Wiki BackProp - Contributions [fr]

AI Agent

2023-09-19T09:33:16Z

Jboscher : Page créée avec « AI agents are artificial entities that sense their environment, make decisions, and take actions. The Rise and Potential of Large Language Model Based Agents: A Survey : https://arxiv.org/pdf/2309.07864.pdf »

AI agents are artificial entities that sense their environment, make decisions, and take actions.

The Rise and Potential of Large Language Model Based Agents: A Survey : https://arxiv.org/pdf/2309.07864.pdf

SFT

2023-08-01T12:10:19Z

Jboscher :

Supervised Fine-Tuning (SFT): Models are trained on a dataset of instructions and responses. It adjusts the weights in the LLM to minimize the difference between the generated answers and ground-truth responses, acting as labels.[1]

However, in some cases, updating the knowledge of the model is not enough and you want to modify the behavior of the LLM. In these situations, you will need a supervised fine-tuning (SFT) dataset, which is a collection of prompts and their corresponding responses. SFT datasets can be manually curated by users or generated by other LLMs. Supervised fine-tuning is especially important for LLMs such as ChatGPT, which have been designed to follow user instructions and stay on a specific task across long stretches of text. This specific type of fine-tuning is also referred to as instruction fine-tuning [2]

== Références ==

* [https://towardsdatascience.com/fine-tune-your-own-llama-2-model-in-a-colab-notebook-df9823a04a32] Fine-Tune Your Own Llama 2 Model in a Colab Notebook
* [https://bdtechtalks.com/2023/07/10/llm-fine-tuning/] The complete guide to LLM fine-tuning

SFT

2023-08-01T12:00:50Z

Jboscher : Page créée avec « Supervised Fine-Tuning (SFT): Models are trained on a dataset of instructions and responses. It adjusts the weights in the LLM to minimize the difference between the generated answers and ground-truth responses, acting as labels. == Références == * [https://towardsdatascience.com/fine-tune-your-own-llama-2-model-in-a-colab-notebook-df9823a04a32] Fine-Tune Your Own Llama 2 Model in a Colab Notebook »

Supervised Fine-Tuning (SFT): Models are trained on a dataset of instructions and responses. It adjusts the weights in the LLM to minimize the difference between the generated answers and ground-truth responses, acting as labels.

== Références ==

* [https://towardsdatascience.com/fine-tune-your-own-llama-2-model-in-a-colab-notebook-df9823a04a32] Fine-Tune Your Own Llama 2 Model in a Colab Notebook

PEFT

2023-07-27T09:34:26Z

Jboscher : Page créée avec « Thus, we, as a com- munity of researchers and engineers, need efficient ways to train on downstream task data. Parameter-efficient fine-tuning, which we denote as PEFT, aims to resolve this problem by only training a small set of parameters which might be a subset of the existing model parameters or a set of newly added parameters. https://arxiv.org/pdf/2303.15647.pdf »

Thus, we, as a com- munity of researchers and engineers, need efficient ways to train on downstream task data.

Parameter-efficient fine-tuning, which we denote as PEFT, aims to resolve this problem by only training a small set of parameters which might be a subset of the existing model parameters or a set of newly added parameters.

https://arxiv.org/pdf/2303.15647.pdf

Agent

2023-07-05T21:19:10Z

Jboscher :

In LangChain, agents are high-level components that use language models (LLMs) to determine which actions to take and in what order. An action can either be using a tool and observing its output or returning it to the user. Tools are functions that perform specific duties, such as Google Search, database lookups, or Python REPL.
Agents involve an LLM making decisions about which Actions to take, taking that Action, seeing an Observation, and repeating that until done.

Several types of agents are available in LangChain:
* The zero-shot-react-description agent uses the ReAct framework to decide which tool to employ based purely on the tool's description. It necessitates a description of each tool.
* The react-docstore agent engages with a docstore through the ReAct framework. It needs two tools: a Search tool and a Lookup tool. The Search tool finds a document, and the Lookup tool searches for a term in the most recently discovered document.
* The self-ask-with-search agent employs a single tool named Intermediate Answer, which is capable of looking up factual responses to queries. It is identical to the original self-ask with the search paper, where a Google search API was provided as the tool.
* The conversational-react-description agent is designed for conversational situations. It uses the ReAct framework to select a tool and uses memory to remember past conversation interactions.

Agent

2023-07-05T21:18:48Z

Jboscher : Page créée avec « In LangChain, agents are high-level components that use language models (LLMs) to determine which actions to take and in what order. An action can either be using a tool and observing its output or returning it to the user. Tools are functions that perform specific duties, such as Google Search, database lookups, or Python REPL. Agents involve an LLM making decisions about which Actions to take, taking that Action, seeing an Observation, and repeating that until... »

In LangChain, agents are high-level components that use language models (LLMs) to determine which actions to take and in what order. An action can either be using a tool and observing its output or returning it to the user. Tools are functions that perform specific duties, such as Google Search, database lookups, or Python REPL.
Agents involve an LLM making decisions about which Actions to take, taking that Action, seeing an Observation, and repeating that until done.

Several types of agents are available in LangChain:
- The zero-shot-react-description agent uses the ReAct framework to decide which tool to employ based purely on the tool's description. It necessitates a description of each tool.
- The react-docstore agent engages with a docstore through the ReAct framework. It needs two tools: a Search tool and a Lookup tool. The Search tool finds a document, and the Lookup tool searches for a term in the most recently discovered document.
- The self-ask-with-search agent employs a single tool named Intermediate Answer, which is capable of looking up factual responses to queries. It is identical to the original self-ask with the search paper, where a Google search API was provided as the tool.
- The conversational-react-description agent is designed for conversational situations. It uses the ReAct framework to select a tool and uses memory to remember past conversation interactions.

Temperature

2023-07-04T10:15:11Z

Jboscher :

Prompt design

2023-07-04T10:14:14Z

Jboscher :

The process of finding and designing the best input text to get the desired response back from the model is called prompt design, which often involves a lot of experimentation.

Please note a few best practices around prompt design. Be concise Be specific and well-defined Ask one task at a time Turn generative tasks into classification tasks. For example, instead of asking what programming language to learn, ask if Python, Java, or C is a better fit for a beginner in programming. and Improve response quality by including examples. Adding instructions and a few examples tends to yield good results however there’s currently no one best way to write a prompt. You may need to experiment with different structures, formats, and examples to see what works best for your use case.

* Be concise
* Be specific and well-defined
* Ask one task at a time
* Turn generative tasks into classification tasks
* Improve response quality by including examples

https://www.cloudskillsboost.google/course_sessions/3264154/video/383122

Zero-shot

2023-07-04T10:13:18Z

Jboscher :

Zero-shot prompting - is a method where the LLM is given no additional data on the specifictask that it is being asked to perform. Instead, it is only given a prompt that describes the task. For example, if you want the LLM to answer a question, you just prompt "what is prompt design?". One-shot prompting - is a method where the LLM is given a single example of the task that it is being asked to perform. For example, if you want the LLM to write a poem, you might provide a single example poem. and Few-shot prompting - is a method where the LLM is given a small number of examples of the task that it is being asked to perform. For example, if you want the LLM to write a news article, you might give it a few news articles to read. You can use the structured mode to design the few-shot prompting by providing a context and additional examples for the model to learn from. The structured prompt contains a few different components:

First we have the context, which instructs how the model should respond. You can specify words the model can or cannot use, topics to focus on or avoid, or a particular response format. And the context applies each time you send a request to the model. Let’s say we want to use an LLM to answer questions based on some background text. In this case, a passage that describes changes in rainforest vegetation in the Amazon. We can paste in the background text as the context. Then, we add some examples of questions that could be answered from this passage Like what does LGM stand for? Or what did the analysis from the sediment deposits indicate? We’ll need to add in the corresponding answers to these questions, to demonstrate how we want the model to respond. Then, we can test out the prompt we’ve designed by sending a new question as input. And there you go, you’ve prototyped a q&a system based on background text in just a few minutes! Please note a few best practices around prompt design. Be concise Be specific and well-defined Ask one task at a time Turn generative tasks into classification tasks. For example, instead of asking what programming language to learn, ask if Python, Java, or C is a better fit for a beginner in programming. and Improve response quality by including examples. Adding instructions and a few examples tends to yield good results however there’s currently no one best way to write a prompt. You may need to experiment with different structures, formats, and examples to see what works best for your use case

- Zero-shot prompting: Provides one single command with no examples.
- One-shot prompting: Provides one example of the task.
- Few-shot prompting: Provides a few examples of the task often with the description of the context.

https://www.cloudskillsboost.google/course_sessions/3264154/video/383122

Prompt design

2023-07-04T10:12:35Z

Jboscher :

The process of finding and designing the best input text to get the desired response back from the model is called prompt design, which often involves a lot of experimentation.

Please note a few best practices around prompt design. Be concise Be specific and well-defined Ask one task at a time Turn generative tasks into classification tasks. For example, instead of asking what programming language to learn, ask if Python, Java, or C is a better fit for a beginner in programming. and Improve response quality by including examples. Adding instructions and a few examples tends to yield good results however there’s currently no one best way to write a prompt. You may need to experiment with different structures, formats, and examples to see what works best for your use case.

https://www.cloudskillsboost.google/course_sessions/3264154/video/383122

Prompt design

2023-07-04T09:20:33Z

Jboscher : Page créée avec « Please note a few best practices around prompt design. Be concise Be specific and well-defined Ask one task at a time Turn generative tasks into classification tasks. For example, instead of asking what programming language to learn, ask if Python, Java, or C is a better fit for a beginner in programming. and Improve response quality by including examples. Adding instructions and a few examples tends to yield good results however there’s currently no one best w... »

Please note a few best practices around prompt design. Be concise Be specific and well-defined Ask one task at a time Turn generative tasks into classification tasks. For example, instead of asking what programming language to learn, ask if Python, Java, or C is a better fit for a beginner in programming. and Improve response quality by including examples. Adding instructions and a few examples tends to yield good results however there’s currently no one best way to write a prompt. You may need to experiment with different structures, formats, and examples to see what works best for your use case.

https://www.cloudskillsboost.google/course_sessions/3264154/video/383122

Zero-shot

2023-07-04T08:58:15Z

Jboscher : Page créée avec « Zero-shot prompting - is a method where the LLM is given no additional data on the specifictask that it is being asked to perform. Instead, it is only given a prompt that describes the task. For example, if you want the LLM to answer a question, you just prompt "what is prompt design?". One-shot prompting - is a method where the LLM is given a single example of the task that it is being asked to perform. For example, if you want the LLM to write a poem, you might... »

Zero-shot prompting - is a method where the LLM is given no additional data on the specifictask that it is being asked to perform. Instead, it is only given a prompt that describes the task. For example, if you want the LLM to answer a question, you just prompt "what is prompt design?". One-shot prompting - is a method where the LLM is given a single example of the task that it is being asked to perform. For example, if you want the LLM to write a poem, you might provide a single example poem. and Few-shot prompting - is a method where the LLM is given a small number of examples of the task that it is being asked to perform. For example, if you want the LLM to write a news article, you might give it a few news articles to read. You can use the structured mode to design the few-shot prompting by providing a context and additional examples for the model to learn from. The structured prompt contains a few different components:

First we have the context, which instructs how the model should respond. You can specify words the model can or cannot use, topics to focus on or avoid, or a particular response format. And the context applies each time you send a request to the model. Let’s say we want to use an LLM to answer questions based on some background text. In this case, a passage that describes changes in rainforest vegetation in the Amazon. We can paste in the background text as the context. Then, we add some examples of questions that could be answered from this passage Like what does LGM stand for? Or what did the analysis from the sediment deposits indicate? We’ll need to add in the corresponding answers to these questions, to demonstrate how we want the model to respond. Then, we can test out the prompt we’ve designed by sending a new question as input. And there you go, you’ve prototyped a q&a system based on background text in just a few minutes! Please note a few best practices around prompt design. Be concise Be specific and well-defined Ask one task at a time Turn generative tasks into classification tasks. For example, instead of asking what programming language to learn, ask if Python, Java, or C is a better fit for a beginner in programming. and Improve response quality by including examples. Adding instructions and a few examples tends to yield good results however there’s currently no one best way to write a prompt. You may need to experiment with different structures, formats, and examples to see what works best for your use case

https://www.cloudskillsboost.google/course_sessions/3264154/video/383122

Top P

2023-06-22T08:30:29Z

Jboscher : Page créée avec « First, there are different models you can choose from. Each model is tuned to perform well on specific tasks. You can also specify the temperature, top P, and top K. These parameters all adjust the randomness of responses by controlling how the output tokens are selected. When you send a prompt to the model, it produces an array of probabilities over the words that could come next. And from this array, we need some strategy to decide what to return. A simple stra... »

First, there are different models you can choose from. Each model is tuned to perform well on specific tasks. You can also specify the temperature, top P, and top K. These parameters all adjust the randomness of responses by controlling how the output tokens are selected. When you send a prompt to the model, it produces an array of probabilities over the words that could come next. And from this array, we need some strategy to decide what to return. A simple strategy might be to select the most likely word at every timestep.

But this method can result in uninteresting and sometimes repetitive answers. On the contrary, if you randomly sample over the distribution returned by the model, you might get some unlikely responses.

By controlling the degree of randomness, you can get more unexpected, and some might say creative, responses. Back to the model parameters, temperature is a number used to tune the degree of randomness.

Low temperature: Means to select the words that are highly possible and more predictable. In this case, those are flowers and the other words that are located at the beginning ofthe list. This setting is generally better for tasks like q&a and summarization where you expect a more “predictable” answer with less variation. … High temperature: Means to select the words that have low possibility and are more unusual. In this case, those are bugs and the other words that that are located at the end of the list. This setting is good if you want to generate more “creative” or unexpected content.

In addition to adjusting the temperature, top K lets the model randomly return a word from the top K number of words in terms of possibility. For example, top 2 means you get a random word from the top 2 possible words including flowers and trees. This approach allows the other high-scoring word a chance of being selected. However, if the probability distribution of the words is highly skewed and you have one word that is very likely and everything else is very unlikely, this approach can result in some strange responses. The difficulty of selecting the best top-k value, leads to another popular approach that dynamically sets the size of the shortlist of words.

Top P allows the model to randomly return a word from the top P probability of words. With top P, you choose from a set of words with the sum of the likelihoods not exceeding P. For example, p of 0.75 means you sample from a set of words that have a cumulative probability greater than 0.75. In this case, it includes three words: flowers, trees, and herbs.

This way, the size of the set of words can dynamically increase and decrease according to the probability distribution of the next word on the list. In sum, Generative AI Studio provides a few model parameters for you to play with such as the model, temperature, top K, and top P. Note that, you are not required to adjust them constantly, especially top k and top p. Now let’s look at the second feature, which creates conversations.

https://www.cloudskillsboost.google/course_sessions/3264154/video/381925

Top K

2023-06-22T08:28:45Z

First, there are different models you can choose from. Each model is tuned to perform well on specific tasks. You can also specify the temperature, top P, and top K. These parameters all adjust the randomness of responses by controlling how the output tokens are selected. When you send a prompt to the model, it produces an array of probabilities over the words that could come next. And from this array, we need some strategy to decide what to return. A simple strategy might be to select the most likely word at every timestep.

But this method can result in uninteresting and sometimes repetitive answers. On the contrary, if you randomly sample over the distribution returned by the model, you might get some unlikely responses.

By controlling the degree of randomness, you can get more unexpected, and some might say creative, responses. Back to the model parameters, temperature is a number used to tune the degree of randomness.

Low temperature: Means to select the words that are highly possible and more predictable. In this case, those are flowers and the other words that are located at the beginning ofthe list. This setting is generally better for tasks like q&a and summarization where you expect a more “predictable” answer with less variation. … High temperature: Means to select the words that have low possibility and are more unusual. In this case, those are bugs and the other words that that are located at the end of the list. This setting is good if you want to generate more “creative” or unexpected content.

In addition to adjusting the temperature, top K lets the model randomly return a word from the top K number of words in terms of possibility. For example, top 2 means you get a random word from the top 2 possible words including flowers and trees. This approach allows the other high-scoring word a chance of being selected. However, if the probability distribution of the words is highly skewed and you have one word that is very likely and everything else is very unlikely, this approach can result in some strange responses. The difficulty of selecting the best top-k value, leads to another popular approach thatdynamically sets the size of the shortlist of words.

https://www.cloudskillsboost.google/course_sessions/3264154/video/381925

Temperature

2023-06-22T08:24:10Z

LangChain

2023-05-11T04:03:50Z

Jboscher :

LangChain est un framework qui sert à développer des applications qui s'interfacent avec des LLM (Large Language Models).

LangChain peut accéder à d'autres sources de données que celles du LLM et peut interagir avec son environnement.

LangChain makes the complicated parts of working & building with AI models easier. It helps do this in two ways:

Integration - Bring external data, such as your files, other applications, and api data, to your LLMs
Agency - Allow your LLMs to interact with its environment via decision making. Use LLMs to help decide which action to take next [3]

== Références ==

* [https://python.langchain.com/en/latest/index.html] Documentation LangChain
* [https://www.youtube.com/watch?v=2xxziIWmaSA&list=PLqZXAkvF1bPNQER9mLmDbntNfSpzdDIU5&index=4] Cours LangChain
* [https://github.com/ia35/langchain-tutorials/blob/main/LangChain%20Cookbook%20Part%201%20-%20Fundamentals.ipynb] Notebook LangChain

LangChain

2023-05-11T04:03:27Z

Jboscher :

LangChain est un framework qui sert à développer des applications qui s'interfacent avec des LLM (Large Language Models).

LangChain peut accéder à d'autres sources de données que celles du LLM et peut interagir avec son environnement.

LangChain makes the complicated parts of working & building with AI models easier. It helps do this in two ways:

Integration - Bring external data, such as your files, other applications, and api data, to your LLMs
Agency - Allow your LLMs to interact with it's environment via decision making. Use LLMs to help decide which action to take next

== Références ==

* [https://python.langchain.com/en/latest/index.html] Documentation LangChain
* [https://www.youtube.com/watch?v=2xxziIWmaSA&list=PLqZXAkvF1bPNQER9mLmDbntNfSpzdDIU5&index=4] Cours LangChain
* [https://github.com/ia35/langchain-tutorials/blob/main/LangChain%20Cookbook%20Part%201%20-%20Fundamentals.ipynb] Notebook LangChain

LangChain

2023-05-11T03:45:02Z

Jboscher :

LangChain est un framework qui sert à développer des applications qui s'interfacent avec des LLM (Large Language Models).

LangChain peut accéder à d'autres sources de données que celles du LLM et peut interagit avec son environnement.

== Références ==

* [https://python.langchain.com/en/latest/index.html] Documentation LangChain
* [https://www.youtube.com/watch?v=2xxziIWmaSA&list=PLqZXAkvF1bPNQER9mLmDbntNfSpzdDIU5&index=4] Cours LangChain

LangChain

2023-05-11T03:30:09Z

Jboscher : Page créée avec «  == Références == * [https://python.langchain.com/en/latest/index.html] Documentation LangChain * [https://www.youtube.com/watch?v=2xxziIWmaSA&list=PLqZXAkvF1bPNQER9mLmDbntNfSpzdDIU5&index=4] Cours LangChain »

== Références ==

* [https://python.langchain.com/en/latest/index.html] Documentation LangChain
* [https://www.youtube.com/watch?v=2xxziIWmaSA&list=PLqZXAkvF1bPNQER9mLmDbntNfSpzdDIU5&index=4] Cours LangChain

Prompt

2023-05-02T12:43:18Z

Jboscher :

Un prompt est une commande textuelle transmise afin d'exécuter une action. Cette commande s'adresse principalement aux LLM (Large Language Model) ou aux logiciels de génération d'images entraînées pour comprendre ces commandes.

Exemples:

Un prompt basique, le LLM (Large Language Model) poursuit le texte commencé.

Prompt:
Four score and seven years ago our

Réponse:
fathers brought forth on this continent, a new nation, conceived in Liberty, and dedicated to the proposition that all men are created

Les prompts peuvent être aussi des instructions simples ou complexes.

Prompt:
Write an introduction for a weekly newsletter.

Réponse:
Welcome to our weekly newsletter. This week we have a special focus on productivity. We'll be sharing tips and tricks on how to be productive in all areas of your life. We hope you enjoy and please share with your friends!

ou dans un cas plus complexe :

Prompt:
Write an introduction for a weekly newsletter to my company named Contoso. It should mention the great all hands meeting we had last week and thank the team for all their hard work over the tough past few months.

Réponse:
Welcome to Contoso's weekly newsletter! We had a great all hands meeting last week and I want to thank the team for all their hard work over the tough past few months. We're looking forward to a bright future and I'm confident that with our team's dedication and commitment, we will achieve great things. Thank you for being a part of our team and I look forward to working with you all in the coming weeks and months.

On peut aussi écrire des prompts du type "Primary Content" où un texte est modifié.

Prompt:
Can you please tell me how to get to the museum?
Translate to French:

Réponse:
Pouvez-vous s'il vous plaît me dire comment aller au musée?

Prompt engineering is a relatively new discipline for developing and optimizing prompts to efficiently use language models (LMs) for a wide variety of applications and research topics. Prompt engineering skills help to better understand the capabilities and limitations of large language models (LLMs). Researchers use prompt engineering to improve the capacity of LLMs on a wide range of common and complex tasks such as question answering and arithmetic reasoning. Developers use prompt engineering to design robust and effective prompting techniques that interface with LLMs and other tools.[4]

Ecrire un prompt, contrairement à ce qu'on peut imaginer, n'est pas chose facile. Il faut faire l'analogie avec le piano. Obtenir des notes est facile, bien jouer est très compliqué.

While these models are extremely powerful, their behavior is also very sensitive to the prompt. This makes prompt construction an important skill to develop.
Prompt construction can be difficult. In practice, the prompt acts to configure the model weights to complete the desired task, but it's more of an art than a science, often requiring experience and intuition to craft a successful prompt. [1]

== Références ==

* [https://learn.microsoft.com/en-us/azure/cognitive-services/openai/concepts/prompt-engineering] Introduction to prompt engineering
* [https://www.deeplearning.ai/short-courses/chatgpt-prompt-engineering-for-developers/] ChatGPT Prompt Engineering for Developers
* [https://learn.microsoft.com/en-us/azure/cognitive-services/openai/concepts/advanced-prompt-engineering?pivots=programming-language-chat-completions] Prompt engineering techniques
* [https://github.com/dair-ai/Prompt-Engineering-Guide] Prompt Engineering Guide
* [https://www.promptingguide.ai/fr] Prompt Engineering Guide (fr)
* [https://www.youtube.com/watch?v=dOxUroR57xs] Prompt Engineering Overview
* [https://learn.microsoft.com/en-us/azure/cognitive-services/openai/how-to/chatgpt?pivots=programming-language-chat-completions] Learn how to work with the ChatGPT and GPT-4 models

Prompt

2023-05-02T09:52:44Z

Jboscher :

Un prompt est une commande textuelle transmise afin d'exécuter une action. Cette commande s'adresse principalement aux LLM (Large Language Model) ou aux logiciels de génération d'images entraînées pour comprendre ces commandes.

Prompt engineering is a relatively new discipline for developing and optimizing prompts to efficiently use language models (LMs) for a wide variety of applications and research topics. Prompt engineering skills help to better understand the capabilities and limitations of large language models (LLMs). Researchers use prompt engineering to improve the capacity of LLMs on a wide range of common and complex tasks such as question answering and arithmetic reasoning. Developers use prompt engineering to design robust and effective prompting techniques that interface with LLMs and other tools.[4]

Ecrire un prompt, contrairement à ce qu'on peut imaginer, n'est pas chose facile. Il faut faire l'analogie avec le piano. Obtenir des notes est facile, bien jouer est très compliqué.

While these models are extremely powerful, their behavior is also very sensitive to the prompt. This makes prompt construction an important skill to develop.
Prompt construction can be difficult. In practice, the prompt acts to configure the model weights to complete the desired task, but it's more of an art than a science, often requiring experience and intuition to craft a successful prompt. [1]

== Références ==

* [https://learn.microsoft.com/en-us/azure/cognitive-services/openai/concepts/prompt-engineering] Introduction to prompt engineering
* [https://www.deeplearning.ai/short-courses/chatgpt-prompt-engineering-for-developers/] ChatGPT Prompt Engineering for Developers
* [https://learn.microsoft.com/en-us/azure/cognitive-services/openai/concepts/advanced-prompt-engineering?pivots=programming-language-chat-completions] Prompt engineering techniques
* [https://github.com/dair-ai/Prompt-Engineering-Guide] Prompt Engineering Guide
* [https://www.promptingguide.ai/fr] Prompt Engineering Guide (fr)
* [https://www.youtube.com/watch?v=dOxUroR57xs] Prompt Engineering Overview
* [https://learn.microsoft.com/en-us/azure/cognitive-services/openai/how-to/chatgpt?pivots=programming-language-chat-completions] Learn how to work with the ChatGPT and GPT-4 models

Prompt

2023-05-02T09:37:18Z

Jboscher :

Un prompt est une commande textuelle transmise afin d'exécuter une action. Cette commande s'adresse principalement aux LLM (Large Language Model) ou aux logiciels de génération d'images entraînées pour comprendre ces commandes.

Prompt engineering is a relatively new discipline for developing and optimizing prompts to efficiently use language models (LMs) for a wide variety of applications and research topics. Prompt engineering skills help to better understand the capabilities and limitations of large language models (LLMs). Researchers use prompt engineering to improve the capacity of LLMs on a wide range of common and complex tasks such as question answering and arithmetic reasoning. Developers use prompt engineering to design robust and effective prompting techniques that interface with LLMs and other tools.[4]

Ecrire un prompt, contrairement à ce qu'on peut imaginer, n'est pas chose facile. Il faut faire l'analogie avec le piano. Obtenir des notes est facile, bien jouer est très compliqué.

While these models are extremely powerful, their behavior is also very sensitive to the prompt. This makes prompt construction an important skill to develop.
Prompt construction can be difficult. In practice, the prompt acts to configure the model weights to complete the desired task, but it's more of an art than a science, often requiring experience and intuition to craft a successful prompt. [1]

== Références ==

* [https://learn.microsoft.com/en-us/azure/cognitive-services/openai/concepts/prompt-engineering] Introduction to prompt engineering
* [https://www.deeplearning.ai/short-courses/chatgpt-prompt-engineering-for-developers/] ChatGPT Prompt Engineering for Developers
* [https://learn.microsoft.com/en-us/azure/cognitive-services/openai/concepts/advanced-prompt-engineering?pivots=programming-language-chat-completions] Prompt engineering techniques
* [https://github.com/dair-ai/Prompt-Engineering-Guide] Prompt Engineering Guide
* [https://www.promptingguide.ai/fr] Prompt Engineering Guide (fr)
* [https://www.youtube.com/watch?v=dOxUroR57xs] Prompt Engineering Overview

LLaMA

2023-05-02T09:25:02Z

Jboscher : Page créée avec « LLaMA (Large Language Model Meta AI) is a language model released by Meta (Facebook). It is Meta’s answer to OpenAI’s GPT models. Like GPT, LLaMA is intended to be a general-purpose foundational model suitable for further fine-tuning. [1] == Références == * [https://agi-sphere.com/llama-models/] A brief history of LLaMA models »

LLaMA (Large Language Model Meta AI) is a language model released by Meta (Facebook). It is Meta’s answer to OpenAI’s GPT models.

Like GPT, LLaMA is intended to be a general-purpose foundational model suitable for further fine-tuning. [1]

== Références ==

* [https://agi-sphere.com/llama-models/] A brief history of LLaMA models

Prompt

2023-05-02T08:56:45Z

Jboscher :

Un prompt est une commande textuelle transmise afin d'exécuter une action. Cette commande s'adresse principalement aux LLM (Large Language Model) ou aux logiciels de génération d'images entraînées pour comprendre ces commandes.

Prompt engineering is a relatively new discipline for developing and optimizing prompts to efficiently use language models (LMs) for a wide variety of applications and research topics. Prompt engineering skills help to better understand the capabilities and limitations of large language models (LLMs). Researchers use prompt engineering to improve the capacity of LLMs on a wide range of common and complex tasks such as question answering and arithmetic reasoning. Developers use prompt engineering to design robust and effective prompting techniques that interface with LLMs and other tools.[4]

== Références ==

* [https://learn.microsoft.com/en-us/azure/cognitive-services/openai/concepts/prompt-engineering] Introduction to prompt engineering
* [https://www.deeplearning.ai/short-courses/chatgpt-prompt-engineering-for-developers/] ChatGPT Prompt Engineering for Developers
* [https://learn.microsoft.com/en-us/azure/cognitive-services/openai/concepts/advanced-prompt-engineering?pivots=programming-language-chat-completions] Prompt engineering techniques
* [https://github.com/dair-ai/Prompt-Engineering-Guide] Prompt Engineering Guide
* [https://www.promptingguide.ai/fr] Prompt Engineering Guide (fr)

Fichier:Cropped-cropped-LogoBackPropTrans-300x120-3.png

2023-05-01T08:02:22Z

Jboscher : Logo BackProp transparent

== Description ==
Logo BackProp transparent

Prompt

2023-05-01T07:36:49Z

Jboscher :

Un prompt est une commande textuelle transmise afin d'exécuter une action. Cette commande s'adresse principalement aux LLM (Large Language Model) ou aux logiciels de génération d'images entraînées pour comprendre ces commandes.

Prompt engineering is a relatively new discipline for developing and optimizing prompts to efficiently use language models (LMs) for a wide variety of applications and research topics. Prompt engineering skills help to better understand the capabilities and limitations of large language models (LLMs). Researchers use prompt engineering to improve the capacity of LLMs on a wide range of common and complex tasks such as question answering and arithmetic reasoning. Developers use prompt engineering to design robust and effective prompting techniques that interface with LLMs and other tools.[4]

== Références ==

* [https://learn.microsoft.com/en-us/azure/cognitive-services/openai/concepts/prompt-engineering] Introduction to prompt engineering
* [https://www.deeplearning.ai/short-courses/chatgpt-prompt-engineering-for-developers/] ChatGPT Prompt Engineering for Developers
* [https://learn.microsoft.com/en-us/azure/cognitive-services/openai/concepts/advanced-prompt-engineering?pivots=programming-language-chat-completions] Prompt engineering techniques
* [https://github.com/dair-ai/Prompt-Engineering-Guide] Prompt Engineering Guide

Prompt

2023-05-01T07:33:15Z

Jboscher :

Un prompt est une commande textuelle transmise afin d'exécuter une action. Cette commande s'adresse principalement aux LLM (Large Language Model) ou aux logiciels de génération d'images entraînées pour comprendre ces commandes.

Prompt engineering is a relatively new discipline for developing and optimizing prompts to efficiently use language models (LMs) for a wide variety of applications and research topics. Prompt engineering skills help to better understand the capabilities and limitations of large language models (LLMs). Researchers use prompt engineering to improve the capacity of LLMs on a wide range of common and complex tasks such as question answering and arithmetic reasoning. Developers use prompt engineering to design robust and effective prompting techniques that interface with LLMs and other tools.[4]

== Références ==

* [https://learn.microsoft.com/en-us/azure/cognitive-services/openai/concepts/prompt-engineering] Introduction to prompt engineering
* [https://www.deeplearning.ai/short-courses/chatgpt-prompt-engineering-for-developers/] ChatGPT Prompt Engineering for Developers
* [https://learn.microsoft.com/en-us/azure/cognitive-services/openai/concepts/advanced-prompt-engineering?pivots=programming-language-chat-completions] Prompt engineering techniques
* [https://github.com/dair-ai/Prompt-Engineering-Guide] Prompt Engineering Guide

Prompt

2023-05-01T07:29:28Z

Jboscher : Page créée avec «  == Références == * [https://learn.microsoft.com/en-us/azure/cognitive-services/openai/concepts/prompt-engineering] Introduction to prompt engineering * [https://www.deeplearning.ai/short-courses/chatgpt-prompt-engineering-for-developers/] ChatGPT Prompt Engineering for Developers * [https://learn.microsoft.com/en-us/azure/cognitive-services/openai/concepts/advanced-prompt-engineering?pivots=programming-language-chat-completions] Prompt engineering techniques... »

== Références ==

* [https://learn.microsoft.com/en-us/azure/cognitive-services/openai/concepts/prompt-engineering] Introduction to prompt engineering
* [https://www.deeplearning.ai/short-courses/chatgpt-prompt-engineering-for-developers/] ChatGPT Prompt Engineering for Developers
* [https://learn.microsoft.com/en-us/azure/cognitive-services/openai/concepts/advanced-prompt-engineering?pivots=programming-language-chat-completions] Prompt engineering techniques
* [https://github.com/dair-ai/Prompt-Engineering-Guide] Prompt Engineering Guide

In-context learning

2023-04-27T16:40:37Z

Jboscher :

[1] The in-context learning (ICL) ability is formally introduced by GPT-3 : assuming that the language model has been provided with a natural language instruction and/or several task demonstrations, it can generate the expected output for the test instances by completing the word sequence of input text, without requiring additional training or gradient update

[[File:A comparative illustration of in-context learning (ICL) and chain-of-thought (CoT) prompting.jpg|500px]]

== Références ==

* [https://arxiv.org/pdf/2303.18223.pdf] A Survey of Large Language Models

In-context learning

2023-04-27T16:40:02Z

Jboscher :

[1] The in-context learning (ICL) ability is formally introduced by GPT-3 : assuming that the language model has been provided with a natural language instruction and/or several task demonstrations, it can generate the expected output for the test instances by completing the word sequence of input text, without requiring additional training or gradient update

[[File:Capture d’écran 2023-04-27 à 18.32.50.png|500px]]

== Références ==

* [https://arxiv.org/pdf/2303.18223.pdf] A Survey of Large Language Models

In-context learning

2023-04-27T16:38:44Z

Jboscher :

Fichier:Capture d’écran 2023-04-27 à 18.32.50.png

2023-04-27T16:36:05Z

Jboscher : A comparative illustration of in-context learning (ICL) and chain-of-thought (CoT) prompting

== Description ==
A comparative illustration of in-context learning (ICL) and chain-of-thought (CoT) prompting

In-context learning

2023-04-27T16:35:13Z

Jboscher : Page créée avec «  [1] The in-context learning (ICL) ability is formally introduced by GPT-3 : assuming that the language model has been provided with a natural language instruction and/or several task demonstrations, it can generate the expected output for the test instances by completing the word sequence of input text, without requiring additional training or gradient update == Références == * [https://arxiv.org/pdf/2303.18223.pdf] A Survey of Large Language Models »

[1] The in-context learning (ICL) ability is formally introduced by GPT-3 : assuming that the language model has been provided with a natural language instruction and/or several task demonstrations, it can generate the expected output for the test instances by completing the word sequence of input text, without requiring additional training or gradient update

== Références ==

* [https://arxiv.org/pdf/2303.18223.pdf] A Survey of Large Language Models

Emergent Abilities of Large Language Models

2023-04-27T16:23:12Z

Jboscher :

On entend par "Emergent Abilities of Large Language Models" une capacité présente dans un LLM qui ne se retrouve pas dans un modèle similaire mais plus petit. Ce qui veut dire aussi qu'on ne peut pas prévoir (extrapoler) cette nouvelle capacité uniquement à partir de celles d'un modèle plus petit.

[1] We consider an ability to be emergent if it is not present in smaller models but is present in larger models.

[1] We will consider the following general definition of emergence, adapted from Steinhardt (2022) and rooted in a 1972 essay called “More Is Different” by Nobel prize-winning physicist Philip Anderson

[1] Emergence is when quantitative changes in a system result in qualitative changes in behavior.

Les paramètres qui servent à changer d'échelle pour les modèles sont : la taille du dataset, le nombre de paramètres du modèle et la puissance de calcul requise pour l'entraîner (les deux derniers étant souvent corrélés).

[2] Although scaling is mainly conducted in model size (with similar architectures and pre-training tasks), these large-sized PLMs display different behaviors from smaller PLMs (e.g., 330M-parameter BERT and 1.5B- parameter GPT-2) and show surprising abilities (called emergent abilities) in solving a series of complex tasks.

[2] For example, GPT-3 can solve few-shot tasks through in-context learning, whereas GPT-2 cannot do well.

== Références ==

* [https://openreview.net/pdf?id=yzkSU5zdwD] Emergent Abilities of Large Language Models
* [https://arxiv.org/pdf/2303.18223.pdf] A Survey of Large Language Models

Emergent Abilities of Large Language Models

2023-04-27T15:58:23Z

Jboscher :

On entend par "Emergent Abilities of Large Language Models" une capacité présente dans un LLM qui ne se retrouve pas dans un modèle similaire mais plus petit. Ce qui veut dire aussi qu'on ne peut pas prévoir (extrapoler) cette nouvelle capacité uniquement à partir de celles d'un modèle plus petit.

[1] We consider an ability to be emergent if it is not present in smaller models but is present in larger models.

[1] We will consider the following general definition of emergence, adapted from Steinhardt (2022) and rooted in a 1972 essay called “More Is Different” by Nobel prize-winning physicist Philip Anderson

[2] Although scaling is mainly conducted in model size (with similar architectures and pre-training tasks), these large-sized PLMs display different behaviors from smaller PLMs (e.g., 330M-parameter BERT and 1.5B- parameter GPT-2) and show surprising abilities (called emergent abilities) in solving a series of complex tasks.

[2] For example, GPT-3 can solve few-shot tasks through in-context learning, whereas GPT-2 cannot do well.

== Références ==

* [https://openreview.net/pdf?id=yzkSU5zdwD] Emergent Abilities of Large Language Models
* [https://arxiv.org/pdf/2303.18223.pdf] A Survey of Large Language Models

Emergent Abilities of Large Language Models

2023-04-27T15:39:06Z

Jboscher : Page créée avec « On entend par "Emergent Abilities of Large Language Models" une capacité présente dans un LLM qui ne se retrouve pas dans un modèle similaire mais plus petit. Ce qui veut dire aussi qu'on ne peut pas prévoir (extrapoler) cette nouvelle capacité uniquement à partir de celles d'un modèle plus petit. [1] We consider an ability to be emergent if it is not present in smaller models but is present in larger models. Although scaling is mainly conducted in mode... »

On entend par "Emergent Abilities of Large Language Models" une capacité présente dans un LLM qui ne se retrouve pas dans un modèle similaire mais plus petit. Ce qui veut dire aussi qu'on ne peut pas prévoir (extrapoler) cette nouvelle capacité uniquement à partir de celles d'un modèle plus petit.

[1] We consider an ability to be emergent if it is not present in smaller models but is present in larger models.

Although scaling is mainly conducted in model size (with similar architectures and pre-training tasks), these large-sized PLMs display different behaviors from smaller PLMs (e.g., 330M-parameter BERT and 1.5B- parameter GPT-2) and show surprising abilities (called emergent abilities) in solving a series of complex tasks.

For example, GPT-3 can solve few-shot tasks through in-context learning, whereas GPT-2 cannot do well.

Thus, the research community coins the term “large language models (LLM)”1 for these large-sized PLMs [32–35]. A remarkable application of LLMs is ChatGPT2 that adapts the LLMs from the GPT series for dialogue, which presents an amazing conversation ability with humans.

== Références ==

* [https://openreview.net/pdf?id=yzkSU5zdwD] Emergent Abilities of Large Language Models

Pre-trained language models

2023-04-27T14:53:16Z

Jboscher :

Les auteurs de "A Survey of Large Language Models" distinguent les "Pre-trained language models (PLM)" des Large language models (LLM).

ELMo et BERT appartiendraient à la 1ère catégorie, un peu comme les pionniers des LLM.

As an early attempt, ELMo was proposed to capture context-aware word representations by first pre-training a bidirectional LSTM (biLSTM) network (instead of learning fixed word representations) and then fine-tuning the biLSTM network according to specific downstream tasks.

Further, based on the highly parallelizable Transformer architecture [22] with self-attention mechanisms, BERT [23] was proposed by pre- training bidirectional language models with specially de- signed pre-training tasks on large-scale unlabeled corpora.

La différence entre les PLM et les LLM c'est l'augmentation de taille des données et des modèles, qui a pour effet d'améliorer significativement les performances sur les tâches.

Researchers find that scaling PLM (e.g., scaling model size or data size) often leads to an improved model capacity on downstream tasks

== Références ==

* [https://arxiv.org/pdf/1802.05365.pdf] Deep contextualized word representations
* [https://aclanthology.org/N19-1423.pdf] BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

Pre-trained language models

2023-04-27T14:44:33Z

Jboscher : Page créée avec « Les auteurs de "A Survey of Large Language Models" distinguent les "Pre-trained language models (PLM)" des Large language models (LLM). ELMo et BERT appartiendraient à la 1ère catégorie, un peu comme les pionniers des LLM. As an early attempt, ELMo was proposed to capture context-aware word representations by first pre-training a bidirectional LSTM (biLSTM) network (instead of learning fixed word representations) and then fine-tuning the biLSTM network acco... »

Fichier:Cropped-LogoBackPropTrans-2.png

2023-04-27T13:31:55Z

Jboscher : Logo BackProp

== Description ==
Logo BackProp

Fichier:Shamentality artificial intelligence beautiful textures and ter 545b2bf2-aece-4268-b49e-78dee7f78862.png

2023-04-27T13:29:41Z

Jboscher : AI logo by MidJourney

== Description ==
AI logo by MidJourney

Reinforcement Learning from Human Feedback

2023-01-22T21:59:21Z

Jboscher : Page créée avec « Reinforcement Learning from Human Feedback (RLHF) https://huggingface.co/blog/rlhf »

Reinforcement Learning from Human Feedback (RLHF)

https://huggingface.co/blog/rlhf

CLIP

2023-01-12T16:21:08Z

Jboscher :

CLIP pour (Contrastive Language–Image Pre-training) a été introduit par OpenAi en 2021.

L'objectif est de fournir un text (caption) pour une nouvelle image présentée lors de la phase d'inférence.

Cette méthode CLIP permet de contourner certains problèmes de l'apprentissage supervisé, par exemple, ceux liés aux coûts des Datasets (voir [[ImageNet]]), à leur limitation de prédiction à l'ensemble des catégories entrainées, et à leur faible performance (parfois).

L'apprentissage décrit par OpenAi s'est fait avec 256 GPU durant 2 semaines.
Il semble que 400M de paires(caption/image) aient été utilisées pour l'apprentissage.

" CLIP is an open source, multi-modal, zero-shot model. Given an image and text descriptions, the model can predict the most relevant text description for that image, without optimizing for a particular task."

https://wiki.backprop.fr/images/1*ag6qUFmmXAr4E410Ll-eSQ.jpeg

Voir le notebook à ce sujet : CLIP
https://colab.research.google.com/drive/1kr0xnHQZ7G-cvlFgl2DECmLUbpgvswcy?usp=sharing

== Références ==

* [https://openai.com/blog/clip/ CLIP: Connecting Text and Images]
* [https://www.kdnuggets.com/2021/03/beginners-guide-clip-model.html CLIP: A Beginner’s Guide to the CLIP Model]
* [https://github.com/openai/CLIP CLIP sur GitHub]
* [https://towardsdatascience.com/clip-the-most-influential-ai-model-from-openai-and-how-to-use-it-f8ee408958b1 CLIP: The Most Influential AI Model From OpenAI — And How To Use It]

CLIP

2023-01-10T16:26:48Z

Jboscher :

CLIP pour (Contrastive Language–Image Pre-training) a été introduit par OpenAi en 2021.

L'objectif est de fournir un text (caption) pour une nouvelle image présentée lors de la phase d'inférence.

Cette méthode CLIP permet de contourner certains problèmes de l'apprentissage supervisé, par exemple, ceux liés aux coûts des Datasets (voir [[ImageNet]]), à leur limitation de prédiction à l'ensemble des catégories entrainées, et à leur faible performance (parfois).

L'apprentissage décrit par OpenAi s'est fait avec 256 GPU durant 2 semaines.
Il semble que 400M de paires(caption/image) aient été utilisées pour l'apprentissage.

" CLIP is an open source, multi-modal, zero-shot model. Given an image and text descriptions, the model can predict the most relevant text description for that image, without optimizing for a particular task."

https://wiki.backprop.fr/images/1*ag6qUFmmXAr4E410Ll-eSQ.jpeg

== Références ==

* [https://openai.com/blog/clip/ CLIP: Connecting Text and Images]
* [https://www.kdnuggets.com/2021/03/beginners-guide-clip-model.html CLIP: A Beginner’s Guide to the CLIP Model]
* [https://github.com/openai/CLIP CLIP sur GitHub]
* [https://towardsdatascience.com/clip-the-most-influential-ai-model-from-openai-and-how-to-use-it-f8ee408958b1 CLIP: The Most Influential AI Model From OpenAI — And How To Use It]

CLIP

2023-01-10T16:26:21Z

Jboscher :

CLIP pour (Contrastive Language–Image Pre-training) a été introduit par OpenAi en 2021.

L'objectif est de fournir un text (caption) pour une nouvelle image présentée lors de la phase d'inférence.

Cette méthode CLIP permet de contourner certains problèmes de l'apprentissage supervisé, par exemple, ceux liés aux coûts des Datasets (voir [[ImageNet]]), à leur limitation de prédiction à l'ensemble des catégories entrainées, et à leur faible performance (parfois).

L'apprentissage décrit par OpenAi s'est fait avec 256 GPU durant 2 semaines.
Il semble que 400M de paires(caption/image) aient été utilisées pour l'apprentissage.

" CLIP is an open source, multi-modal, zero-shot model. Given an image and text descriptions, the model can predict the most relevant text description for that image, without optimizing for a particular task."

[[File:https://wiki.backprop.fr/images/1*ag6qUFmmXAr4E410Ll-eSQ.jpeg]]

== Références ==

* [https://openai.com/blog/clip/ CLIP: Connecting Text and Images]
* [https://www.kdnuggets.com/2021/03/beginners-guide-clip-model.html CLIP: A Beginner’s Guide to the CLIP Model]
* [https://github.com/openai/CLIP CLIP sur GitHub]
* [https://towardsdatascience.com/clip-the-most-influential-ai-model-from-openai-and-how-to-use-it-f8ee408958b1 CLIP: The Most Influential AI Model From OpenAI — And How To Use It]

CLIP

2023-01-10T16:21:23Z

Jboscher :

CLIP pour (Contrastive Language–Image Pre-training) a été introduit par OpenAi en 2021.

L'objectif est de fournir un text (caption) pour une nouvelle image présentée lors de la phase d'inférence.

Cette méthode CLIP permet de contourner certains problèmes de l'apprentissage supervisé, par exemple, ceux liés aux coûts des Datasets (voir [[ImageNet]]), à leur limitation de prédiction à l'ensemble des catégories entrainées, et à leur faible performance (parfois).

L'apprentissage décrit par OpenAi s'est fait avec 256 GPU durant 2 semaines.
Il semble que 400M de paires(caption/image) aient été utilisées pour l'apprentissage.

" CLIP is an open source, multi-modal, zero-shot model. Given an image and text descriptions, the model can predict the most relevant text description for that image, without optimizing for a particular task."

[[File:https://wiki.backprop.fr/images/1*ag6qUFmmXAr4E410Ll-eSQ.jpeg|caption]]

== Références ==

* [https://openai.com/blog/clip/ CLIP: Connecting Text and Images]
* [https://www.kdnuggets.com/2021/03/beginners-guide-clip-model.html CLIP: A Beginner’s Guide to the CLIP Model]
* [https://github.com/openai/CLIP CLIP sur GitHub]
* [https://towardsdatascience.com/clip-the-most-influential-ai-model-from-openai-and-how-to-use-it-f8ee408958b1 CLIP: The Most Influential AI Model From OpenAI — And How To Use It]

CLIP

2023-01-10T15:55:57Z

Jboscher :

CLIP pour (Contrastive Language–Image Pre-training) a été introduit par OpenAi en 2021.

L'objectif est de fournir un text (caption) pour une nouvelle image présentée lors de la phase d'inférence.

Cette méthode CLIP permet de contourner certains problèmes de l'apprentissage supervisé, par exemple, ceux liés aux coûts des Datasets (voir [[ImageNet]]), à leur limitation de prédiction à l'ensemble des catégories entrainées, et à leur faible performance (parfois).

L'apprentissage décrit par OpenAi s'est fait avec 256 GPU durant 2 semaines.
Il semble que 400M de paires(caption/image) aient été utilisées pour l'apprentissage.

" CLIP is an open source, multi-modal, zero-shot model. Given an image and text descriptions, the model can predict the most relevant text description for that image, without optimizing for a particular task."

[[File:1*ag6qUFmmXAr4E410Ll-eSQ.jpeg|caption]]

== Références ==

* [https://openai.com/blog/clip/ CLIP: Connecting Text and Images]
* [https://www.kdnuggets.com/2021/03/beginners-guide-clip-model.html CLIP: A Beginner’s Guide to the CLIP Model]
* [https://github.com/openai/CLIP CLIP sur GitHub]
* [https://towardsdatascience.com/clip-the-most-influential-ai-model-from-openai-and-how-to-use-it-f8ee408958b1 CLIP: The Most Influential AI Model From OpenAI — And How To Use It]

CLIP

2023-01-10T15:55:17Z

Jboscher :

CLIP pour (Contrastive Language–Image Pre-training) a été introduit par OpenAi en 2021.

L'objectif est de fournir un text (caption) pour une nouvelle image présentée lors de la phase d'inférence.

Cette méthode CLIP permet de contourner certains problèmes de l'apprentissage supervisé, par exemple, ceux liés aux coûts des Datasets (voir [[ImageNet]]), à leur limitation de prédiction à l'ensemble des catégories entrainées, et à leur faible performance (parfois).

L'apprentissage décrit par OpenAi s'est fait avec 256 GPU durant 2 semaines.
Il semble que 400M de paires(caption/image) aient été utilisées pour l'apprentissage.

" CLIP is an open source, multi-modal, zero-shot model. Given an image and text descriptions, the model can predict the most relevant text description for that image, without optimizing for a particular task."

[[File:images/1*ag6qUFmmXAr4E410Ll-eSQ.jpeg|caption]]

== Références ==

* [https://openai.com/blog/clip/ CLIP: Connecting Text and Images]
* [https://www.kdnuggets.com/2021/03/beginners-guide-clip-model.html CLIP: A Beginner’s Guide to the CLIP Model]
* [https://github.com/openai/CLIP CLIP sur GitHub]
* [https://towardsdatascience.com/clip-the-most-influential-ai-model-from-openai-and-how-to-use-it-f8ee408958b1 CLIP: The Most Influential AI Model From OpenAI — And How To Use It]

CLIP

2023-01-10T15:54:55Z

Jboscher :

CLIP pour (Contrastive Language–Image Pre-training) a été introduit par OpenAi en 2021.

L'objectif est de fournir un text (caption) pour une nouvelle image présentée lors de la phase d'inférence.

Cette méthode CLIP permet de contourner certains problèmes de l'apprentissage supervisé, par exemple, ceux liés aux coûts des Datasets (voir [[ImageNet]]), à leur limitation de prédiction à l'ensemble des catégories entrainées, et à leur faible performance (parfois).

L'apprentissage décrit par OpenAi s'est fait avec 256 GPU durant 2 semaines.
Il semble que 400M de paires(caption/image) aient été utilisées pour l'apprentissage.

" CLIP is an open source, multi-modal, zero-shot model. Given an image and text descriptions, the model can predict the most relevant text description for that image, without optimizing for a particular task."

[[File:https://wiki.backprop.fr/images/1*ag6qUFmmXAr4E410Ll-eSQ.jpeg|caption]]

== Références ==

* [https://openai.com/blog/clip/ CLIP: Connecting Text and Images]
* [https://www.kdnuggets.com/2021/03/beginners-guide-clip-model.html CLIP: A Beginner’s Guide to the CLIP Model]
* [https://github.com/openai/CLIP CLIP sur GitHub]
* [https://towardsdatascience.com/clip-the-most-influential-ai-model-from-openai-and-how-to-use-it-f8ee408958b1 CLIP: The Most Influential AI Model From OpenAI — And How To Use It]

CLIP

2023-01-10T15:54:14Z

Jboscher :

CLIP pour (Contrastive Language–Image Pre-training) a été introduit par OpenAi en 2021.

L'objectif est de fournir un text (caption) pour une nouvelle image présentée lors de la phase d'inférence.

Cette méthode CLIP permet de contourner certains problèmes de l'apprentissage supervisé, par exemple, ceux liés aux coûts des Datasets (voir [[ImageNet]]), à leur limitation de prédiction à l'ensemble des catégories entrainées, et à leur faible performance (parfois).

L'apprentissage décrit par OpenAi s'est fait avec 256 GPU durant 2 semaines.
Il semble que 400M de paires(caption/image) aient été utilisées pour l'apprentissage.

" CLIP is an open source, multi-modal, zero-shot model. Given an image and text descriptions, the model can predict the most relevant text description for that image, without optimizing for a particular task."

... text text text
[[File:http://wiki.backprop.fr/images/1*ag6qUFmmXAr4E410Ll-eSQ.jpeg|caption]]
text text text ...

== Références ==

* [https://openai.com/blog/clip/ CLIP: Connecting Text and Images]
* [https://www.kdnuggets.com/2021/03/beginners-guide-clip-model.html CLIP: A Beginner’s Guide to the CLIP Model]
* [https://github.com/openai/CLIP CLIP sur GitHub]
* [https://towardsdatascience.com/clip-the-most-influential-ai-model-from-openai-and-how-to-use-it-f8ee408958b1 CLIP: The Most Influential AI Model From OpenAI — And How To Use It]

CLIP

2023-01-10T15:53:30Z

Jboscher :

CLIP pour (Contrastive Language–Image Pre-training) a été introduit par OpenAi en 2021.

L'objectif est de fournir un text (caption) pour une nouvelle image présentée lors de la phase d'inférence.

Cette méthode CLIP permet de contourner certains problèmes de l'apprentissage supervisé, par exemple, ceux liés aux coûts des Datasets (voir [[ImageNet]]), à leur limitation de prédiction à l'ensemble des catégories entrainées, et à leur faible performance (parfois).

L'apprentissage décrit par OpenAi s'est fait avec 256 GPU durant 2 semaines.
Il semble que 400M de paires(caption/image) aient été utilisées pour l'apprentissage.

" CLIP is an open source, multi-modal, zero-shot model. Given an image and text descriptions, the model can predict the most relevant text description for that image, without optimizing for a particular task."

[[File:http://wiki.backprop.fr/images/1*ag6qUFmmXAr4E410Ll-eSQ.jpeg|caption]]

== Références ==

* [https://openai.com/blog/clip/ CLIP: Connecting Text and Images]
* [https://www.kdnuggets.com/2021/03/beginners-guide-clip-model.html CLIP: A Beginner’s Guide to the CLIP Model]
* [https://github.com/openai/CLIP CLIP sur GitHub]
* [https://towardsdatascience.com/clip-the-most-influential-ai-model-from-openai-and-how-to-use-it-f8ee408958b1 CLIP: The Most Influential AI Model From OpenAI — And How To Use It]