Curiosity Trace
About
Curiosity Trace
Categories
All
(8)
computer-science
(1)
cuda
(1)
datasets
(1)
fastai
(2)
huggingface
(2)
litellm
(1)
llm
(1)
math
(2)
neural-network
(1)
openai
(1)
python
(1)
Benchmark CUDA kernel with Python
cuda
python
In this tutorial, we’ll see the impact of block size on the performance of a CUDA kernel. We’ll use
subprocess
module in Python standard library to compile and execute a…
Mar 16, 2024
Serving open source LLM as OpenAI API
llm
huggingface
openai
litellm
In this tutorial, we’ll see how to serve an open source language model with OpenAI compatible API using HuggingFace Text Generation Inference and LiteLLM OpenAI proxy…
Jan 8, 2024
How to use HuggingFace datasets with fastai?
fastai
huggingface
datasets
In this tutorial, we will show you how to use a HuggingFace dataset with fastai to train a model for image classification. We will use the Beans dataset, which consists of…
Apr 30, 2023
How to use a custom model with fastai cnn_learner?
fastai
fastai library offers many pre-trained models for vision tasks. However, we sometimes need to use a custom model available in another library or created from scratch. In…
Feb 11, 2022
How many coin flips does high confidence need?
math
You walked into a casino, there is a game of tossing coins. You don’t know whether the coin is biased or fair. But, you can observe the game and keep the stats. How many…
Jul 25, 2020
Why is initialization important in neural networks?
neural-network
Today, when we build a deep neural network with standard layers and activation functions, we don’t worry much about the initialization of network’s parameters. But, this…
Jun 25, 2020
Fourier Transform
math
A minimal introduction
Jun 23, 2020
Halting problem
computer-science
Halting problem was proved to be unsolvable by Alan Turing in his seminal paper, in 1936. It states that there cannot be a general algorithm to decide whether an arbitrary…
Apr 7, 2020
No matching items