restore original implementation with new names

Author: ejones

examples/common.cpp

@@ -128,12 +128,8 @@ bool gpt_params_parse(int argc, char ** argv, gpt_params & params) {
                 break;
             }
             params.path_prompt_cache = argv[i];
-        } else if (arg == "--session") {
-            if (++i >= argc) {
-                invalid_param = true;
-                break;
-            }
-            params.path_session = argv[i];
+        } else if (arg == "--prompt-cache-all") {
+            params.prompt_cache_save_all = true;
         } else if (arg == "-f" || arg == "--file") {
             if (++i >= argc) {
                 invalid_param = true;
@@ -350,11 +346,6 @@ bool gpt_params_parse(int argc, char ** argv, gpt_params & params) {
         gpt_print_usage(argc, argv, default_params);
         exit(1);
     }
-    if (!params.path_session.empty() && !params.path_prompt_cache.empty()) {
-        fprintf(stderr, "error: only one of --prompt-cache or --session may be specified\n");
-        gpt_print_usage(argc, argv, default_params);
-        exit(1);
-    }
     if (escape_prompt) {
         process_escapes(params.prompt);
     }
@@ -380,7 +371,7 @@ void gpt_print_usage(int /*argc*/, char ** argv, const gpt_params & params) {
     fprintf(stderr, "                        prompt to start generation with (default: empty)\n");
     fprintf(stderr, "  -e                    process prompt escapes sequences (\\n, \\r, \\t, \\', \\\", \\\\)\n");
     fprintf(stderr, "  --prompt-cache FNAME  file to cache prompt state for faster startup (default: none)\n");
-    fprintf(stderr, "  --session FNAME       file to store prompt and generations, allowing continuation (default: none)\n");
+    fprintf(stderr, "  --prompt-cache-all    if specified, saves user input and generations to cache as well\n");
     fprintf(stderr, "  --random-prompt       start with a randomized prompt.\n");
     fprintf(stderr, "  --in-prefix STRING    string to prefix user inputs with (default: empty)\n");
     fprintf(stderr, "  --in-suffix STRING    string to suffix after user inputs with (default: empty)\n");

examples/common.h

@@ -42,18 +42,18 @@ struct gpt_params {
     std::string model  = "models/lamma-7B/ggml-model.bin"; // model path
     std::string prompt = "";
     std::string path_prompt_cache = "";  // path to file for saving/loading prompt eval state
-    std::string path_session      = "";  // file for saving/loading prompt and generations
     std::string input_prefix      = "";  // string to prefix user inputs with
     std::string input_suffix      = "";  // string to suffix user inputs with
     std::vector<std::string> antiprompt; // string upon seeing which more user input is prompted
 
     std::string lora_adapter = "";  // lora adapter path
     std::string lora_base = "";     // base model path for the lora adapter
 
-    bool memory_f16        = true;  // use f16 instead of f32 for memory kv
-    bool random_prompt     = false; // do not randomize prompt if none provided
-    bool use_color         = false; // use color to distinguish generations and inputs
-    bool interactive       = false; // interactive mode
+    bool memory_f16            = true;  // use f16 instead of f32 for memory kv
+    bool random_prompt         = false; // do not randomize prompt if none provided
+    bool use_color             = false; // use color to distinguish generations and inputs
+    bool interactive           = false; // interactive mode
+    bool prompt_cache_save_all = false; // save user input and generations to prompt cache
 
     bool embedding         = false; // get only sentence embedding
     bool interactive_first = false; // wait for user input immediately

examples/main/main.cpp

@@ -140,10 +140,9 @@ int main(int argc, char ** argv) {
     // Add a space in front of the first character to match OG llama tokenizer behavior
     params.prompt.insert(0, 1, ' ');
 
-    std::string path_session =
-        !params.path_session.empty() ? params.path_session : params.path_prompt_cache;
+    std::string path_session    = params.path_prompt_cache;
+    const bool session_save_all = params.prompt_cache_save_all;
     std::vector<llama_token> session_tokens;
-    bool resume_session = !params.path_session.empty();
 
     if (!path_session.empty()) {
         fprintf(stderr, "%s: attempting to load saved session from '%s'\n", __func__, path_session.c_str());
@@ -316,8 +315,8 @@ int main(int argc, char ** argv) {
                 // insert n_left/2 tokens at the start of embd from last_n_tokens
                 embd.insert(embd.begin(), last_n_tokens.begin() + n_ctx - n_left/2 - embd.size(), last_n_tokens.end() - embd.size());
 
-                // stop saving session if we run out of context, saving whatever was evaled
-                if (!path_session.empty() && resume_session) {
+                // stop saving session if we run out of context
+                if (!path_session.empty() && session_save_all) {
                     llama_save_session_file(ctx, path_session.c_str(),
                         session_tokens.data(), session_tokens.size());
                 }
@@ -621,7 +620,7 @@ int main(int argc, char ** argv) {
         }
     }
 
-    if (!path_session.empty() && resume_session) {
+    if (!path_session.empty() && session_save_all) {
         fprintf(stderr, "\n%s: saving final output to session file '%s'\n", __func__, path_session.c_str());
         llama_save_session_file(ctx, path_session.c_str(), session_tokens.data(), session_tokens.size());
     }