Skip to content

Switch back to gpt2 512 model. Fixes gibberish gpt2 output #29

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Switch back to gpt2 512 model. Fixes gibberish gpt2 output #29

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 12 additions & 0 deletions CoreMLBert.xcodeproj/project.pbxproj
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -68,6 +68,10 @@
79F2CCA022C666C7009F8551 /* question_tokens.json in Resources */ = {isa = PBXBuildFile; fileRef = 79F2CC9F22C666C7009F8551 /* question_tokens.json */; };
79F2CCA222C6717E009F8551 /* LoaderView.swift in Sources */ = {isa = PBXBuildFile; fileRef = 79F2CCA122C6717D009F8551 /* LoaderView.swift */; };
79F7060E22EA0CA900C4432C /* BERTSQUADFP16.mlmodel in Sources */ = {isa = PBXBuildFile; fileRef = 79F2CC9022C5590C009F8551 /* BERTSQUADFP16.mlmodel */; };
7E7B9A4029099D44004914F5 /* gpt2-512.mlmodel in Sources */ = {isa = PBXBuildFile; fileRef = 7E7B9A3F29099D44004914F5 /* gpt2-512.mlmodel */; };
7E7B9A4129099D44004914F5 /* gpt2-512.mlmodel in Sources */ = {isa = PBXBuildFile; fileRef = 7E7B9A3F29099D44004914F5 /* gpt2-512.mlmodel */; };
7E7B9A4329099ED1004914F5 /* gpt2.mlmodel in Sources */ = {isa = PBXBuildFile; fileRef = 7E7B9A4229099ED1004914F5 /* gpt2.mlmodel */; };
7E7B9A4429099ED1004914F5 /* gpt2.mlmodel in Sources */ = {isa = PBXBuildFile; fileRef = 7E7B9A4229099ED1004914F5 /* gpt2.mlmodel */; };
/* End PBXBuildFile section */

/* Begin PBXContainerItemProxy section */
Expand Down Expand Up @@ -136,6 +140,8 @@
79F2CC9D22C57825009F8551 /* BertForQATests.swift */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.swift; path = BertForQATests.swift; sourceTree = "<group>"; };
79F2CC9F22C666C7009F8551 /* question_tokens.json */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = text.json; path = question_tokens.json; sourceTree = "<group>"; };
79F2CCA122C6717D009F8551 /* LoaderView.swift */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.swift; path = LoaderView.swift; sourceTree = "<group>"; };
7E7B9A3F29099D44004914F5 /* gpt2-512.mlmodel */ = {isa = PBXFileReference; lastKnownFileType = file.mlmodel; path = "gpt2-512.mlmodel"; sourceTree = "<group>"; };
7E7B9A4229099ED1004914F5 /* gpt2.mlmodel */ = {isa = PBXFileReference; lastKnownFileType = file.mlmodel; path = gpt2.mlmodel; sourceTree = "<group>"; };
/* End PBXFileReference section */

/* Begin PBXFrameworksBuildPhase section */
Expand Down Expand Up @@ -179,7 +185,9 @@
79F2CC8422C50F87009F8551 /* vocab.txt */,
79F2CC8722C51E03009F8551 /* basic_tokenized_questions.json */,
79F2CC8D22C55413009F8551 /* tokenized_questions.json */,
7E7B9A3F29099D44004914F5 /* gpt2-512.mlmodel */,
79F2CC9F22C666C7009F8551 /* question_tokens.json */,
7E7B9A4229099ED1004914F5 /* gpt2.mlmodel */,
79F2CC9022C5590C009F8551 /* BERTSQUADFP16.mlmodel */,
79908C13234E95FB00D0FE5B /* distilbert-squad-384.mlmodel */,
79908C18234EAB5300D0FE5B /* distilbert-squad-384_FP16.mlmodel */,
Expand Down Expand Up @@ -462,7 +470,9 @@
796DF55022E1026700140C02 /* AppDelegate.swift in Sources */,
796DF57222E1039C00140C02 /* Utils.swift in Sources */,
79D94AD6234CE4830033EA7D /* gpt2-64-12.mlmodel in Sources */,
7E7B9A4129099D44004914F5 /* gpt2-512.mlmodel in Sources */,
796DF57422E1039C00140C02 /* MLMultiArray+Utils.swift in Sources */,
7E7B9A4429099ED1004914F5 /* gpt2.mlmodel in Sources */,
796DF57022E1039C00140C02 /* BertTokenizer.swift in Sources */,
796DF57522E1039C00140C02 /* GPT2Tokenizer.swift in Sources */,
796DF55222E1026700140C02 /* SceneDelegate.swift in Sources */,
Expand All @@ -486,6 +496,7 @@
buildActionMask = 2147483647;
files = (
79F2CC9A22C57132009F8551 /* MLMultiArray+Utils.swift in Sources */,
7E7B9A4329099ED1004914F5 /* gpt2.mlmodel in Sources */,
79F2CCA222C6717E009F8551 /* LoaderView.swift in Sources */,
79F2CC6022C50078009F8551 /* ViewController.swift in Sources */,
79F2CC8122C5041C009F8551 /* SquadDataset.swift in Sources */,
Expand All @@ -495,6 +506,7 @@
79F2CC5C22C50078009F8551 /* AppDelegate.swift in Sources */,
796DF51022E0EB1D00140C02 /* GPT2Tokenizer.swift in Sources */,
79F2CC9422C56693009F8551 /* BertForQuestionAnswering.swift in Sources */,
7E7B9A4029099D44004914F5 /* gpt2-512.mlmodel in Sources */,
79908C19234EAB5400D0FE5B /* distilbert-squad-384_FP16.mlmodel in Sources */,
79F2CC9122C5590C009F8551 /* BERTSQUADFP16.mlmodel in Sources */,
79F2CC8322C50E00009F8551 /* BertTokenizer.swift in Sources */,
Expand Down
4 changes: 2 additions & 2 deletions Sources/GPT2.swift
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,9 +21,9 @@ class GPT2 {
case topP(Double)
}

private let model = distilgpt2_64_6()
private let model = gpt2_512()
public let tokenizer = GPT2Tokenizer()
public let seqLen = 64
public let seqLen = 512
private let strategy: DecodingStrategy

init(strategy: DecodingStrategy = .greedy) {
Expand Down