Amaresh PatnaikinDataDrivenInvestorRAG using GGUFA generative AI model generates text on the basis of the data it has been trained on. Over and above that, we can provide the model with…4 min read·Mar 4, 2024----
Amaresh PatnaikinDataDrivenInvestorCode Llama 7B on Free ColabThis article is a sequel to Mistral 7B on Free Colab and Zephyr 7B on Free Colab, though this article can also be read independently. In…5 min read·Dec 13, 2023--1--1
Amaresh PatnaikinDataDrivenInvestorZephyr 7B on Free ColabThis article is a sequel to Mistral 7B on Free Colab. In the previous article, I had explained that Large Language Models (LLMs) are…12 min read·Dec 3, 2023--1--1
Amaresh PatnaikinDataDrivenInvestorMistral 7B on Free ColabLarge Language Models (LLMs) can perform many natural language processing (NLP) tasks with a surprising degree of near-human performance…7 min read·Nov 7, 2023----
Amaresh PatnaikEncoder Decoder Sequences: How Long is Too Long? | Data Driven InvestorIn machine learning many times we deal with the input being a sequence and the output also a sequence. We call such a problem a…11 min read·Mar 25, 2020----
Amaresh PatnaikThe Math in ML: Cosine Similarity | Data Driven InvestorIn Machine Learning, we frequently express the similarity between two vectors as cosine similarity. What exactly does that mean? What has…7 min read·Nov 20, 2019----
Amaresh PatnaikNeural Network: A Counter ExampleNeural networks find pattern in your data and help you with unseen examples using the pattern they have discovered. What if there is no…3 min read·Jul 30, 2019----
Amaresh PatnaikinDataDrivenInvestorNeural Network: Check Before You RunYour neural network is not giving you the accuracy and performance you expected? Since a lot of the neural network is a black box hidden…7 min read·Jul 8, 2019----
Amaresh PatnaikinDataDrivenInvestorGradients in TensorFlowA neural network starts with an initial set of weights and calculates its first output as per the architecture of the network. It then…3 min read·Jul 3, 2019----
Amaresh PatnaikinDataDrivenInvestorMatch TensorFlow and Keras: SigmoidFor the same neural network, you should get the same results whether you use TensorFlow or you use Keras to implement your network. We saw…4 min read·Jun 27, 2019----