add logging and model parameter tutorials to docu (#536)

Adding Model Parameter Setting tutorial and Logging explanation to
documentation

docs/docs/tutorials/image-models.md

@@ -4,6 +4,8 @@ sidebar_position: 21
 
 # 12. Image Models
 
-Creating and interpreting
+Creating images: example in https://github.com/langchain4j/langchain4j-examples/blob/main/tutorials/src/main/java/_02_OpenAiImageModelExamples.java
 
-Coming soon
+Interpreting images: coming very soon
+
+More elaborated content coming soon - or help us by adding it <3

docs/docs/tutorials/logging.md

@@ -0,0 +1,56 @@
+---
+sidebar_position: 30
+---
+
+# 15. Logging
+
+### Model requests and responses
+Console output can be switched on and off by setting `.logRequests()` and `.logResponses()` on the model
+
+```
+        ChatLanguageModel model = OpenAiChatModel.builder()
+                .apiKey(ApiKeys.OPENAI_API_KEY)
+                .logRequests(true)
+                .logResponses(true)
+                .build();
+```
+
+### Default logging: slf4j 
+LangChain4j comes with an slf4j facade, but the user is free to use any logging backend (e.g. log4j, logback, tinylog, etc.)
+
+An example of Tinylog backend can be found in langchain4j-examples/tutorials, where logging properties are set in `tinylog.properties`, as follows:
+```
+writer.level = info
+```
+
+Typical log level settings are `error`, `warn`, `info` and `debug`. 
+
+An overview of all the options:
+- `off`: No log messages will be written. This effectively disables logging.
+- `trace`: All log messages, including trace, debug, info, warn, and error, will be written to the log output.
+- `debug`: Log messages of debug, info, warn, and error levels will be written to the log output. Trace messages will be ignored.
+- `info`: Log messages of info, warn, and error levels will be written to the log output. Debug and trace messages will be ignored.
+- `warn`: Log messages of warn and error levels will be written to the log output. Info, debug, and trace messages will be ignored.
+- `error`: Only log messages of error level will be written to the log output. Warn, info, debug, and trace messages will be ignored.
+- `fatal`: This level is not part of the standard log levels in Tinylog. You can use it to specify a custom level for log messages. By default, it behaves the same as the `error` level.
+
+## Quarkus
+In Quarkus examples, logging properties are set in the `application.properties` file:
+```
+quarkus.log.console.enable = true
+quarkus.log.file.enable = false
+quarkus.langchain4j.openai.chat-model.log-responses = true
+quarkus.langchain4j.openai.chat-model.log-requests = true
+```
+
+These properties can also be set and changed in the Quarkus Dev UI, when running the application in dev mode (command: `quarkus dev`).
+The Dev UI can then be accessed via `host:port/q/dev-ui`.
+
+## Spring Boot
+In Spring Boot examples, logging properties are set in the `application.properties` file
+```
+logging.level.dev.langchain4j=INFO
+logging.level.dev.ai4j.openai4j=INFO
+```
+
+_This documentation page is a stub - help us make it better_

docs/docs/tutorials/response-streaming.md

@@ -4,6 +4,6 @@ sidebar_position: 9
 
 # 4. Response Streaming
 
-[Streaming of LLM responses](https://github.com/langchain4j/langchain4j-examples/blob/main/other-examples/src/main/java/StreamingExamples.java)
+[Streaming of LLM responses](https://github.com/langchain4j/langchain4j-examples/blob/main/tutorials/src/main/java/_04_Streaming.java)
 
 Tutorial coming soon

docs/docs/tutorials/set-model-parameters.md

@@ -3,5 +3,81 @@ sidebar_position: 4
 ---
 
 # 2. Set Model Parameters
+An example of specifying model parameters can be found [here](https://github.com/langchain4j/langchain4j-examples/blob/main/tutorials/src/main/java/_01_ModelParameters.java
+).
 
-Coming soon
+## What are Parameters in LLMs
+Depending on which model and which model provider you use, you can set a lot of parameters that will influence the model's output, speed, logging, etc.
+Typically, you will find all the parameters and their meaning on the provider's website.
+
+
+For example, OpenAI API's parameters can be found at https://platform.openai.com/docs/api-reference/chat (most up-to-date version)
+and include options like
+
+| Parameter          | Description                                                                                                                                                                                          | Type      |
+|--------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------|
+| `modelName`        | The name of the model to use (gtp-3.5-turbo, gpt-4-1106-preview, ...)                                                                                                                                | `String`  |
+| `temperature`      | What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.                 | `Double`  |
+| `max_tokens`       | The maximum number of tokens that can be generated in the chat completion.                                                                                                                           | `Integer` |
+| `frequencyPenalty` | Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim.           | `Double`  |
+| `...`              | ...                                                                                                                                                                                                  | `...`     |
+
+For the full list of parameters in OpenAI LLMs, see the [OpenAI Language Model page](/docs/integrations/language-models/openai).
+Full lists of parameters and default values per model can be found under the separate model pages (under Integration, Language Model and Image Model).
+
+## Default Parameter Settings
+The LangChain4j framework offers very easy model constructors with a lot of parameters set under the hood to sensible defaults. The fastest way to construct a model object is
+```
+ChatLanguageModel model = OpenAiChatModel.withApiKey("demo");
+```
+In this case of an OpenAI Chat Model for example, some of the defaults are
+
+| Parameter      | Default Value | 
+|----------------|---------------|
+| `timeout`      | 60s           |
+| `modelName`    | gpt-3.5-turbo |
+| `temperature`  | 0.7           |
+| `logRequests`  | false         |
+| `logResponses` | false         |
+| `...`          | ...           |
+
+Defaults for all language and image models can be found on the pages of the respective providers under [Integrations](/docs/integrations).
+
+## How to Set Parameter Values
+When we use the builder pattern, we will be able to set all the available parameters of the model as follows:
+```
+ChatLanguageModel model = OpenAiChatModel.builder()
+        .apiKey(ApiKeys.OPENAI_API_KEY)
+        .modelName(GPT_3_5_TURBO)
+        .temperature(0.3)
+        .timeout(ofSeconds(60))
+        .logRequests(true)
+        .logResponses(true)
+        .build();
+```
+
+## Parameter Settings in Quarkus
+LangChain4j parameters in Quarkus applications can be set in the `application.properties` file as follows:
+```
+quarkus.langchain4j.openai.chat-model.temperature=0.5
+quarkus.langchain4j.openai.timeout=60s
+quarkus.langchain4j.openai.api-key=${OPENAI_API_KEY}
+```
+
+Interestingly, for debugging, tweaking or even just knowing all the available parameters, one can have a look in the quarkus DEV UI.
+In this dashboard, you can make changes that will be immediately reflected in your running instance, and your changes are automatically ported to the code.
+The DEV UI can be accessed by running your Quarkus application with the command `quarkus dev`, then you can find it on localhost:8080/q/dev-ui (or wherever you deploy your application).
+
+
+[![](/img/quarkus-dev-ui-parameters.png)](/docs/tutorials/set-model-parameters)
+
+## Parameter Settings in Spring Boot
+LangChain4j parameters in Spring Boot applications can be set in the `application.properties` file as follows:
+```
+langchain4j.open-ai.chat-model.api-key=${OPENAI_API_KEY}
+langchain4j.open-ai.chat-model.model-name=gpt-4-1106-preview
+langchain4j.open-ai.chat-model.temperature=0.0
+langchain4j.open-ai.chat-model.timeout=PT60S
+langchain4j.open-ai.chat-model.log-requests=false
+langchain4j.open-ai.chat-model.log-responses=false
+```