Add classification support (#11)

.github/workflows/test.yml

@@ -1,4 +1,4 @@
-name: Publish
+name: Test
 
 on:
   push:

api/src/main/java/com/theokanning/openai/classification/ClassificationRequest.java

@@ -0,0 +1,120 @@
+package com.theokanning.openai.classification;
+
+import lombok.*;
+
+import java.util.List;
+import java.util.Map;
+
+/**
+ * A request for OpenAi to classify text based on provided examples
+ * All fields are nullable.
+ *
+ * Documentation taken from
+ * https://beta.openai.com/docs/api-reference/classifications/create
+ */
+@Builder
+@NoArgsConstructor
+@AllArgsConstructor
+@Data
+public class ClassificationRequest {
+
+    /**
+     * ID of the engine to use for completion
+     */
+    @NonNull
+    String model;
+
+    /**
+     * Query to be classified
+     */
+    @NonNull
+    String query;
+
+    /**
+     * A list of examples with labels, in the following format:
+     *
+     * [["The movie is so interesting.", "Positive"], ["It is quite boring.", "Negative"], ...]
+     *
+     * All the label strings will be normalized to be capitalized.
+     *
+     * You should specify either examples or file, but not both.
+     */
+    List<List<String>> examples;
+
+    /**
+     * The ID of the uploaded file that contains training examples.
+     * See upload file for how to upload a file of the desired format and purpose.
+     *
+     * You should specify either examples or file, but not both.
+     */
+    String file;
+
+    /**
+     * The set of categories being classified.
+     * If not specified, candidate labels will be automatically collected from the examples you provide.
+     * All the label strings will be normalized to be capitalized.
+     */
+    List<String> labels;
+
+    /**
+     * ID of the engine to use for Search. You can select one of ada, babbage, curie, or davinci.
+     */
+    String searchModel;
+
+    /**
+     * What sampling temperature to use. Higher values means the model will take more risks.
+     * Try 0.9 for more creative applications, and 0 (argmax sampling) for ones with a well-defined answer.
+     *
+     * We generally recommend using this or {@link top_p} but not both.
+     */
+    Double temperature;
+
+    /**
+     * Include the log probabilities on the logprobs most likely tokens, as well the chosen tokens.
+     * For example, if logprobs is 10, the API will return a list of the 10 most likely tokens.
+     * The API will always return the logprob of the sampled token,
+     * so there may be up to logprobs+1 elements in the response.
+     */
+    Integer logprobs;
+
+    /**
+     * The maximum number of examples to be ranked by Search when using file.
+     * Setting it to a higher value leads to improved accuracy but with increased latency and cost.
+     */
+    Integer maxExamples;
+
+    /**
+     * Modify the likelihood of specified tokens appearing in the completion.
+     *
+     * Accepts a json object that maps tokens (specified by their token ID in the GPT tokenizer) to an
+     * associated bias value from -100 to 100.
+     */
+    Map<String, Double> logitBias;
+
+    /**
+     * If set to true, the returned JSON will include a "prompt" field containing the final prompt that was
+     * used to request a completion. This is mainly useful for debugging purposes.
+     */
+    Boolean returnPrompt;
+
+    /**
+     * A special boolean flag for showing metadata.
+     * If set to true, each document entry in the returned JSON will contain a "metadata" field.
+     *
+     * This flag only takes effect when file is set.
+     */
+    Boolean returnMetadata;
+
+    /**
+     * If an object name is in the list, we provide the full information of the object;
+     * otherwise, we only provide the object ID.
+     *
+     * Currently we support completion and file objects for expansion.
+     */
+    List<String> expand;
+
+    /**
+     * A unique identifier representing your end-user, which will help OpenAI to monitor and detect abuse.
+     */
+    String user;
+}

api/src/main/java/com/theokanning/openai/classification/ClassificationResult.java

@@ -0,0 +1,45 @@
+package com.theokanning.openai.classification;
+
+import com.theokanning.openai.completion.CompletionChoice;
+import lombok.Data;
+
+import java.util.List;
+
+/**
+ * An object containing a response from the classification api
+ * <p>
+ * https://beta.openai.com/docs/api-reference/classifications/create
+ */
+@Data
+public class ClassificationResult {
+
+    /**
+     * A unique id assigned to this completion
+     */
+    String completion;
+
+    /**
+     * The predicted label for the query text.
+     */
+    String label;
+
+    /**
+     * The GPT-3 model used for completion
+     */
+    String model;
+
+    /**
+     * The type of object returned, should be "classification"
+     */
+    String object;
+
+    /**
+     * The GPT-3 model used for search
+     */
+    String searchModel;
+
+    /**
+     * A list of the most relevant examples for the query text.
+     */
+    List<Example> selectedExamples;
+}

api/src/main/java/com/theokanning/openai/classification/Example.java

@@ -0,0 +1,26 @@
+package com.theokanning.openai.classification;
+
+import lombok.Data;
+
+/**
+ * Represents an example returned by the classification api
+ *
+ * https://beta.openai.com/docs/api-reference/classifications/create
+ */
+@Data
+public class Example {
+    /**
+     * The position of this example in the example list
+     */
+    Integer document;
+
+    /**
+     * The label of the example
+     */
+    String label;
+
+    /**
+     * The text of the example
+     */
+    String text;
+}

api/src/main/java/com/theokanning/openai/finetune/FineTuneRequest.java

@@ -1,9 +1,6 @@
 package com.theokanning.openai.finetune;
 
-import lombok.AllArgsConstructor;
+import lombok.*;
-import lombok.Builder;
-import lombok.Data;
-import lombok.NoArgsConstructor;
 
 import java.util.List;
 
@@ -22,6 +19,7 @@ public class FineTuneRequest {
     /**
      * The ID of an uploaded file that contains training data.
      */
+    @NonNull
     String trainingFile;
 
     /**

client/src/main/java/com/theokanning/openai/OpenAiApi.java

@@ -1,5 +1,7 @@
 package com.theokanning.openai;
 
+import com.theokanning.openai.classification.ClassificationRequest;
+import com.theokanning.openai.classification.ClassificationResult;
 import com.theokanning.openai.engine.Engine;
 import com.theokanning.openai.file.File;
 import com.theokanning.openai.finetune.FineTuneRequest;
@@ -28,6 +30,9 @@ public interface OpenAiApi {
     @POST("/v1/engines/{engine_id}/search")
     Single<OpenAiResponse<SearchResult>> search(@Path("engine_id") String engineId, @Body SearchRequest request);
 
+    @POST("v1/classifications")
+    Single<ClassificationResult> createClassification(@Body ClassificationRequest request);
+
     @GET("/v1/files")
     Single<OpenAiResponse<File>> listFiles();
 

client/src/main/java/com/theokanning/openai/OpenAiService.java

@@ -4,6 +4,8 @@ import com.fasterxml.jackson.annotation.JsonInclude;
 import com.fasterxml.jackson.databind.DeserializationFeature;
 import com.fasterxml.jackson.databind.ObjectMapper;
 import com.fasterxml.jackson.databind.PropertyNamingStrategy;
+import com.theokanning.openai.classification.ClassificationRequest;
+import com.theokanning.openai.classification.ClassificationResult;
 import com.theokanning.openai.file.File;
 import com.theokanning.openai.finetune.FineTuneRequest;
 import com.theokanning.openai.finetune.FineTuneEvent;
@@ -62,6 +64,10 @@ public class OpenAiService {
         return api.search(engineId, request).blockingGet().data;
     }
 
+    public ClassificationResult createClassification(ClassificationRequest request) {
+        return api.createClassification(request).blockingGet();
+    }
+
     public List<File> listFiles() {
         return api.listFiles().blockingGet().data;
     }

client/src/test/java/com/theokanning/openai/ClassificationTest.java

@@ -0,0 +1,39 @@
+package com.theokanning.openai;
+
+import com.theokanning.openai.classification.ClassificationRequest;
+import com.theokanning.openai.classification.ClassificationResult;
+import com.theokanning.openai.completion.CompletionChoice;
+import com.theokanning.openai.completion.CompletionRequest;
+import org.junit.jupiter.api.Test;
+
+import java.util.Arrays;
+import java.util.List;
+
+import static org.junit.jupiter.api.Assertions.assertFalse;
+import static org.junit.jupiter.api.Assertions.assertNotNull;
+
+
+public class ClassificationTest {
+
+    String token = System.getenv("OPENAI_TOKEN");
+    OpenAiService service = new OpenAiService(token);
+
+    @Test
+    void createCompletion() {
+        ClassificationRequest classificationRequest = ClassificationRequest.builder()
+                .examples(Arrays.asList(
+                        Arrays.asList("A happy moment", "Positive"),
+                        Arrays.asList("I am sad.", "Negative"),
+                        Arrays.asList("I am feeling awesome", "Positive")
+                ))
+                .query("It is a raining day :(")
+                .model("curie")
+                .searchModel("ada")
+                .labels(Arrays.asList("Positive", "Negative", "Neutral"))
+                .build();
+
+        ClassificationResult result = service.createClassification(classificationRequest);
+
+        assertNotNull(result.getCompletion());
+    }
+}

client/src/test/java/com/theokanning/openai/FineTuneTest.java

@@ -6,6 +6,7 @@ import com.theokanning.openai.finetune.FineTuneResult;
 import org.junit.jupiter.api.*;
 
 import java.util.List;
+import java.util.concurrent.TimeUnit;
 
 import static org.junit.jupiter.api.Assertions.*;
 
@@ -17,10 +18,13 @@ public class FineTuneTest {
 
 
     @BeforeAll
-   static void setup() {
+   static void setup() throws Exception {
         String token = System.getenv("OPENAI_TOKEN");
         service = new OpenAiService(token);
         fileId = service.uploadFile("fine-tune", "src/test/resources/fine-tuning-data.jsonl").getId();
+
+        // wait for file to be processed
+        TimeUnit.SECONDS.sleep(10);
     }
 
     @AfterAll