From 7fb12cbb668dcb29a68f99650e1d7e29ab2b59e6 Mon Sep 17 00:00:00 2001
From: a252jain <a252jain@uwaterloo.ca>
Date: Fri, 22 Mar 2024 09:27:06 -0400
Subject: [PATCH] broken commit
---
Package.resolved | 45 +++
Package.swift | 4 +-
.../SwiftNLP/2. Encoding/CoreMLEncoder.swift | 15 +-
.../Tokenizers/BertTokenizer.swift | 5 +-
.../2. Encoding/all-MiniLM-L6-v2.swift | 306 ------------------
.../SwiftNLP/2. Encoding/generic-model.swift | 140 --------
.../SwiftNLPTests/AllMiniLM_sampleTest.swift | 1 -
7 files changed, 60 insertions(+), 456 deletions(-)
delete mode 100644 Sources/SwiftNLP/2. Encoding/all-MiniLM-L6-v2.swift
delete mode 100644 Sources/SwiftNLP/2. Encoding/generic-model.swift
diff --git a/Package.resolved b/Package.resolved
index 187ae88b..377c4a1a 100644
--- a/Package.resolved
+++ b/Package.resolved
@@ -27,6 +27,15 @@
"version" : "2.2.0"
}
},
+ {
+ "identity" : "faissmobile",
+ "kind" : "remoteSourceControl",
+ "location" : "https://github.com/jkrukowski/FaissMobile",
+ "state" : {
+ "revision" : "9d5b9925305eea9398cc92ce4a8e51c8a4b043af",
+ "version" : "0.0.1"
+ }
+ },
{
"identity" : "similarity-topology",
"kind" : "remoteSourceControl",
@@ -36,6 +45,24 @@
"version" : "0.1.14"
}
},
+ {
+ "identity" : "swift-argument-parser",
+ "kind" : "remoteSourceControl",
+ "location" : "https://github.com/apple/swift-argument-parser",
+ "state" : {
+ "revision" : "c8ed701b513cf5177118a175d85fbbbcd707ab41",
+ "version" : "1.3.0"
+ }
+ },
+ {
+ "identity" : "swift-log",
+ "kind" : "remoteSourceControl",
+ "location" : "https://github.com/apple/swift-log",
+ "state" : {
+ "revision" : "e97a6fcb1ab07462881ac165fdbb37f067e205d5",
+ "version" : "1.5.4"
+ }
+ },
{
"identity" : "swift-numerics",
"kind" : "remoteSourceControl",
@@ -62,6 +89,24 @@
"revision" : "6d90636e22510c2f0798f9f8ff072109e345750a",
"version" : "1.1.0"
}
+ },
+ {
+ "identity" : "swiftfaiss",
+ "kind" : "remoteSourceControl",
+ "location" : "https://github.com/jkrukowski/SwiftFaiss.git",
+ "state" : {
+ "revision" : "d3831c1e9898695ae7f680b6353e48e873d3f1d3",
+ "version" : "0.0.8"
+ }
+ },
+ {
+ "identity" : "swiftformat",
+ "kind" : "remoteSourceControl",
+ "location" : "https://github.com/nicklockwood/SwiftFormat",
+ "state" : {
+ "revision" : "dbc9a4406d21cc52f16caf1e299172b097145e5e",
+ "version" : "0.53.3"
+ }
}
],
"version" : 2
diff --git a/Package.swift b/Package.swift
index db7eef19..45c096b0 100644
--- a/Package.swift
+++ b/Package.swift
@@ -15,6 +15,7 @@ let package = Package(
],
dependencies: [
.package(url: "https://github.com/jbadger3/SwiftAnnoy", .upToNextMajor(from: "1.0.1")),
+ .package(url: "https://github.com/jkrukowski/SwiftFaiss.git", from: "0.0.7"),
.package(url: "https://github.com/L1MeN9Yu/Elva", .upToNextMajor(from: "2.1.3")),
.package(url: "https://github.com/JadenGeller/similarity-topology", .upToNextMajor(from: "0.1.14")),
],
@@ -29,7 +30,8 @@ let package = Package(
],
resources: [
- .process("Resources"),
+ .process("Resources/bert_vocab.txt"),
+ .process("Resources/glove.6B.50d.mmap"),
]
),
.testTarget(
diff --git a/Sources/SwiftNLP/2. Encoding/CoreMLEncoder.swift b/Sources/SwiftNLP/2. Encoding/CoreMLEncoder.swift
index d073978d..7871e3a8 100644
--- a/Sources/SwiftNLP/2. Encoding/CoreMLEncoder.swift
+++ b/Sources/SwiftNLP/2. Encoding/CoreMLEncoder.swift
@@ -59,9 +59,9 @@ class CoreMLEncoder<Scalar: BinaryFloatingPoint & Codable>: SNLPEncoder {
@available(macOS 13.0, *)
public class MiniLMEmbeddings {
- private let model: GenericLLMModel
+ private let model: LLMModel
public let tokenizer: BertTokenizer
- public let inputDimention: Int = 512
+ public let inputDimention: Int = 128
public let outputDimention: Int = 384
public init() {
@@ -69,12 +69,12 @@ public class MiniLMEmbeddings {
modelConfig.computeUnits = .all
do {
- self.model = try GenericLLMModel(contentsOf: URL(fileURLWithPath: "Sources/SwiftNLP/Models/all-MiniLM-L6-v2.mlmodelc"), model_name: "all-MiniLM-L6-v2", input_size: inputDimention, output_size: outputDimention)
+ self.model = try LLMModel<all_MiniLM_L6_v2>()
} catch {
fatalError("Failed to load the Core ML model. Error: \(error.localizedDescription)")
}
- self.tokenizer = BertTokenizer()
+ self.tokenizer = BertTokenizer(maxLen: self.inputDimention)
}
// MARK: - Dense Embeddings
@@ -84,15 +84,18 @@ public class MiniLMEmbeddings {
let inputTokens = tokenizer.buildModelTokens(sentence: sentence)
let (inputIds, attentionMask) = tokenizer.buildModelInputs(from: inputTokens)
+ print(inputIds.count, attentionMask.count)
+
// Send tokens through the MLModel
let embeddings = generateEmbeddings(inputIds: inputIds, attentionMask: attentionMask)
+ print(inputIds.count, attentionMask.count)
+
return embeddings
}
public func generateEmbeddings(inputIds: MLMultiArray, attentionMask: MLMultiArray) -> [Float]? {
-
- let output : GenericLLMModelOutput? = try? model.prediction(input: GenericLLMModelInput(input_ids: inputIds, attention_mask: attentionMask))
+ let output : LLMModelOutput? = try? model.prediction(input: LLMModelInput(input_ids: inputIds, attention_mask: attentionMask))
guard let embeddings = output?.embeddings else {
return nil
}
diff --git a/Sources/SwiftNLP/2. Encoding/Tokenizers/BertTokenizer.swift b/Sources/SwiftNLP/2. Encoding/Tokenizers/BertTokenizer.swift
index 376cdff5..8c1e01d3 100644
--- a/Sources/SwiftNLP/2. Encoding/Tokenizers/BertTokenizer.swift
+++ b/Sources/SwiftNLP/2. Encoding/Tokenizers/BertTokenizer.swift
@@ -6,12 +6,12 @@ import CoreML
public class BertTokenizer {
private let basicTokenizer = BasicTokenizer()
private let wordpieceTokenizer: WordpieceTokenizer
- private let maxLen = 512
+ private var maxLen = 512
private let vocab: [String: Int]
private let ids_to_tokens: [Int: String]
- public init() {
+ public init(maxLen: Int) {
let url = Bundle.module.url(forResource: "bert_vocab", withExtension: "txt")!
let vocabTxt = try! String(contentsOf: url)
let tokens = vocabTxt.split(separator: "\n").map { String($0) }
@@ -24,6 +24,7 @@ public class BertTokenizer {
self.vocab = vocab
self.ids_to_tokens = ids_to_tokens
self.wordpieceTokenizer = WordpieceTokenizer(vocab: self.vocab)
+ self.maxLen = maxLen
}
public func buildModelTokens(sentence: String) -> [Int] {
diff --git a/Sources/SwiftNLP/2. Encoding/all-MiniLM-L6-v2.swift b/Sources/SwiftNLP/2. Encoding/all-MiniLM-L6-v2.swift
deleted file mode 100644
index c2f0c441..00000000
--- a/Sources/SwiftNLP/2. Encoding/all-MiniLM-L6-v2.swift
+++ /dev/null
@@ -1,306 +0,0 @@
-//
-// all_MiniLM_L6_v2.swift
-//
-// This file was automatically generated and should not be edited.
-//
-
-import CoreML
-
-
-/// Model Prediction Input Type
-@available(macOS 12.0, iOS 15.0, tvOS 15.0, watchOS 8.0, *)
-class all_MiniLM_L6_v2Input : MLFeatureProvider {
-
- /// input_ids as 1 by 512 matrix of floats
- var input_ids: MLMultiArray
-
- /// attention_mask as 1 by 512 matrix of floats
- var attention_mask: MLMultiArray
-
- var featureNames: Set<String> {
- get {
- return ["input_ids", "attention_mask"]
- }
- }
-
- func featureValue(for featureName: String) -> MLFeatureValue? {
- if (featureName == "input_ids") {
- return MLFeatureValue(multiArray: input_ids)
- }
- if (featureName == "attention_mask") {
- return MLFeatureValue(multiArray: attention_mask)
- }
- return nil
- }
-
- init(input_ids: MLMultiArray, attention_mask: MLMultiArray) {
- self.input_ids = input_ids
- self.attention_mask = attention_mask
- }
-
- convenience init(input_ids: MLShapedArray<Float>, attention_mask: MLShapedArray<Float>) {
- self.init(input_ids: MLMultiArray(input_ids), attention_mask: MLMultiArray(attention_mask))
- }
-
-}
-
-
-/// Model Prediction Output Type
-@available(macOS 12.0, iOS 15.0, tvOS 15.0, watchOS 8.0, *)
-class all_MiniLM_L6_v2Output : MLFeatureProvider {
-
- /// Source provided by CoreML
- private let provider : MLFeatureProvider
-
- /// embeddings as multidimensional array of floats
- var embeddings: MLMultiArray {
- return self.provider.featureValue(for: "embeddings")!.multiArrayValue!
- }
-
- /// embeddings as multidimensional array of floats
- var embeddingsShapedArray: MLShapedArray<Float> {
- return MLShapedArray<Float>(self.embeddings)
- }
-
- var featureNames: Set<String> {
- return self.provider.featureNames
- }
-
- func featureValue(for featureName: String) -> MLFeatureValue? {
- return self.provider.featureValue(for: featureName)
- }
-
- init(embeddings: MLMultiArray) {
- self.provider = try! MLDictionaryFeatureProvider(dictionary: ["embeddings" : MLFeatureValue(multiArray: embeddings)])
- }
-
- init(features: MLFeatureProvider) {
- self.provider = features
- }
-}
-
-
-/// Class for model loading and prediction
-@available(macOS 12.0, iOS 15.0, tvOS 15.0, watchOS 8.0, *)
-class all_MiniLM_L6_v2 {
- let model: MLModel
-
- /// URL of model assuming it was installed in the same bundle as this class
- class var urlOfModelInThisBundle : URL {
- let bundle = Bundle(for: self)
- return bundle.url(forResource: "all-MiniLM-L6-v2", withExtension:"mlmodelc")!
- }
-
- /**
- Construct all_MiniLM_L6_v2 instance with an existing MLModel object.
-
- Usually the application does not use this initializer unless it makes a subclass of all_MiniLM_L6_v2.
- Such application may want to use `MLModel(contentsOfURL:configuration:)` and `all_MiniLM_L6_v2.urlOfModelInThisBundle` to create a MLModel object to pass-in.
-
- - parameters:
- - model: MLModel object
- */
- init(model: MLModel) {
- self.model = model
- }
-
- /**
- Construct a model with configuration
-
- - parameters:
- - configuration: the desired model configuration
-
- - throws: an NSError object that describes the problem
- */
- convenience init(configuration: MLModelConfiguration = MLModelConfiguration()) throws {
- try self.init(contentsOf: type(of:self).urlOfModelInThisBundle, configuration: configuration)
- }
-
- /**
- Construct all_MiniLM_L6_v2 instance with explicit path to mlmodelc file
- - parameters:
- - modelURL: the file url of the model
-
- - throws: an NSError object that describes the problem
- */
- convenience init(contentsOf modelURL: URL) throws {
- try self.init(model: MLModel(contentsOf: modelURL))
- }
-
- /**
- Construct a model with URL of the .mlmodelc directory and configuration
-
- - parameters:
- - modelURL: the file url of the model
- - configuration: the desired model configuration
-
- - throws: an NSError object that describes the problem
- */
- convenience init(contentsOf modelURL: URL, configuration: MLModelConfiguration) throws {
- try self.init(model: MLModel(contentsOf: modelURL, configuration: configuration))
- }
-
- /**
- Construct all_MiniLM_L6_v2 instance asynchronously with optional configuration.
-
- Model loading may take time when the model content is not immediately available (e.g. encrypted model). Use this factory method especially when the caller is on the main thread.
-
- - parameters:
- - configuration: the desired model configuration
- - handler: the completion handler to be called when the model loading completes successfully or unsuccessfully
- */
- class func load(configuration: MLModelConfiguration = MLModelConfiguration(), completionHandler handler: @escaping (Swift.Result<all_MiniLM_L6_v2, Error>) -> Void) {
- return self.load(contentsOf: self.urlOfModelInThisBundle, configuration: configuration, completionHandler: handler)
- }
-
- /**
- Construct all_MiniLM_L6_v2 instance asynchronously with optional configuration.
-
- Model loading may take time when the model content is not immediately available (e.g. encrypted model). Use this factory method especially when the caller is on the main thread.
-
- - parameters:
- - configuration: the desired model configuration
- */
- class func load(configuration: MLModelConfiguration = MLModelConfiguration()) async throws -> all_MiniLM_L6_v2 {
- return try await self.load(contentsOf: self.urlOfModelInThisBundle, configuration: configuration)
- }
-
- /**
- Construct all_MiniLM_L6_v2 instance asynchronously with URL of the .mlmodelc directory with optional configuration.
-
- Model loading may take time when the model content is not immediately available (e.g. encrypted model). Use this factory method especially when the caller is on the main thread.
-
- - parameters:
- - modelURL: the URL to the model
- - configuration: the desired model configuration
- - handler: the completion handler to be called when the model loading completes successfully or unsuccessfully
- */
- class func load(contentsOf modelURL: URL, configuration: MLModelConfiguration = MLModelConfiguration(), completionHandler handler: @escaping (Swift.Result<all_MiniLM_L6_v2, Error>) -> Void) {
- MLModel.load(contentsOf: modelURL, configuration: configuration) { result in
- switch result {
- case .failure(let error):
- handler(.failure(error))
- case .success(let model):
- handler(.success(all_MiniLM_L6_v2(model: model)))
- }
- }
- }
-
- /**
- Construct all_MiniLM_L6_v2 instance asynchronously with URL of the .mlmodelc directory with optional configuration.
-
- Model loading may take time when the model content is not immediately available (e.g. encrypted model). Use this factory method especially when the caller is on the main thread.
-
- - parameters:
- - modelURL: the URL to the model
- - configuration: the desired model configuration
- */
- class func load(contentsOf modelURL: URL, configuration: MLModelConfiguration = MLModelConfiguration()) async throws -> all_MiniLM_L6_v2 {
- let model = try await MLModel.load(contentsOf: modelURL, configuration: configuration)
- return all_MiniLM_L6_v2(model: model)
- }
-
- /**
- Make a prediction using the structured interface
-
- - parameters:
- - input: the input to the prediction as all_MiniLM_L6_v2Input
-
- - throws: an NSError object that describes the problem
-
- - returns: the result of the prediction as all_MiniLM_L6_v2Output
- */
- func prediction(input: all_MiniLM_L6_v2Input) throws -> all_MiniLM_L6_v2Output {
- return try self.prediction(input: input, options: MLPredictionOptions())
- }
-
- /**
- Make a prediction using the structured interface
-
- - parameters:
- - input: the input to the prediction as all_MiniLM_L6_v2Input
- - options: prediction options
-
- - throws: an NSError object that describes the problem
-
- - returns: the result of the prediction as all_MiniLM_L6_v2Output
- */
- func prediction(input: all_MiniLM_L6_v2Input, options: MLPredictionOptions) throws -> all_MiniLM_L6_v2Output {
- let outFeatures = try model.prediction(from: input, options:options)
- return all_MiniLM_L6_v2Output(features: outFeatures)
- }
-
- /**
- Make an asynchronous prediction using the structured interface
-
- - parameters:
- - input: the input to the prediction as all_MiniLM_L6_v2Input
- - options: prediction options
-
- - throws: an NSError object that describes the problem
-
- - returns: the result of the prediction as all_MiniLM_L6_v2Output
- */
- @available(macOS 14.0, iOS 17.0, tvOS 17.0, watchOS 10.0, *)
- func prediction(input: all_MiniLM_L6_v2Input, options: MLPredictionOptions = MLPredictionOptions()) async throws -> all_MiniLM_L6_v2Output {
- let outFeatures = try await model.prediction(from: input, options:options)
- return all_MiniLM_L6_v2Output(features: outFeatures)
- }
-
- /**
- Make a prediction using the convenience interface
-
- - parameters:
- - input_ids as 1 by 512 matrix of floats
- - attention_mask as 1 by 512 matrix of floats
-
- - throws: an NSError object that describes the problem
-
- - returns: the result of the prediction as all_MiniLM_L6_v2Output
- */
- func prediction(input_ids: MLMultiArray, attention_mask: MLMultiArray) throws -> all_MiniLM_L6_v2Output {
- let input_ = all_MiniLM_L6_v2Input(input_ids: input_ids, attention_mask: attention_mask)
- return try self.prediction(input: input_)
- }
-
- /**
- Make a prediction using the convenience interface
-
- - parameters:
- - input_ids as 1 by 512 matrix of floats
- - attention_mask as 1 by 512 matrix of floats
-
- - throws: an NSError object that describes the problem
-
- - returns: the result of the prediction as all_MiniLM_L6_v2Output
- */
-
- func prediction(input_ids: MLShapedArray<Float>, attention_mask: MLShapedArray<Float>) throws -> all_MiniLM_L6_v2Output {
- let input_ = all_MiniLM_L6_v2Input(input_ids: input_ids, attention_mask: attention_mask)
- return try self.prediction(input: input_)
- }
-
- /**
- Make a batch prediction using the structured interface
-
- - parameters:
- - inputs: the inputs to the prediction as [all_MiniLM_L6_v2Input]
- - options: prediction options
-
- - throws: an NSError object that describes the problem
-
- - returns: the result of the prediction as [all_MiniLM_L6_v2Output]
- */
- func predictions(inputs: [all_MiniLM_L6_v2Input], options: MLPredictionOptions = MLPredictionOptions()) throws -> [all_MiniLM_L6_v2Output] {
- let batchIn = MLArrayBatchProvider(array: inputs)
- let batchOut = try model.predictions(from: batchIn, options: options)
- var results : [all_MiniLM_L6_v2Output] = []
- results.reserveCapacity(inputs.count)
- for i in 0..<batchOut.count {
- let outProvider = batchOut.features(at: i)
- let result = all_MiniLM_L6_v2Output(features: outProvider)
- results.append(result)
- }
- return results
- }
-}
diff --git a/Sources/SwiftNLP/2. Encoding/generic-model.swift b/Sources/SwiftNLP/2. Encoding/generic-model.swift
deleted file mode 100644
index 9c7308cc..00000000
--- a/Sources/SwiftNLP/2. Encoding/generic-model.swift
+++ /dev/null
@@ -1,140 +0,0 @@
-import CoreML
-
-
-/// Model Prediction Input Type
-@available(macOS 12.0, iOS 15.0, tvOS 15.0, watchOS 8.0, *)
-class GenericLLMModelInput : MLFeatureProvider {
-
- var input_ids: MLMultiArray
-
- var attention_mask: MLMultiArray
-
- var featureNames: Set<String> {
- get {
- return ["input_ids", "attention_mask"]
- }
- }
-
- func featureValue(for featureName: String) -> MLFeatureValue? {
- if (featureName == "input_ids") {
- return MLFeatureValue(multiArray: input_ids)
- }
- if (featureName == "attention_mask") {
- return MLFeatureValue(multiArray: attention_mask)
- }
- return nil
- }
-
- init(input_ids: MLMultiArray, attention_mask: MLMultiArray) {
- self.input_ids = input_ids
- self.attention_mask = attention_mask
- }
-
- convenience init(input_ids: MLShapedArray<Float>, attention_mask: MLShapedArray<Float>) {
- self.init(input_ids: MLMultiArray(input_ids), attention_mask: MLMultiArray(attention_mask))
- }
-}
-
-/// Model Prediction Output Type
-@available(macOS 12.0, iOS 15.0, tvOS 15.0, watchOS 8.0, *)
-class GenericLLMModelOutput : MLFeatureProvider {
-
- private let provider : MLFeatureProvider
-
- var embeddings: MLMultiArray {
- return self.provider.featureValue(for: "embeddings")!.multiArrayValue!
- }
-
- var embeddingsShapedArray: MLShapedArray<Float> {
- return MLShapedArray<Float>(self.embeddings)
- }
-
- var featureNames: Set<String> {
- return self.provider.featureNames
- }
-
- func featureValue(for featureName: String) -> MLFeatureValue? {
- return self.provider.featureValue(for: featureName)
- }
-
- init(embeddings: MLMultiArray) {
- self.provider = try! MLDictionaryFeatureProvider(dictionary: ["embeddings" : MLFeatureValue(multiArray: embeddings)])
- }
-
- init(features: MLFeatureProvider) {
- self.provider = features
- }
-}
-
-
-/// Class for model loading and prediction
-@available(macOS 12.0, iOS 15.0, tvOS 15.0, watchOS 8.0, *)
-class GenericLLMModel {
- let model: MLModel
- let model_name: String
- let input_size: Int
- let output_size: Int
-
- init(model: MLModel, model_name: String, input_size: Int, output_size: Int) {
- self.model = model
- self.model_name = model_name
- self.input_size = input_size
- self.output_size = output_size
- }
-
- convenience init(contentsOf modelURL: URL, model_name: String, input_size: Int, output_size: Int) throws {
- try self.init(model: MLModel(contentsOf: modelURL), model_name: model_name, input_size: input_size, output_size: output_size)
- }
-
- convenience init(contentsOf modelURL: URL, configuration: MLModelConfiguration, model_name: String, input_size: Int, output_size: Int) throws {
- try self.init(model: MLModel(contentsOf: modelURL, configuration: configuration), model_name: model_name, input_size: input_size, output_size: output_size)
- }
-
- class func load(contentsOf modelURL: URL, model_name: String, input_size: Int, output_size: Int, configuration: MLModelConfiguration = MLModelConfiguration()) async throws -> GenericLLMModel {
- let model = try await MLModel.load(contentsOf: modelURL, configuration: configuration)
- return GenericLLMModel(model: model, model_name: model_name, input_size: input_size, output_size: output_size)
- }
-
- func prediction(input: GenericLLMModelInput) throws -> GenericLLMModelOutput {
- return try self.prediction(input: input, options: MLPredictionOptions())
- }
-
- func prediction(input: GenericLLMModelInput, options: MLPredictionOptions) throws -> GenericLLMModelOutput {
- let outFeatures = try model.prediction(from: input, options:options)
- return GenericLLMModelOutput(features: outFeatures)
- }
-
- @available(macOS 14.0, iOS 17.0, tvOS 17.0, watchOS 10.0, *)
- func prediction(input: GenericLLMModelInput, options: MLPredictionOptions = MLPredictionOptions()) async throws -> GenericLLMModelOutput {
- let outFeatures = try await model.prediction(from: input, options:options)
- return GenericLLMModelOutput(features: outFeatures)
- }
-
- func prediction(input_ids: MLMultiArray, attention_mask: MLMultiArray) throws -> GenericLLMModelOutput {
- let input_ = GenericLLMModelInput(input_ids: input_ids, attention_mask: attention_mask)
- return try self.prediction(input: input_)
- }
-
- func prediction(input_ids: MLShapedArray<Float>, attention_mask: MLShapedArray<Float>) throws -> GenericLLMModelOutput {
- let input_ = GenericLLMModelInput(input_ids: input_ids, attention_mask: attention_mask)
- return try self.prediction(input: input_)
- }
-
- func predictions(inputs: [GenericLLMModelInput], options: MLPredictionOptions = MLPredictionOptions()) throws -> [GenericLLMModelOutput] {
- for input in inputs {
- assert(input.input_ids.count == self.input_size)
- }
-
- let batchIn = MLArrayBatchProvider(array: inputs)
- let batchOut = try model.predictions(from: batchIn, options: options)
- var results : [GenericLLMModelOutput] = []
- results.reserveCapacity(inputs.count)
- for i in 0..<batchOut.count {
- let outProvider = batchOut.features(at: i)
- let result = GenericLLMModelOutput(features: outProvider)
- results.append(result)
- }
-
- return results
- }
-}
diff --git a/Tests/SwiftNLPTests/AllMiniLM_sampleTest.swift b/Tests/SwiftNLPTests/AllMiniLM_sampleTest.swift
index 8e837333..64e401f1 100644
--- a/Tests/SwiftNLPTests/AllMiniLM_sampleTest.swift
+++ b/Tests/SwiftNLPTests/AllMiniLM_sampleTest.swift
@@ -35,7 +35,6 @@ final class BERT_test: XCTestCase {
var embedding_dim: Int = 384
var model = MiniLMEmbeddings()
-
query_embedding = await model.encode(sentence: query[0])!
--
GitLab