Introduction What is a whisper model? Whisper is an open source speech recognition model trained by OpenAI. It enables transcription in multiple languages, as well as translation from those languages into English. Because Whisper was trained on a large and diverse dataset and was not fine-tuned to any specific one, it does not beat models that... Read more 13 Apr 2024 - 2 minute read
Introduction: tsai is an open source deep learning package built on top of Pytorch and fastai focused on tasks like classification, regression, forecasting, imputation and many more. The aim of the blog post is to provide solution in performing inference of tsai trained classification model. The training method is simple, but I was facing pr... Read more 03 Apr 2024 - 2 minute read
Hi everyone, I have written this blog post to share steps and my knowledge on how to fine tune a Large language model at ease. Introduction I am using the huggingface library to fine tune the “bigscience/bloom-7b1” model using LoRA (Low rank adaptation of large language models). bigscience/bloom-7b1 is a BigScience Large Open science op... Read more 07 Oct 2023 - 7 minute read
Paper: Xtreme Margin: A Tunable Loss Function for Binary Classification Problems This is a paper summary generated from summarizepaper.com. I edited for better understanding. Introduction Loss functions are crucial in optimizing machine learning algorithms. The choice of loss function impacts the training process and model learning. Bi... Read more 16 Jul 2023 - 3 minute read
This is a code snippet to generate image augmentation on your training data. from imgaug import augmenters as iaa from skimage.io import imread_collection import cv2 import numpy as np # list .PNG files from the folder IMAGE_FOLDER_PATH = "/content/*.PNG" seq = iaa.Sequential([ iaa.Crop(px=(0, 16)), # crop images from each side by 0 to 1... Read more 04 Jul 2023 - less than 1 minute read
In this blog post, I explain the steps that I followed to install google pegasus summatization model from the officical github page. I failed to complete the installation, since the pip package “tensorflow_text” failed to import, I will mention the error it threw. I cloned the github repository. I downloaded the pretrained model weights. ... Read more 25 Jun 2023 - 10 minute read
I was working on a project, where I wanted to analyze the decisions from the decision tree. I thought, converting a decision tree to decision table helps in the interpretation and analysis. There was just one stack-overflow solution for converting a decision tree to decision table, but it failed for my use case. Hence I wrote a parsing code to... Read more 16 Jan 2023 - 2 minute read
In this blog post I will provide the procedures that I followed to inference the DiffusionDet Model for object detection. This is based on the Github repo Paper I created a conda environment conda create --name diffusionDetection python=3.7 Detectron repo needs python >= 3.7 Activate the conda environment conda activate... Read more 25 Nov 2022 - 12 minute read
PyTorch LSTM model architecture LSTM1( (lstm1): LSTM(2800, 5, batch_first=True) (linear): Linear(in_features=5, out_features=5, bias=True) (softmax): Softmax(dim=None) ) class LSTM1(nn.Module): def __init__(self, input_size, hidden_size, output_size): super().__init__() self.lstm1 = torch.nn.LSTM(input_size, hi... Read more 27 Sep 2022 - 1 minute read