From f73fbf28a46384d2c07c29c84b3576730916018e Mon Sep 17 00:00:00 2001
From: Adam Treat <treat.adam@gmail.com>
Date: Mon, 17 Apr 2023 14:11:41 -0400
Subject: [PATCH] Fix the context.

---
 llm.cpp  | 19 ++++++++++++++++---
 llm.h    |  5 ++++-
 main.qml |  2 +-
 3 files changed, 21 insertions(+), 5 deletions(-)

diff --git a/llm.cpp b/llm.cpp
index 1d904511..db13807a 100644
--- a/llm.cpp
+++ b/llm.cpp
@@ -64,7 +64,7 @@ bool LLMObject::isModelLoaded() const
     return m_llmodel->isModelLoaded();
 }
 
-void LLMObject::resetResponse()
+void LLMObject::regenerateResponse()
 {
     s_ctx.n_past -= m_responseTokens;
     s_ctx.n_past = std::max(0, s_ctx.n_past);
@@ -75,9 +75,17 @@ void LLMObject::resetResponse()
     emit responseChanged();
 }
 
+void LLMObject::resetResponse()
+{
+    m_responseTokens = 0;
+    m_responseLogits = 0;
+    m_response = std::string();
+    emit responseChanged();
+}
+
 void LLMObject::resetContext()
 {
-    resetResponse();
+    regenerateResponse();
     s_ctx = LLModel::PromptContext();
 }
 
@@ -142,7 +150,6 @@ bool LLMObject::prompt(const QString &prompt, const QString &prompt_template, in
     auto func = std::bind(&LLMObject::handleResponse, this, std::placeholders::_1);
     emit responseStarted();
     qint32 logitsBefore = s_ctx.logits.size();
-    qInfo() << instructPrompt << "\n";
     m_llmodel->prompt(instructPrompt.toStdString(), func, s_ctx, n_predict, top_k, top_p, temp, n_batch);
     m_responseLogits += s_ctx.logits.size() - logitsBefore;
     std::string trimmed = trim_whitespace(m_response);
@@ -167,6 +174,7 @@ LLM::LLM()
     connect(m_llmodel, &LLMObject::modelNameChanged, this, &LLM::modelNameChanged, Qt::QueuedConnection);
 
     connect(this, &LLM::promptRequested, m_llmodel, &LLMObject::prompt, Qt::QueuedConnection);
+    connect(this, &LLM::regenerateResponseRequested, m_llmodel, &LLMObject::regenerateResponse, Qt::BlockingQueuedConnection);
     connect(this, &LLM::resetResponseRequested, m_llmodel, &LLMObject::resetResponse, Qt::BlockingQueuedConnection);
     connect(this, &LLM::resetContextRequested, m_llmodel, &LLMObject::resetContext, Qt::BlockingQueuedConnection);
 }
@@ -182,6 +190,11 @@ void LLM::prompt(const QString &prompt, const QString &prompt_template, int32_t
     emit promptRequested(prompt, prompt_template, n_predict, top_k, top_p, temp, n_batch);
 }
 
+void LLM::regenerateResponse()
+{
+    emit regenerateResponseRequested(); // blocking queued connection
+}
+
 void LLM::resetResponse()
 {
     emit resetResponseRequested(); // blocking queued connection
diff --git a/llm.h b/llm.h
index 03db9f9a..33aa95c6 100644
--- a/llm.h
+++ b/llm.h
@@ -18,6 +18,7 @@ public:
 
     bool loadModel();
     bool isModelLoaded() const;
+    void regenerateResponse();
     void resetResponse();
     void resetContext();
     void stopGenerating() { m_stopGenerating = true; }
@@ -63,8 +64,9 @@ public:
     Q_INVOKABLE bool isModelLoaded() const;
     Q_INVOKABLE void prompt(const QString &prompt, const QString &prompt_template, int32_t n_predict, int32_t top_k, float top_p,
                             float temp, int32_t n_batch);
-    Q_INVOKABLE void resetContext();
+    Q_INVOKABLE void regenerateResponse();
     Q_INVOKABLE void resetResponse();
+    Q_INVOKABLE void resetContext();
     Q_INVOKABLE void stopGenerating();
 
     QString response() const;
@@ -80,6 +82,7 @@ Q_SIGNALS:
     void responseInProgressChanged();
     void promptRequested(const QString &prompt, const QString &prompt_template, int32_t n_predict, int32_t top_k, float top_p,
                          float temp, int32_t n_batch);
+    void regenerateResponseRequested();
     void resetResponseRequested();
     void resetContextRequested();
     void modelNameChanged();
diff --git a/main.qml b/main.qml
index a1d2be79..18b48af1 100644
--- a/main.qml
+++ b/main.qml
@@ -666,7 +666,7 @@ Window {
                 if (LLM.responseInProgress)
                     LLM.stopGenerating()
                 else {
-                    LLM.resetResponse()
+                    LLM.regenerateResponse()
                     if (chatModel.count) {
                         var listElement = chatModel.get(chatModel.count - 1)
                         if (listElement.name === qsTr("Response: ")) {