Feng Zhu blog

Jul 24, 2022 Vision Language Pre-training Model
Multimodal model pre-training introduction
Sep 12, 2021 pre-training model in NLP
from word embeddings to pre-training model based on large-scale data set to better capture and represent the contextual information including Word2Vec, ELMo, Transformer, Transformer-XL, GPT, BERT, XLNet and ERNIE
Feb 2, 2021 InceptionNet
The Explanation of GoogLeNet, Inception-v2, Inception-v3, Inception-v4 and Inception-ResNet
Feb 1, 2021 evaluation metrics in language model
common evaluation metrics in language model: BLEU, ROUGE, Perplexity
Jan 25, 2021 object detection and segmentation
A brief introduction about R-CNN, Fast R-CNN, Faster R-CNN, SSD, YOLO and Mask R-CNN
Jan 22, 2021 MobileNets, EfficientNet and EfficientDet
suitable network architecture design to achieve the trade-off between model efficiency (computing and storage resources, model size and latency) and model accuracy
Jan 8, 2021 ranking algorithm
Ranking methods including PageRank, BM25 and Learning to Rank (L2R)
Jan 7, 2021 named entity recognition methods
Short summary of NER: HMM, CRF, BI-LSTM, BI-LSTM-CRF, BERT-BI-LSTM-CRF
Jan 5, 2021 Hyperparameter optimization
the common hyperparameter tuning methods are as following: Grid search, Random search and Bayesian optimization
Jan 4, 2021 git command to recover the modification
git command without using git add, with using git add and git commit
Jan 4, 2021 optimization algorithm for neural network
A brief introduction of gradient descent optimization algorithms
May 31, 2016 Deep Reinforcement Learning: Pong from Pixels
fork from Andrej Karpathy blog