Más contenido relacionado

Similar a Deep Learning.pptx(20)


Deep Learning.pptx

  2. Introduction Motivation History of Deep Learning Data Framework History of Data Frame Works Era of Deep Learning Applications Tools HTTPS://WWW.GOOGLE.COM/URL?SA=I&URL=HTTPS%3A%2F%2FSTOCK.ADOBE.COM%2FPL%2FSEARCH%3FK%3DAGENDA&PSIG=AOVVAW0RULYIUOO82OCEBP7R9XXP&UST=1680122141330000&SO URCE=IMAGES&CD=VFE&VED=0CBAQJRXQFWOTCNIZUU68__0CFQAAAAADAAAAABAD 2
  3. Introduction  Deep learning is an artificial intelligence (AI) technology that trains computers to analyze data in a manner similar to that of the human brain.  Deep learning models are capable of recognizing complex patterns in images, text, sounds, and other data in order to generate accurate insights and predictions. HTTPS://WWW.GOOGLE.COM/URL?SA=I&URL=HTTPS%3A%2F%2FWWW.FORBES.COM%2FSITES%2FBERNARDMARR%2F2018%2F10%2F01%2FWHAT-IS-DEEP-LEARNING-AI-A-SIMPLE-GUIDE-WITH-8- PRACTICAL-EXAMPLES%2F&PSIG=AOVVAW3QTSODC3BYEAFZACRUD2KI&UST=1680122173650000&SOURCE=IMAGES&CD=VFE&VED=0CBAQJRXQFWOTCNJ52IO9__0CFQAAAAADAAAAABAH 3
  5. Diagnosis Deep learning algorithms are particularly well-suited for medical diagnosis jobs because they can learn to extract significant features from enormous datasets without explicit programming. Deep learning systems, for example, can learn to identify abnormalities in X-rays or MRIs by evaluating hundreds of images and identifying patterns that suggest sickness. HTTPS://WWW.GOOGLE.COM/URL?SA=I&URL=HTTPS%3A%2F%2FHEALTH.SUNNYBROOK.CA%2FMEMORY-DOCTOR%2FBRAIN-SCAN-ALZHEIMERS- DISEASE%2F&PSIG=AOVVAW2MKJW3XI5F0VUA-8O3IL0X&UST=1680122971247000&SOURCE=IMAGES&CD=VFE&VED=0CBAQJRXQFWOTCNI65_Y___0CFQAAAAADAAAAABAI 5
  6. Segmentation § Deep learning-based segmentation approaches have gained traction in recent years, particularly in medical imaging applications. Typically, these methods require training a neural network to recognize and segment specific structures or features in an image. § Segmentation is the process of identifying and isolating various items or structures within an image or video. HTTPS://DEEPLEARNINGOFPYTHON.BLOGSPOT.COM/2023/03/HISTORY-INTRODUCTION-DEEP%20LEARNING.HTML 6
  7. Exploring • Deep learning exploration is the process of evaluating and comprehending the behavior and performance of deep learning models. This process entails assessing the model's strengths, shortcomings, and limits, as well as identifying areas for development. • Visualization techniques, performance indicators, and explainability tools are among the tools and strategies available for analyzing deep learning models. HTTPS://DEEPLEARNINGOFPYTHON.BLOGSPOT.COM/2023/03/HISTORY-INTRODUCTION-DEEP%20LEARNING.HTML 7
  8. History of Deep Learning HTTP://BEAMLAB.ORG/DEEPLEARNING/2017/02/23/DEEP_LEARNING_101_PART1.HTML 8
  9. Deep learning data frames are a sort of data structure that is extensively used in machine learning and data analysis. Adata frame is a two-dimensional table-like structure with rows and columns, each of which represents an observation and each column a variable. Data frames are commonly used in deep learning to hold training data, which comprises of input characteristics and output labels. The data required to train the model is represented by the input features, while the output labels reflect the desired output for each input. Data frames can be built from a variety of data sources, including CSV files, databases, and web APIs. Data frames can be preprocessed and changed after they are created, employing techniques such as normalization, feature scaling, and one-hot encoding, among others. Data Frame HTTPS://DEEPLEARNINGOFPYTHON.BLOGSPOT.COM/2023/03/HISTORY-INTRODUCTION-DEEP%20LEARNING.HTML 9
  10. History of Data Frameworks Deep learning frameworks like Kera's, MX Net, and CNTK have proliferated in recent years. Each framework has advantages and disadvantages, and the framework chosen typically relies on the project's specific requirements. Caffe, the first deep learning framework, was created in 2014 by the Berkeley Vision and Learning Center. Google launched TensorFlow as an open-source deep learning framework in 2015. Facebook published PyTorch as an open-source deep- learning platform in 2016. The Montreal Institute for Learning Algorithms also released Theano in 2015. (MILA). HTTPS://DEEPLEARNINGOFPYTHON.BLOGSPOT.COM/2023/03/HISTORY-INTRODUCTION-DEEP%20LEARNING.HTML 10
  11. Why this is the era of Deep Learning? Deep learning has gained popularity in recent years for numerous reasons: Big Data Better Hardware Improved Algorithms HTTPS://DEEPLEARNINGOFPYTHON.BLOGSPOT.COM/2023/03/HISTORY-INTRODUCTION-DEEP%20LEARNING.HTML 11
  12. Applications Natural Language Processing Computer Vision Health Care Finance Gaming Robotics Marketing Agriculture Energy Social Media HTTPS://DEEPLEARNINGOFPYTHON.BLOGSPOT.COM/2023/03/HISTORY-INTRODUCTION-DEEP%20LEARNING.HTML 12
  14. Popular Tool TensorFlow is the most widely used deep learning technology. TensorFlow is a Google open-source software library for constructing and training neural networks. It supports a wide range of platforms, including CPUs, GPUs, and TPUs, and provides a high-level interface for constructing and deploying deep learning models (Tensor Processing Units). TensorFlow is widely used in academia and industry, and it has a large user and developer community. It includes data preparation, data augmentation, visualization, and model tweaking tools and features for developing and training deep learning models. HTTPS://DEEPLEARNINGOFPYTHON.BLOGSPOT.COM/2023/03/HISTORY-INTRODUCTION-DEEP%20LEARNING.HTML 14
  15. 15