Changed configuration file name

.gitignore

@@ -147,7 +147,7 @@ tmp/
 temp/
 # configurations other than the default one
 configs/*
-!configs/default.yaml
+!configs/config.yaml
 
 # personalities other than the default one
 personalities/*

api/__init__.py

@@ -334,7 +334,7 @@ class ModelProcess:
         else:
             print("No model is installed or selected. Please make sure to install a model and select it inside your configuration before attempting to communicate with the model.")
             print("To do this: Install the model to your models/<binding name> folder.")
-            print("Then set your model information in your local configuration file that you can find in configs/local_default.yaml")
+            print("Then set your model information in your local configuration file that you can find in configs/local_config.yaml")
             print("You can also use the ui to set your model in the settings page.")
             output = ""
         return output

app.py

@@ -933,13 +933,13 @@ if __name__ == "__main__":
 
     # The default configuration must be kept unchanged as it is committed to the repository, 
     # so we have to make a copy that is not comitted
-    default_config = load_config(f"configs/default.yaml")
+    default_config = load_config(f"configs/config.yaml")
 
-    if args.config=="default":
-        args.config = "local_default"
-        if not Path(f"configs/local_default.yaml").exists():
+    if args.config!="local_config":
+        args.config = "local_config"
+        if not Path(f"configs/local_config.yaml").exists():
             print("No local configuration file found. Building from scratch")
-            shutil.copy(f"configs/default.yaml", f"configs/local_default.yaml")
+            shutil.copy(f"configs/config.yaml", f"configs/local_config.yaml")
 
     config_file_path = f"configs/{args.config}.yaml"
     config = load_config(config_file_path)

docs/dev/server_endpoints.md

@@ -167,7 +167,7 @@ This Flask server provides various endpoints to manage and interact with the cha
 {
   "auto_read": false,
   "binding": "llama_cpp",
-  "config": "local_default",
+  "config": "local_config.yaml",
   "ctx_size": 2048,
   "db_path": "databases/database.db",
   "debug": false,

docs/youtube/script_install.md

@@ -32,7 +32,7 @@ You may notice that the installer does not prompt you to install the default mod
 
 Now let's take a look at the new ui. As the previous one, you have a light mode and a dark mode. You also have direct access to the github repository.
 
-On the right side, you have the discussions panel, and on the center side, you have the messages flow.
+On the left side, you have the discussions panel, and on the center side, you have the messages flow.
 
 Before starting to use the tool, we need to download a model for a specific binding.
 To do this, go to settings. Then open the models zoo tab.
@@ -40,9 +40,34 @@ You need to select a binding from the list. For example the llama-cpp-official.
 
 Once the installation is done, you should install a model by pressing install and waiting for it to finish.
 This may take some time.
-Once the model is installed, you can select it and press Apply modifications.
+
+Once the model is installed, you can select it and press Apply changes.
 
 Notice that applying modifications does not save the configuration, so You need to press the save button and confirm.
 
 Now your model is selected and you are ready to start your first discussion.
 
+Let's go back to discussions view. To create a new discussion, press the + button. You should see the personality welcome message. This is a predefined welcome message that you can find in the personality configuration file. by default, we use the GPT4All personality which is conditioned to be a helpful personality. Let's ask it something. For example, who is Abraham Lincoln?
+
+You can stop the generation at any time by pressing the Stop Generating button.
+
+Now that the generation is done, we can see that it did a good job. In fact there is a little generation starting time at the beginning, but one it starts, you get around two words a second depending on your pc or mac configuration. As this is a discussion, we can follow up by asking to give more details.
+
+You can see that the model did give a more detailed text about Abraham Lincoln. You can fact check the information on wikipedia, as those models may sometimes hallucinate or invent things that are not true. It depends on the quality of the data used to train the model and to fine tune it. Some models are better than others. You can test by yourself to choose the best model for your needs. This is one of the advantages of being able to test multiple bindings and multiple models.
+
+
+Just as previous version, you can give the answer a thumb up or a thumb down, you can also edit the message, copy the message to the clip board, and remove the message. Notice that we removed the option to have the message read by the AI, but this feature should be added in the release version. I think of giving different voices to each personality and use a fullly local library as the previous version used a remote text to speach synthesizer. Our phylosophy is, your data should stay on your PC and never be sent out unless you explicitly do it yourself.
+
+
+Now let's focus on the discussions sidebar. As we saw, we can create new discussion. When you create a new discussion, its name is always new discussion, but as soon as you put your initial prompt, the discussion title take the first message you typed as the title. You can edit the title at any time, or delete the discussion. All your discussions are stored in a local sqlite3 database on your databases/database.db file. You can change the database path in your configs/local_config.yaml file. We will make this possible from the UI in the release version.
+
+In chat-GPT ui, one complaint I have is the difficulty to find out your discussion after days of discussions. So we have thought this up by adding a search button that allows you to search your discussions at any time. You can also delete multiple discussions or export them. Exporting messages may be interesting for training, but also for contributing to datalakes to help the community get more quality data. you can do any multi-discussions operation by pressing the check mode button. now you can select one or many discussions then you can export them or delete them.
+
+In this tutorial we have seen how to install the new ui, how to select a binding, download and select a model, how to start a discussion and follow up with more details, how to manipulate discussions.
+In the next tutorial we will take a look at the personalities zoo and all the 250 personalities that you can interact with.
+
+Thank you very much for your attention. I hope you like this tool, test it, share your experience with it and report bugs. This is an amateur project, so don't be harsh on us as we do this in our free time.
+
+Don't forget to hit like and subscribe and share the video if you want it to reach more people.
+
+See ya