上(precise model) 這個期間,不太在乎車速,只要確定車子沿著線條中間,大約開了兩圈車道,占了資料量10%左右。 模式2 : 在車道小幅震盪(small-oscillation model) 故意不要用一致的車速,而且也不一定開在中央車道線上,這樣的訓練資料,可以讓你的神經網路(Neural Network)學習車道上,不同角度的視角,以及如何操作回中央車道。 在這個模型,不需要過度蒐集資料,大約2至3圈車道的數據就足夠。 模式3 : 較多震盪(large-oscillation model) 除了正常駕駛之外,每次過度震盪的駕駛時間,都能夠拉長,並且加上一些車速的變換,例如慢速震盪、快速震盪,讓每次的震盪都能觸及邊界,又再度修正回軌道。並且有時不要過度修正震盪。 有些Donkey Car的車友,可能試圖讓車子開出界線,但是我覺得不大需要,依照上述的訓練模式,就能訓練出經得起考驗的模型了。 影片出處 Source: Deep Learning on Medium
Read moreInstead of obsessing over materials, the new technique takes a hard look at the picture itself – specifically, the thousands of tiny individual strokes that
Read moreI’ve read almost all blog posts explaining Transfer Learning and its immense practicality. However, I feel that Transfer Learning isn’t just ‘dump-your-data-in-a-pre-trained-network-and-train-for-some-epochs’; rather, it’s a
Read more