[examples] Finalise the Ollama example application (#713)

examples/applications/docker-chatbot/chatbot.yaml

@@ -25,6 +25,7 @@ errors:
     on-failure: "skip"
 pipeline:
   - name: "convert-to-structure"
+    id: "chatbot"
     type: "document-to-json"
     input: "questions-topic"
     configuration:

examples/applications/docker-chatbot/crawler.yaml

@@ -58,7 +58,6 @@ pipeline:
   - name: "Detect language"
     type: "language-detector"
     configuration:
-      allowedLanguages: ["en", "fr"]
       property: "language"
   - name: "Split into chunks"
     type: "text-splitter"

examples/applications/ollama-chatbot/README.md

@@ -9,19 +9,36 @@ In this example we are using [HerdDB](ps://github.com/diennea/herddb) as a vecto
 but you can use any Vector databases.
 
 As LLM we are using [Ollama](https://ollama.ai), that is a service that runs on your machine. 
-
+We are using OpenAI to compute the embeddings of the texts.
 
 ## Install Ollama
 
-
 Follow the instructions on the Ollama.ai website to install Ollama.
 
-Then start Ollama with the llama2 model
+Then start Ollama with the llama2:13b model
 
 ```
-ollama run llama2
+ollama run llama2:13b
 ```
 
+If you want to use another model export this variable before starting the application.
+
+```bash
+export OLLAMA_MODEL=llama2:13b
+```
+
+## Configure you OpenAI API Key
+
+At the moment it the embeddings computed by Ollama models are not performing well, so we are using OpenAI to compute them. 
+
+Export to the ENV the access key to OpenAI
+
+```
+export OPEN_AI_ACCESS_KEY=...
+```
+
+The default [secrets file](../../secrets/secrets.yaml) reads from the ENV. Check out the file to learn more about
+the default settings, you can change them by exporting other ENV variables.
 
 ## Deploy the LangStream application in docker
 
@@ -37,3 +54,8 @@ Since the application opens a gateway, we can use the gateway API to send and co
 ```
 ./bin/langstream gateway chat test -cg bot-output -pg user-input -p sessionId=$(uuidgen)
 ```
+
+
+## Testing Ollama embeddings
+If you want to use Ollama for the embeddings you can change the "ai-service" to "ollama" in the pipeline files and set the 
+model for the embeddings.

examples/applications/ollama-chatbot/chatbot.yaml

@@ -32,7 +32,8 @@ pipeline:
   - name: "compute-embeddings"
     type: "compute-ai-embeddings"
     configuration:
-      model: "${secrets.ollama.model}"
+      ai-service: "openai"
+      model: "${secrets.open-ai.embeddings-model}"
       embeddings-field: "value.question_embeddings"
       text: "{{ value.question }}"
       flush-interval: 0
@@ -47,7 +48,7 @@ pipeline:
   - name: "re-rank documents with MMR"
     type: "re-rank"
     configuration:
-      max: 5 # keep only the top 5 documents, because we have an hard limit on the prompt size
+      max: 10 # keep only the top 5 documents, because we have and hard limit on the prompt size
       field: "value.related_documents"
       query-text: "value.question"
       query-embeddings: "value.question_embeddings"
@@ -58,10 +59,13 @@ pipeline:
       lambda: 0.5
       k1: 1.2
       b: 0.75
+  - name: "log documents"
+    type: "log-event"
   - name: "ai-chat-completions"
     type: "ai-chat-completions"
 
     configuration:
+      ai-service: "ollama"
       model: "${secrets.ollama.model}"
       # on the log-topic we add a field with the answer
       completion-field: "value.answer"
@@ -79,20 +83,21 @@ pipeline:
       # eventually we want to send bigger messages to reduce the overhead of each message on the topic
       min-chunks-per-message: 20
       messages:
-        - role: system
-          content: |
-              An user is going to perform a questions, The documents below may help you in answering to their questions.
-              Please try to leverage them in your answer as much as possible.
-              Take into consideration that the user is always asking questions about the LangStream project.
-              If you provide code or YAML snippets, please explicitly state that they are examples.
-              Do not provide information that is not related to the LangStream project.
+        - content: |            
+              I have this list of documents that may help you in answering to my question below.
             
               Documents:
               {{# value.related_documents}}
               {{ text}}
               {{/ value.related_documents}}
-        - role: user
-          content: "{{ value.question}}"
+            
+              Please try to leverage the content above in your answer as much as possible.
+              Take into consideration that I am always asking questions about the LangStream project.              
+              Do not provide information that is not related to the LangStream project.
+            
+              This is my question:
+              {{ value.question}}
+
   - name: "cleanup-response"
     type: "drop-fields"
     output: "log-topic"

examples/applications/ollama-chatbot/configuration.yaml

@@ -27,8 +27,16 @@ configuration:
         password: "${secrets.herddb.password}"
     - type: "ollama-configuration"
       name: "ollama"
+      id: "ollama"
       configuration:
         url: "${secrets.ollama.url}"
+    - type: "open-ai-configuration"
+      name: "OpenAI Azure configuration"
+      id: "openai"
+      configuration:
+        url: "${secrets.open-ai.url}"
+        access-key: "${secrets.open-ai.access-key}"
+        provider: "${secrets.open-ai.provider}"
   dependencies:
     - name: "HerdDB.org JDBC Driver"
       url: "https://repo1.maven.org/maven2/org/herddb/herddb-jdbc/0.28.0/herddb-jdbc-0.28.0-thin.jar"

examples/applications/ollama-chatbot/crawler.yaml

@@ -41,7 +41,6 @@ pipeline:
   - name: "Detect language"
     type: "language-detector"
     configuration:
-      allowedLanguages: ["en", "fr"]
       property: "language"
   - name: "Split into chunks"
     type: "text-splitter"
@@ -77,7 +76,8 @@ pipeline:
     id: "step1"
     type: "compute-ai-embeddings"
     configuration:
-      model: "${secrets.ollama.model}"
+      ai-service: "openai"
+      model: "${secrets.open-ai.embeddings-model}"
       embeddings-field: "value.embeddings_vector"
       text: "{{ value.text }}"
       batch-size: 10

examples/secrets/secrets.yaml

@@ -148,7 +148,7 @@ secrets:
     id: ollama
     data:
       url: "${OLLAMA_URL:-http://host.docker.internal:11434}"
-      model: "${OLLAMA_MODEL:-mistral}"
+      model: "${OLLAMA_MODEL:-llama2:13b}"
   - name: camel-github-source
     id: camel-github-source
     data: