Taeyoon.Kim.DS

랜덤 포레스트는 여러가지 Decision Tree의 합이다. 여러 가지 Decision Tree를 만든 후에 voting을 통해서 결과물을 어는다. Bagging은 샘플에서 샘플1에서 나온 모델1과 그 다음 샘플 2에서 나온 모델 2 그리고 모델 3처럼 parallel하게 만들어지고 더 로버스트한 모델을 만든다. Boosting은 Training이 sequential하게 각 분류기에 해당된다는 것이다. 샘플에서 모델1 -> 모델2 -> 모델3으로 가면서 성능이 향상된다. XGBoost와 Logistics Regression의 성능을 비교해볼 수 있다. Bagging은 Parallisation하는 것이다.

AWS Certified Machine Learning Specialty 2023. 10. 30. 01:39

https://wikidocs.net/24603 05-01 코사인 유사도(Cosine Similarity) BoW에 기반한 단어 표현 방법인 DTM, TF-IDF, 또는 뒤에서 배우게 될 Word2Vec 등과 같이 단어를 수치화할 수 있는 방법을 이해했다면 이러한 표현 방법에 대해서 … wikidocs.net Focusing on the vector's directions, thus cosine similarity can be used for two different length documentation and sentences. 분모는 Vector간의 크기의 곱 분자는 Vector간의 내적 (Inner dot product) What is dot product? https://wikidocs.net/2..

데이터 과학 2023. 10. 18. 00:11

The discussion on Amazon S3 Security covers several key aspects: User-Based Security: IAM policies are used to authorize specific API calls for individual IAM users. This means you can control what actions a user is allowed to perform within Amazon S3. Resource-Based Security: A newer approach involves using S3 Bucket policies. These policies are applied at the bucket level and can be configured..

AWS Certified Machine Learning Specialty 2023. 10. 9. 19:08

https://www.coursera.org/learn/generative-ai-with-llms/lecture/WVsMi/optional-video-aws-sagemaker-jumpstart Optional video: AWS Sagemaker JumpStart - Week 3 | Coursera Video created by deeplearning.ai, Amazon Web Services for the course "Generative AI with Large Language Models". Reinforcement learning and LLM-powered applications www.coursera.org Building an LLM application needs multiple compo..

Generative AI with Large Language Models 2023. 9. 20. 18:20

https://www.coursera.org/learn/generative-ai-with-llms/lecture/qojKp/model-optimizations-for-deployment Model optimizations for deployment - Week 3 | Coursera Video created by deeplearning.ai, Amazon Web Services for the course "Generative AI with Large Language Models". Reinforcement learning and LLM-powered applications www.coursera.org Before deploying a large language model (LLM) for applica..

Generative AI with Large Language Models 2023. 9. 19. 21:18

https://www.youtube.com/watch?v=Dh9CL8fyG80&list=PLo2EIpI_JMQvWfQndUesu0nPBAtZ9gP1o&index=30 from datasets import load_dataset load_dataset from transformers import AutoTokenizer, DataCollatorWithPadding raw_datasets = load_dataset("glue", "mrpc") # 데이터셋 checkpoint = "bert-base-cased" # 모델명 - 대소문자를 구분하는 버트 모델. tokenizer = AutoTokenizer.from_pretrained(checkpoint) # 토크나이저 - 사전학습된 모델을 이용한 토크나이저. 그..

Hugging Face Course 2023. 9. 19. 20:49

https://www.youtube.com/watch?v=nx10eh4CoOs&list=PLo2EIpI_JMQvWfQndUesu0nPBAtZ9gP1o&index=29 # ['logits']라면 우선 토크나이즈 된 데이터셋의 validation 데이터셋을 숫자로 만든 후, preds라는 변수에 저장. preds = model.predict(tokenized_datasets['validation'])['logits'] # 결과 확률값 = tensorflow의 nn 그리고 softmax값에 preds를 넣어서, 각 확률값이 도출되게 만듬. probabilities = tf.nn.softmax(preds) # class_preds는, probabilities에 도출된 확률값들에 대해서 argmax를 제공. ar..

Hugging Face Course 2023. 9. 19. 18:51

Transitioning Objects Between Storage Classes: Objects in Amazon S3 can be moved between storage classes, such as from Standard to Standard IA, to Intelligent Tiering, to One-Zone IA, and from there to Flexible Retrieval or Deep Archive. Various permutations are possible based on your data access and archiving needs. Automation with Lifecycle Rules: Transitioning objects can be done manually, bu..

AWS Certified Machine Learning Specialty 2023. 9. 18. 22:59