Merge pull request #4 from DevlinRocha/ollama

Ollama

Author: DevlinRocha

package.json

@@ -1,6 +1,6 @@
 {
   "name": "create-lilypad-module",
-  "version": "0.0.30",
+  "version": "0.0.39",
   "description": "Create Lilypad modules with a modern Docker setup and minimal configuration.",
   "bin": {
     "create-lilypad-module": "src/create_lilypad_module/scaffold"

src/create_lilypad_module/templates/.env

@@ -3,4 +3,4 @@ MODEL_VERSION=
 DOCKER_HUB_USERNAME=
 DOCKER_IMAGE=
 GITHUB_REPO=
-VERSION=
+VERSION=v0.0.0

src/create_lilypad_module/templates/Dockerfile

@@ -29,7 +29,7 @@ RUN mkdir -p ./outputs && chmod 777 ./outputs
 # Set outputs directory as a volume
 VOLUME ./outputs
 
-# Copy a script to start ollama and handle input
+# Copy source code and handle request
 COPY src ./src
 RUN chmod +x ./src/run_model
 

src/create_lilypad_module/templates/README.md

@@ -15,24 +15,59 @@ To build and run a module on Lilypad Network, you'll need to have the [Lilypad C
 
 Your module's ready! 🎉
 
-Once your Docker image has been pushed to Docker Hub, you can run your module on Lilypad Network:
+Once your Docker image has been pushed to Docker Hub, you can run your module on Lilypad Network.
 
-> Run `git log` in your terminal to easily find the latest commit hash to use as the GitHub tag.
+> Make sure that you Base64 encode your request.
 
 ```sh
 export WEB3_PRIVATE_KEY=WEB3_PRIVATE_KEY
 
-lilypad run github.com/github_username/module_repo:github_tag -i prompt="What animal order do frogs belong to"
+lilypad run github.com/GITHUB_USERNAME/MODULE_REPO:TAG \
+-i request="$(echo -n '{
+  "model": "MODEL_NAME:MODEL_VERSION",
+  "messages": [{
+    "role": "system",
+    "content": "you are a helpful AI assistant"
+  },
+  {
+  "role": "user",
+  "content": "what is the animal order of the frog?"
+  }],
+  "stream": false,
+  "options": {
+    "temperature": 1.0
+  }
+}' | base64 -w 0)"
 ```
 
+### Valid Options Parameters and Default Values
+
+- [Ollama Modelfile](https://github.com/ollama/ollama/blob/main/docs/modelfile.md#valid-parameters-and-values)
+
+| Parameter      | Description                                                                                                                                                                                                                                                                                                                                                 | Default |
+| -------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------- |
+| mirostat       | Enable Mirostat sampling for controlling perplexity. (0 = disabled, 1 = Mirostat, 2 = Mirostat 2.0)                                                                                                                                                                                                                                                         | `0`     |
+| mirostat_eta   | Influences how quickly the algorithm responds to feedback from the generated text. A lower learning rate will result in slower adjustments, while a higher learning rate will make the algorithm more responsive.                                                                                                                                           | `0.1`   |
+| mirostat_tau   | Controls the balance between coherence and diversity of the output. A lower value will result in more focused and coherent text.                                                                                                                                                                                                                            | `5`     |
+| num_ctx        | Sets the size of the context window used to generate the next token.                                                                                                                                                                                                                                                                                        | `2048`  |
+| repeat_last_n  | Sets how far back for the model to look back to prevent repetition. (0 = disabled, -1 = num_ctx)                                                                                                                                                                                                                                                            | `64`    |
+| repeat_penalty | Sets how strongly to penalize repetitions. A higher value (e.g., 1.5) will penalize repetitions more strongly, while a lower value (e.g., 0.9) will be more lenient.                                                                                                                                                                                        | `1.1`   |
+| temperature    | The temperature of the model. Increasing the temperature will make the model answer more creatively.                                                                                                                                                                                                                                                        | `0.8`   |
+| seed           | Sets the random number seed to use for generation. Setting this to a specific number will make the model generate the same text for the same prompt.                                                                                                                                                                                                        | `0`     |
+| stop           | Sets the stop sequences to use. When this pattern is encountered the LLM will stop generating text and return. Multiple stop patterns may be set by specifying multiple separate stop parameters in a modelfile.                                                                                                                                            |         |
+| num_predict    | Maximum number of tokens to predict when generating text. (-1 = infinite generation)                                                                                                                                                                                                                                                                        | `-1`    |
+| top_k          | Reduces the probability of generating nonsense. A higher value (e.g. 100) will give more diverse answers, while a lower value (e.g. 10) will be more conservative.                                                                                                                                                                                          | `40`    |
+| top_p          | Works together with top-k. A higher value (e.g., 0.95) will lead to more diverse text, while a lower value (e.g., 0.5) will generate more focused and conservative text.                                                                                                                                                                                    | `0.9`   |
+| min_p          | Alternative to the top_p, and aims to ensure a balance of quality and variety. The parameter p represents the minimum probability for a token to be considered, relative to the probability of the most likely token. For example, with p=0.05 and the most likely token having a probability of 0.9, logits with a value less than 0.045 are filtered out. | `0.0`   |
+
 ## Available Scripts
 
 In the project directory, you can run:
 
 ### [`scripts/configure`](scripts/configure)
 
-Configure your module.
-Set the following values in the [`.env` file](.env)
+Configures your module.
+Sets the following values in the [`.env` file](.env)
 
 ```
 MODEL_NAME
@@ -48,15 +83,15 @@ Builds the Docker image and pushes it to Docker Hub.
 
 ### `--major`, `--minor`, and `--patch` Flags
 
-Increment the specified version before building the Docker image.
+Increments the specified version before building the Docker image.
 
 #### `--local` Flag
 
 Loads the built Docker image into the local Docker daemon.
 
 ### [`scripts/run`](scripts/run)
 
-Run your module.
+Runs your module.
 
 ## Learn More
 

src/create_lilypad_module/templates/help

@@ -9,6 +9,6 @@ else
     echo "Available commands:"
     echo -e "\tscripts/configure                                        Configure the module"
     echo -e "\tscripts/build [--local] [--major] [--minor] [--patch]    Build and push a new Docker image"
-    echo -e "\tscripts/run [--local] <input>                            Run the module"
+    echo -e "\tscripts/run [--local] <request>                          Run the module"
     exit 1
 fi

src/create_lilypad_module/templates/lilypad_module.json.tmpl

@@ -5,18 +5,10 @@
     "Spec": {
       "Deal": { "Concurrency": 1 },
       "Docker": {
-        "WorkingDirectory": "/app",
         "Entrypoint": [
-          "/app/src/run_model",
-          "--prompt", {{ .prompt }}
-          {{- if .temperature -}},
-          "--temperature", {{ .temperature }}
-          {{- end -}}
-          {{- if .max_tokens -}},
-          "--max_tokens", {{ .max_tokens }}
-          {{- end -}}
+          "/app/src/run_model", {{ .request }}
         ],
-        "Image": "dockerhub_username/image:tag"
+        "Image": "dockerhub_username/image@"
       },
       "Engine": "Docker",
       "Network": { "Type": "None" },

src/create_lilypad_module/templates/scripts/build

@@ -13,7 +13,7 @@ if [ -z $MODEL_NAME ] || [ -z $MODEL_VERSION ] || [ -z $DOCKER_IMAGE ]; then
 fi
 
 for arg in $@; do
-  if [ $arg == "--local" ]; then
+  if [ $1 == "--local" ] || [ $1 == "-l" ]; then
     echo "Building the Docker image and loading it into the local Docker daemon..."
     local=true
   fi
@@ -37,7 +37,6 @@ for arg in $@; do
 
     VERSION="v$MAJOR.$MINOR.$PATCH"
     echo "New version: $VERSION"
-    sed -i "" "s/^VERSION=.*/VERSION=$VERSION/" .env
   fi
 
   if [ $arg == "--no-cache" ]; then
@@ -91,6 +90,7 @@ if [ -z $local ]; then
   echo "✅ Docker image built and published to Docker Hub successfully."
   echo -e "\thttps://hub.docker.com/repository/docker/$DOCKER_HUB_REPO/general"
   echo -e "\n\tscripts/run_module"
+  sed -i "" "s/^VERSION=.*/VERSION=$VERSION/" .env
 else
   echo "✅ Docker image built and loaded into local daemon successfully."
   echo -e "\n\tscripts/run_module --local"

src/create_lilypad_module/templates/scripts/run

@@ -1,36 +1,75 @@
 #!/usr/bin/env bash
 
-if [ "$#" -lt 1 ] || [ "$#" -gt 2 ]; then
-    echo "Usage: scripts/run [--local] <input>"
+CONFIG_FILE=".env"
+source $CONFIG_FILE
+
+function request {
+    if ! printenv | grep -q "WEB3_PRIVATE_KEY=."; then
+        printf "Enter your wallet private key: "
+        read -r private_key
+        WEB3_PRIVATE_KEY=$private_key
+        echo "Private key set"
+        echo "(Hint: Use 'export WEB3_PRIVATE_KEY=<PRIVATE_KEY>' to avoid this prompt in the future)"
+    fi
+    echo "Copy the JSON below to form your request:"
+    echo '
+  "messages": [{
+    "role": "system",
+    "content": "you are a helpful AI assistant"
+  },
+  {
+    "role": "user",
+    "content": "what is the animal order of the frog?"
+  }],
+  "options": {
+    "temperature": 1.0
+  }
+'
+    printf "(Paste JSON as one line) ➔ "
+    read -r request
+    request="{\"model\": \"$MODEL_NAME:$MODEL_VERSION\", $request, \"stream\": false}"
+}
+
+if [ $# -gt 2 ]; then
+    echo "Usage: scripts/run [--local] <request>"
     echo "Example: scripts/run 'What animal order do frogs belong to?'"
     exit 1
-fi
+elif [ $# -eq 0 ]; then
+    request
+else
+    while [[ $# -gt 0 ]]; do
+        case $1 in
+        --local | -l)
+            echo "Running the Lilypad module Docker image locally..."
+            local=true
+            shift
+            ;;
+        *)
+            request=$1
+            shift
+            ;;
+        esac
+    done
 
-if [ $1 == "--local" ] || [ $1 == "-l" ]; then
-    if [ "$#" -ne 2 ]; then
-        echo "❌ Error: Input is required."
-        echo "Example: scripts/run --local 'What animal order do frogs belong to?'"
-        exit 1
+    if [[ -z $request ]]; then
+        request
     fi
-    echo "Running the Lilypad module Docker image locally..."
-    local=true
-    INPUT=$2
-else
-    INPUT=$1
 fi
 
-commit_hash=$(git log --pretty=format:%H | head -n 1)
+# Base64 encode the request
+base64_request=$(echo $request | base64 -w 0)
 
-if [ $local != true ]; then
+if [ -z $local ]; then
+    commit_hash=$(git log --pretty=format:%H | head -n 1)
     MODULE=$GITHUB_REPO:$commit_hash
     echo "Running $MODULE on Lilypad Network..."
-    echo "Original input: $JSON_INPUT"
-    echo "Base64 encoded: $BASE64_INPUT"
-    lilypad run $MODULE -i prompt=$INPUT
+    echo "Original request: $request"
+    echo "Base64 encoded: $base64_request"
+    lilypad run $MODULE -i request=$base64_request --web3-private-key=$WEB3_PRIVATE_KEY
 else
     MODULE=$DOCKER_IMAGE:$VERSION
     echo "Running $MODULE locally..."
-    echo "Original input: $JSON_INPUT"
-    echo "Base64 encoded: $BASE64_INPUT"
-    docker run $MODULE $INPUT
+    echo "Original request: $request"
+    echo "Base64 encoded: $base64_request"
+    docker run $MODULE $base64_request
 fi

src/create_lilypad_module/templates/src/run_model

@@ -1,39 +1,11 @@
 #!/usr/bin/env bash
 
-# Initialize default values
-PROMPT=""
-TEMPERATURE="0.7"
-MAX_TOKENS="2048"
-
 # Create output directory if it doesn't exist
 mkdir -p /outputs
 
-echo "Input: $1" >&2
-
-# Parse command-line arguments
-while [[ $# -gt 0 ]]; do
-    case "$1" in
-    --prompt)
-        PROMPT="$2"
-        shift 2
-        ;;
-    --temperature)
-        TEMPERATURE="$2"
-        shift 2
-        ;;
-    --max_tokens)
-        MAX_TOKENS="$2"
-        shift 2
-        ;;
-    *)
-        echo "Unknown flag: $1" >&2
-        exit 1
-        ;;
-    esac
-done
-
-# Extract values from input JSON with defaults
-messages="[{\"role\": \"user\", \"content\": \"$PROMPT\"}]"
+# Parse Base64 request argument and decode to JSON
+echo "Raw request (Base64): $1" >&2
+request=$(echo "$1" | base64 -d)
 
 # Start the ollama server in the background
 echo "Starting Ollama server..." >&2
@@ -42,7 +14,7 @@ nohup bash -c "ollama serve &" >&2
 # Wait for server with timeout
 timeout=30
 start_time=$(date +%s)
-while ! curl -s http://127.0.0.1:11434 >/dev/null; do
+while ! curl -s http://127.0.0.1:11434 >/dev/null 2>&1; do
     current_time=$(date +%s)
     elapsed=$((current_time - start_time))
     if [ $elapsed -gt $timeout ]; then
@@ -55,49 +27,35 @@ done
 
 echo "Ollama server started" >&2
 
-# Prepare the chat completion request
-request=$(
-    cat <<EOF
-{
-  "model": "$MODEL_ID",
-  "messages": $messages,
-  "temperature": $TEMPERATURE,
-  "max_tokens": $MAX_TOKENS,
-  "stream": false
-}
-EOF
-)
-
 # Make the API call to Ollama's chat endpoint
 echo "Making request to Ollama..." >&2
 response=$(curl -s http://127.0.0.1:11434/api/chat \
     -H "Content-Type: application/json" \
     -d "$request")
 
-# Create JSON structure following OpenAI format
-escaped_response=$(echo "$response" | sed 's/"/\\"/g')
 formatted_response="{
-    \"id\": \"cmpl-$(openssl rand -hex 12)\",
-    \"object\": \"text_completion\",
-    \"created\": "$(date +%s)",
-    \"model\": \"$MODEL_ID\",
-    \"choices\": [{
-        \"text\": \"$escaped_response\",
-        \"index\": 0,
-        \"logprobs\": null,
-        \"finish_reason\": \"stop\"
+    'id': 'cmpl-$(openssl rand -hex 12)',
+    'object': 'text_completion',
+    'created': "$(date +%s)",
+    'model': '$MODEL_ID',
+    'choices': [{
+        'text': '$escaped_response',
+        'index': 0,
+        'logprobs': null,
+        'finish_reason': 'stop'
     }],
-    \"usage\": {
-        \"prompt_tokens\": null,
-        \"completion_tokens\": null,
-        \"total_tokens\": null
+    'usage': {
+        'prompt_tokens': null,
+        'completion_tokens': null,
+        'total_tokens': null
     }
 }"
 
 # Save debug info
 {
     echo "=== Debug Info ==="
-    echo "Input: $1"
+    date
+    echo "Request (Base64): $1"
     echo "Request to Ollama: $request"
     echo "Response from Ollama:"
     echo "$response"