More on Ollama

examples/applications/ollama-chatbot/.gitignore

@@ -0,0 +1 @@
+java/lib/*

examples/applications/ollama-chatbot/README.md

@@ -0,0 +1,96 @@
+# Running your own Chat bot using docker
+
+This sample application shows how to build a chat bot over the content of a website.
+In this case you are going to crawl the LangStream.ai documentation website.
+
+The Chat bot will be able to help you with LangStream.
+
+In this example we are using [HerdDB](ps://github.com/diennea/herddb) as a vector database using the JDBC driver,
+but you can use any Vector databases.
+
+
+## Configure you OpenAI API Key
+
+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
+
+The default docker runner starts Minio, Kafka and HerdDB, so you can run the application locally.
+
+```
+./bin/langstream docker run test -app examples/applications/docker-chatbot -s examples/secrets/secrets.yaml
+```
+
+
+## Talk with the Chat bot using the CLI
+Since the application opens a gateway, we can use the gateway API to send and consume messages.
+
+```
+./bin/langstream gateway chat test -cg bot-output -pg user-input -p sessionId=$(uuidgen)
+```
+
+Responses are streamed to the output-topic. If you want to inspect the history of the raw answers you can
+consume from the log-topic using the llm-debug gateway:
+
+```
+./bin/langstream gateway consume test llm-debug
+```
+
+## Application flow chart
+
+```mermaid
+flowchart TB
+
+  subgraph JdbcDatasource["<b>⛁ JdbcDatasource</b>"]
+    documents
+  end
+
+  subgraph streaming-cluster["<b>✉️ streaming cluster</b>"]
+    questions-topic
+    answers-topic
+    log-topic
+    chunks-topic
+  end
+
+  subgraph gateways["<b>gateways</b>"]
+    user-input --> questions-topic
+    answers-topic --> bot-output
+    log-topic --> llm-debug
+  end
+
+  subgraph chatbot["<b>chatbot</b>"]
+    A("convert-to-structure<br><i>document-to-json</i>") --> B
+    B("compute-embeddings<br><i>compute-ai-embeddings</i>") --> C
+    C("lookup-related-documents<br><i>query-vector-db</i>") --> D
+    D("re-rank documents with MMR<br><i>re-rank</i>") --> E
+    E("ai-chat-completions<br><i>ai-chat-completions</i>") --> F
+    F("cleanup-response<br><i>drop-fields</i>")
+  end
+  questions-topic --> A
+  JdbcDatasource --> C
+  E --> answers-topic
+  F --> log-topic
+
+  subgraph crawler["<b>crawler</b>"]
+    G("Crawl the WebSite<br><i>webcrawler-source</i>") --> H
+    H("Extract text<br><i>text-extractor</i>") --> I
+    I("Normalise text<br><i>text-normaliser</i>") --> J
+    J("Detect language<br><i>language-detector</i>") --> K
+    K("Split into chunks<br><i>text-splitter</i>") --> L
+    L("Convert to structured data<br><i>document-to-json</i>") --> M
+    M("prepare-structure<br><i>compute</i>") --> N
+    N("compute-embeddings<br><i>compute-ai-embeddings</i>")
+    O("Write<br><i>vector-db-sink</i>")
+  end
+  P["🌐 web site"] --> G
+  N --> chunks-topic
+  chunks-topic --> O
+  O --> documents
+```

examples/applications/ollama-chatbot/assets.yaml

@@ -0,0 +1,32 @@
+#
+# Copyright DataStax, Inc.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+assets:
+  - name: "documents-table"
+    asset-type: "jdbc-table"
+    creation-mode: create-if-not-exists
+    config:
+      table-name: "documents"
+      datasource: "JdbcDatasource"
+      create-statements:
+        - |
+          CREATE TABLE documents (
+          filename TEXT,
+          chunk_id int,
+          num_tokens int,
+          lang TEXT,
+          text TEXT,
+          embeddings_vector FLOATA,
+          PRIMARY KEY (filename, chunk_id));

examples/applications/ollama-chatbot/chatbot.yaml

@@ -0,0 +1,109 @@
+#
+# Copyright DataStax, Inc.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+
+topics:
+  - name: "questions-topic"
+    creation-mode: create-if-not-exists
+  - name: "answers-topic"
+    creation-mode: create-if-not-exists
+  - name: "log-topic"
+    creation-mode: create-if-not-exists
+errors:
+    on-failure: "skip"
+pipeline:
+  - name: "convert-to-structure"
+    type: "document-to-json"
+    input: "questions-topic"
+    configuration:
+      text-field: "question"
+  - name: "compute-embeddings"
+    type: "compute-ai-embeddings"
+    resources:
+      disk:
+        size: 256M
+        enabled: true
+    configuration:
+      model: "multilingual-e5-small"
+      model-url: "djl://ai.djl.huggingface.pytorch/intfloat/multilingual-e5-small"
+      ai-service: "huggingface"
+      embeddings-field: "value.question_embeddings"
+      text: "{{ value.question }}"
+      flush-interval: 0
+  - name: "lookup-related-documents"
+    type: "query-vector-db"
+    configuration:
+      datasource: "JdbcDatasource"
+      query: "SELECT text,embeddings_vector FROM documents ORDER BY cosine_similarity(embeddings_vector, CAST(? as FLOAT ARRAY)) DESC LIMIT 20"
+      fields:
+        - "value.question_embeddings"
+      output-field: "value.related_documents"
+  - 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
+      field: "value.related_documents"
+      query-text: "value.question"
+      query-embeddings: "value.question_embeddings"
+      output-field: "value.related_documents"
+      text-field: "record.text"
+      embeddings-field: "record.embeddings_vector"
+      algorithm: "MMR"
+      lambda: 0.5
+      k1: 1.2
+      b: 0.75
+  - 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"
+      # we are also logging the prompt we sent to the LLM
+      log-field: "value.prompt"
+      # here we configure the streaming behavior
+      # as soon as the LLM answers with a chunk we send it to the answers-topic
+      stream-to-topic: "answers-topic"
+      # on the streaming answer we send the answer as whole message
+      # the 'value' syntax is used to refer to the whole value of the message
+      stream-response-completion-field: "value"
+      # we want to stream the answer as soon as we have 20 chunks
+      # in order to reduce latency for the first message the agent sends the first message
+      # with 1 chunk, then with 2 chunks....up to the min-chunks-per-message value
+      # 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.
+            
+              Documents:
+              {{# value.related_documents}}
+              {{ text}}
+              {{/ value.related_documents}}
+        - role: user
+          content: "{{ value.question}}"
+  - name: "cleanup-response"
+    type: "drop-fields"
+    output: "log-topic"
+    configuration:
+      fields:
+        - "question_embeddings"
+        - "related_documents"

examples/applications/ollama-chatbot/configuration.yaml

@@ -0,0 +1,40 @@
+#
+#
+# Copyright DataStax, Inc.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+
+configuration:
+  resources:
+    - type: "datasource"
+      name: "JdbcDatasource"
+      configuration:
+        service: "jdbc"
+        driverClass: "herddb.jdbc.Driver"
+        url: "${secrets.herddb.url}"
+        user: "${secrets.herddb.user}"
+        password: "${secrets.herddb.password}"
+    - type: "ollama-configuration"
+      name: "ollama"
+      configuration:
+        url: "${secrets.ollama.url}"
+    - type: "hugging-face-configuration"
+      name: "huggingface"
+      configuration:
+        provider: "local"
+  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"
+      sha512sum: "d8ea8fbb12eada8f860ed660cbc63d66659ab3506bc165c85c420889aa8a1dac53dab7906ef61c4415a038c5a034f0d75900543dd0013bdae50feafd46f51c8e"
+      type: "java-library"