Add parallel requests

R/ollama.R

@@ -64,7 +64,7 @@ create_request <- function(endpoint, host = NULL) {
 #' @param stream Enable response streaming. Default is FALSE.
 #' @param raw If TRUE, no formatting will be applied to the prompt. You may choose to use the raw parameter if you are specifying a full templated prompt in your request to the API. Default is FALSE.
 #' @param keep_alive The time to keep the connection alive. Default is "5m" (5 minutes).
-#' @param output A character vector of the output format. Default is "resp". Options are "resp", "jsonlist", "raw", "df", "text".
+#' @param output A character vector of the output format. Default is "resp". Options are "resp", "jsonlist", "raw", "df", "text", "req" (httr2_request object).
 #' @param endpoint The endpoint to generate the completion. Default is "/api/generate".
 #' @param host The base URL to use. Default is NULL, which uses Ollama's default base URL.
 #' @param ... Additional options to pass to the model.
@@ -80,10 +80,10 @@ create_request <- function(endpoint, host = NULL) {
 #' generate("llama3", "The sky is...", stream = TRUE, output = "text")
 #' generate("llama3", "The sky is...", stream = TRUE, output = "text", temperature = 2.0)
 #' generate("llama3", "The sky is...", stream = FALSE, output = "jsonlist")
-generate <- function(model, prompt, suffix = "", images = "", system = "", template = "", context = list(), stream = FALSE, raw = FALSE, keep_alive = "5m", output = c("resp", "jsonlist", "raw", "df", "text"), endpoint = "/api/generate", host = NULL, ...) {
+generate <- function(model, prompt, suffix = "", images = "", system = "", template = "", context = list(), stream = FALSE, raw = FALSE, keep_alive = "5m", output = c("resp", "jsonlist", "raw", "df", "text", "req"), endpoint = "/api/generate", host = NULL, ...) {
     output <- output[1]
-    if (!output %in% c("df", "resp", "jsonlist", "raw", "text")) {
-        stop("Invalid output format specified. Supported formats: 'df', 'resp', 'jsonlist', 'raw', 'text'")
+    if (!output %in% c("df", "resp", "jsonlist", "raw", "text", "req")) {
+        stop("Invalid output format specified. Supported formats: 'df', 'resp', 'jsonlist', 'raw', 'text', 'req'")
     }
 
     req <- create_request(endpoint, host)
@@ -118,6 +118,10 @@ generate <- function(model, prompt, suffix = "", images = "", system = "", templ
 
     req <- httr2::req_body_json(req, body_json, stream = stream)
 
+    if (output == "req") {
+        return(req)
+    }
+
     if (!stream) {
         tryCatch(
             {
@@ -160,7 +164,7 @@ generate <- function(model, prompt, suffix = "", images = "", system = "", templ
 #' @param tools Tools for the model to use if supported. Requires stream = FALSE. Default is an empty list.
 #' @param stream Enable response streaming. Default is FALSE.
 #' @param keep_alive The duration to keep the connection alive. Default is "5m".
-#' @param output The output format. Default is "resp". Other options are "jsonlist", "raw", "df", "text".
+#' @param output The output format. Default is "resp". Other options are "jsonlist", "raw", "df", "text", "req" (httr2_request object).
 #' @param endpoint The endpoint to chat with the model. Default is "/api/chat".
 #' @param host The base URL to use. Default is NULL, which uses Ollama's default base URL.
 #' @param ... Additional options to pass to the model.
@@ -191,9 +195,9 @@ generate <- function(model, prompt, suffix = "", images = "", system = "", templ
 #'     list(role = "user", content = "List all the previous messages.")
 #' )
 #' chat("llama3", messages, stream = TRUE)
-chat <- function(model, messages, tools = list(), stream = FALSE, keep_alive = "5m", output = c("resp", "jsonlist", "raw", "df", "text"), endpoint = "/api/chat", host = NULL, ...) {
+chat <- function(model, messages, tools = list(), stream = FALSE, keep_alive = "5m", output = c("resp", "jsonlist", "raw", "df", "text", "req"), endpoint = "/api/chat", host = NULL, ...) {
     output <- output[1]
-    if (!output %in% c("df", "resp", "jsonlist", "raw", "text")) {
+    if (!output %in% c("df", "resp", "jsonlist", "raw", "text", "req")) {
         stop("Invalid output format specified. Supported formats: 'df', 'resp', 'jsonlist', 'raw', 'text'")
     }
 
@@ -218,6 +222,9 @@ chat <- function(model, messages, tools = list(), stream = FALSE, keep_alive = "
     }
 
     req <- httr2::req_body_json(req, body_json, stream = stream)
+    if (output == "req") {
+        return(req)
+    }
 
     if (!stream) {
         tryCatch(

README.Rmd

@@ -284,3 +284,87 @@ messages[[4]]  # get 2nd message
 # delete a message at a specific index/position (2nd position in the example below)
 messages <- delete_message(messages, 2)
 ```
+
+## Advanced usage
+
+### Parallel requests
+
+For the `generate()` and `chat()` endpoints/functions, you can make parallel requests with the `req_perform_parallel` function from the `httr2` library. You need to specify `output = 'req'` in the function so the functions return `httr2_request` objects instead of `httr2_response` objects.
+
+```{r eval=FALSE}
+library(httr2)
+
+prompt <- "Tell me a 5-word story"
+
+# create 5 httr2_request objects that generate a response to the same prompt
+reqs <- lapply(1:5, function(r) generate("llama3.1", prompt, output = "req"))
+
+# make parallel requests and get response
+resps <- req_perform_parallel(reqs)  # list of httr2_request objects
+
+# process the responses
+sapply(resps, resp_process, "text")  # get responses as text
+# [1] "She found him in Paris."         "She found the key upstairs."    
+# [3] "She found her long-lost sister." "She found love on Mars."        
+# [5] "She found the diamond ring."    
+
+```
+
+Example sentiment analysis with parallel requests with `generate()` function
+
+```{r eval=FALSE}
+library(httr2)
+library(glue)
+library(dplyr)
+
+# text to classify
+texts <- c('I love this product', 'I hate this product', 'I am neutral about this product')
+
+# create httr2_request objects for each text, using the same system prompt
+reqs <- lapply(texts, function(text) {
+  prompt <- glue("Your only task/role is to evaluate the sentiment of product reviews, and your response should be one of the following:'positive', 'negative', or 'other'. Product review: {text}")
+  generate("llama3.1", prompt, output = "req")
+})
+
+# make parallel requests and get response
+resps <- req_perform_parallel(reqs)  # list of httr2_request objects
+
+# process the responses
+sapply(resps, resp_process, "text")  # get responses as text
+# [1] "Positive"                            "Negative."                          
+# [3] "'neutral' translates to... 'other'."
+
+
+```
+
+Example sentiment analysis with parallel requests with `chat()` function
+
+```{r eval=FALSE}
+library(httr2)
+library(dplyr)
+
+# text to classify
+texts <- c('I love this product', 'I hate this product', 'I am neutral about this product')
+
+# create system prompt
+chat_history <- create_message("Your only task/role is to evaluate the sentiment of product reviews provided by the user. Your response should simply be 'positive', 'negative', or 'other'.", "system")
+
+# create httr2_request objects for each text, using the same system prompt
+reqs <- lapply(texts, function(text) {
+  messages <- append_message(text, "user", chat_history)
+  chat("llama3.1", messages, output = "req")
+})
+
+# make parallel requests and get response
+resps <- req_perform_parallel(reqs)  # list of httr2_request objects
+
+# process the responses
+bind_rows(lapply(resps, resp_process, "df"))  # get responses as dataframes
+# # A tibble: 3 × 4
+#   model    role      content  created_at                 
+#   <chr>    <chr>     <chr>    <chr>                      
+# 1 llama3.1 assistant Positive 2024-08-05T17:54:27.758618Z
+# 2 llama3.1 assistant negative 2024-08-05T17:54:27.657525Z
+# 3 llama3.1 assistant other    2024-08-05T17:54:27.657067Z
+
+```

README.md

@@ -323,3 +323,88 @@ messages[[4]]  # get 2nd message
 # delete a message at a specific index/position (2nd position in the example below)
 messages <- delete_message(messages, 2)
 ```
+
+## Advanced usage
+
+### Parallel requests
+
+For the `generate()` and `chat()` endpoints/functions, you can make
+parallel requests with the `req_perform_parallel` function from the
+`httr2` library. You need to specify `output = 'req'` in the function so
+the functions return `httr2_request` objects instead of `httr2_response`
+objects.
+
+``` r
+library(httr2)
+
+prompt <- "Tell me a 5-word story"
+
+# create 5 httr2_request objects that generate a response to the same prompt
+reqs <- lapply(1:5, function(r) generate("llama3.1", prompt, output = "req"))
+
+# make parallel requests and get response
+resps <- req_perform_parallel(reqs)  # list of httr2_request objects
+
+# process the responses
+sapply(resps, resp_process, "text")  # get responses as text
+# [1] "She found him in Paris."         "She found the key upstairs."    
+# [3] "She found her long-lost sister." "She found love on Mars."        
+# [5] "She found the diamond ring."    
+```
+
+Example sentiment analysis with parallel requests with `generate()`
+function
+
+``` r
+library(httr2)
+library(glue)
+library(dplyr)
+
+# text to classify
+texts <- c('I love this product', 'I hate this product', 'I am neutral about this product')
+
+# create httr2_request objects for each text, using the same system prompt
+reqs <- lapply(texts, function(text) {
+  prompt <- glue("Your only task/role is to evaluate the sentiment of product reviews, and your response should be one of the following:'positive', 'negative', or 'other'. Product review: {text}")
+  generate("llama3.1", prompt, output = "req")
+})
+
+# make parallel requests and get response
+resps <- req_perform_parallel(reqs)  # list of httr2_request objects
+
+# process the responses
+sapply(resps, resp_process, "text")  # get responses as text
+# [1] "Positive"                            "Negative."                          
+# [3] "'neutral' translates to... 'other'."
+```
+
+Example sentiment analysis with parallel requests with `chat()` function
+
+``` r
+library(httr2)
+library(dplyr)
+
+# text to classify
+texts <- c('I love this product', 'I hate this product', 'I am neutral about this product')
+
+# create system prompt
+chat_history <- create_message("Your only task/role is to evaluate the sentiment of product reviews provided by the user. Your response should simply be 'positive', 'negative', or 'other'.", "system")
+
+# create httr2_request objects for each text, using the same system prompt
+reqs <- lapply(texts, function(text) {
+  messages <- append_message(text, "user", chat_history)
+  chat("llama3.1", messages, output = "req")
+})
+
+# make parallel requests and get response
+resps <- req_perform_parallel(reqs)  # list of httr2_request objects
+
+# process the responses
+bind_rows(lapply(resps, resp_process, "df"))  # get responses as dataframes
+# # A tibble: 3 × 4
+#   model    role      content  created_at                 
+#   <chr>    <chr>     <chr>    <chr>                      
+# 1 llama3.1 assistant Positive 2024-08-05T17:54:27.758618Z
+# 2 llama3.1 assistant negative 2024-08-05T17:54:27.657525Z
+# 3 llama3.1 assistant other    2024-08-05T17:54:27.657067Z
+```

tests/testthat/test-chat.R

@@ -11,6 +11,8 @@ test_that("chat function works with basic input", {
     # incorrect output type
     expect_error(chat("llama3", messages, output = "abc"))
 
+    expect_s3_class(chat("llama3.1", messages, output = "req"), "httr2_request")
+
     # not streaming
     expect_s3_class(chat("llama3", messages), "httr2_response")
     expect_s3_class(chat("llama3", messages, output = "resp"), "httr2_response")

tests/testthat/test-generate.R

@@ -7,6 +7,8 @@ test_that("generate function works with different outputs and resp_process", {
     # incorrect output type
     expect_error(generate("llama3", "The sky is...", output = "abc"))
 
+    expect_s3_class(generate("llama3.1", "tell me a 5-word story", output = "req"), "httr2_request")
+
     # not streaming
     expect_s3_class(generate("llama3", "The sky is..."), "httr2_response")
     expect_s3_class(generate("llama3", "The sky is...", output = "resp"), "httr2_response")