add command line parser, simplify code

examples/simple/simple.cpp

@@ -1,37 +1,84 @@
 #include "llama.h"
 #include <cstdio>
+#include <cstring>
 #include <string>
 #include <vector>
 
 static void print_usage(int, char ** argv) {
     printf("\nexample usage:\n");
-    printf("\n    %s <model.gguf> [prompt]\n", argv[0]);
+    printf("\n    %s -m model.gguf [-n n_predict] [-ngl n_gpu_layers] [prompt]\n", argv[0]);
     printf("\n");
 }
 
 int main(int argc, char ** argv) {
+    // path to the model gguf file
     std::string model_path;
+    // prompt to generate text from
     std::string prompt = "Hello my name is";
+    // number of layers to offload to the GPU
+    int ngl = 99;
+    // number of tokens to predict
     int n_predict = 32;
 
-    if (argc < 2) {
-        print_usage(argc, argv);
-        return 1;
-    }
-    model_path = argv[1];
+    // parse command line arguments
 
-    if (argc > 2) {
-        prompt = argv[2];
-        for (int i = 3; i < argc; i++) {
-            prompt += " ";
-            prompt += argv[i];
+    {
+        int i = 1;
+        for (; i < argc; i++) {
+            if (strcmp(argv[i], "-m") == 0) {
+                if (i + 1 < argc) {
+                    model_path = argv[++i];
+                } else {
+                    print_usage(argc, argv);
+                    return 1;
+                }
+            } else if (strcmp(argv[i], "-n") == 0) {
+                if (i + 1 < argc) {
+                    try {
+                        n_predict = std::stoi(argv[++i]);
+                    } catch (...) {
+                        print_usage(argc, argv);
+                        return 1;
+                    }
+                } else {
+                    print_usage(argc, argv);
+                    return 1;
+                }
+            } else if (strcmp(argv[i], "-ngl") == 0) {
+                if (i + 1 < argc) {
+                    try {
+                        ngl = std::stoi(argv[++i]);
+                    } catch (...) {
+                        print_usage(argc, argv);
+                        return 1;
+                    }
+                } else {
+                    print_usage(argc, argv);
+                    return 1;
+                }
+            } else {
+                // prompt starts here
+                break;
+            }
+        }
+        if (model_path.empty()) {
+            print_usage(argc, argv);
+            return 1;
+        }
+        if (i < argc) {
+            prompt = argv[i++];
+            for (; i < argc; i++) {
+                prompt += " ";
+                prompt += argv[i];
+            }
         }
     }
 
     // initialize the model
 
     llama_model_params model_params = llama_model_default_params();
-    model_params.n_gpu_layers = 99; // offload all layers to GPU
+    model_params.n_gpu_layers = ngl;
+
     llama_model * model = llama_load_model_from_file(model_path.c_str(), model_params);
 
     if (model == NULL) {
@@ -39,11 +86,28 @@ int main(int argc, char ** argv) {
         return 1;
     }
 
+    // tokenize the prompt
+
+    // find the number of tokens in the prompt
+    const int n_prompt = -llama_tokenize(model, prompt.c_str(), prompt.size(), NULL, 0, true, true);
+
+    // allocate space for the tokens and tokenize the prompt
+    std::vector<llama_token> prompt_tokens(n_prompt);
+    if (llama_tokenize(model, prompt.c_str(), prompt.size(), prompt_tokens.data(), prompt_tokens.size(), true, true) < 0) {
+        fprintf(stderr, "%s: error: failed to tokenize the prompt\n", __func__);
+        return 1;
+    }
+
     // initialize the context
 
     llama_context_params ctx_params = llama_context_default_params();
-    ctx_params.n_ctx = 512; // maximum context size
+    // n_ctx is the context size
+    ctx_params.n_ctx = n_prompt + n_predict - 1;
+    // n_batch is the maximum number of tokens that can be processed in a single call to llama_decode
+    ctx_params.n_batch = n_prompt;
+    // enable performance counters
     ctx_params.no_perf = false;
+
     llama_context * ctx = llama_new_context_with_model(model, ctx_params);
 
     if (ctx == NULL) {
@@ -51,40 +115,17 @@ int main(int argc, char ** argv) {
         return 1;
     }
 
+    // initialize the sampler
+
     auto sparams = llama_sampler_chain_default_params();
     sparams.no_perf = false;
     llama_sampler * smpl = llama_sampler_chain_init(sparams);
 
     llama_sampler_chain_add(smpl, llama_sampler_init_greedy());
 
-    // tokenize the prompt
-
-    std::vector<llama_token> tokens_list;
-    int n_tokens = llama_tokenize(model, prompt.c_str(), prompt.size(), NULL, 0, true, true);
-    tokens_list.resize(-n_tokens);
-    if (llama_tokenize(model, prompt.c_str(), prompt.size(), tokens_list.data(), tokens_list.size(), true, true) < 0) {
-        fprintf(stderr, "%s: error: failed to tokenize the prompt\n", __func__);
-        return 1;
-    }
-
-    const int n_ctx    = llama_n_ctx(ctx);
-    const int n_kv_req = tokens_list.size() + (n_predict - tokens_list.size());
-
-    fprintf(stderr, "%s: n_predict = %d, n_ctx = %d, n_kv_req = %d\n", __func__, n_predict, n_ctx, n_kv_req);
-
-
-    // make sure the KV cache is big enough to hold all the prompt and generated tokens
-    if (n_kv_req > n_ctx) {
-        fprintf(stderr, "%s: error: n_kv_req > n_ctx, the required KV cache size is not big enough\n", __func__);
-        fprintf(stderr, "%s:        either reduce n_predict or increase n_ctx\n", __func__);
-        return 1;
-    }
-
     // print the prompt token-by-token
 
-    fprintf(stderr, "\n");
-
-    for (auto id : tokens_list) {
+    for (auto id : prompt_tokens) {
         char buf[128];
         int n = llama_token_to_piece(model, id, buf, sizeof(buf), 0, true);
         if (n < 0) {
@@ -95,34 +136,31 @@ int main(int argc, char ** argv) {
         printf("%s", s.c_str());
     }
 
-    // create a llama_batch with size 512
-    // we use this object to submit token data for decoding
+    // prepare a batch for the prompt
 
-    llama_batch batch = llama_batch_get_one(tokens_list.data(), tokens_list.size(), 0, 0);
-
-    // evaluate the initial prompt
-
-    if (llama_decode(ctx, batch) != 0) {
-        fprintf(stderr, "%s: llama_decode() failed\n", __func__);
-        return 1;
-    }
+    llama_batch batch = llama_batch_get_one(prompt_tokens.data(), prompt_tokens.size(), 0, 0);
 
     // main loop
 
-    int n_cur    = batch.n_tokens;
-    int n_decode = 0;
-
     const auto t_main_start = ggml_time_us();
+    int n_decode = 0;
+    llama_token new_token_id;
+
+    for (int n_pos = 0; n_pos + batch.n_tokens < n_prompt + n_predict; ) {
+        // evaluate the current batch with the transformer model
+        if (llama_decode(ctx, batch)) {
+            fprintf(stderr, "%s : failed to eval, return code %d\n", __func__, 1);
+            return 1;
+        }
+
+        n_pos += batch.n_tokens;
 
-    while (n_cur <= n_predict) {
         // sample the next token
-        llama_token new_token_id = llama_sampler_sample(smpl, ctx, -1);
         {
+            new_token_id = llama_sampler_sample(smpl, ctx, -1);
 
             // is it an end of generation?
-            if (llama_token_is_eog(model, new_token_id) || n_cur == n_predict) {
-                fprintf(stderr, "\n");
-
+            if (llama_token_is_eog(model, new_token_id)) {
                 break;
             }
 
@@ -136,22 +174,14 @@ int main(int argc, char ** argv) {
             printf("%s", s.c_str());
             fflush(stdout);
 
-            // prepare the next batch
-            batch = llama_batch_get_one(&new_token_id, 1, n_cur, 0);
+            // prepare the next batch with the sampled token
+            batch = llama_batch_get_one(&new_token_id, 1, n_pos, 0);
 
             n_decode += 1;
         }
-
-        n_cur += 1;
-
-        // evaluate the current batch with the transformer model
-        if (llama_decode(ctx, batch)) {
-            fprintf(stderr, "%s : failed to eval, return code %d\n", __func__, 1);
-            return 1;
-        }
     }
 
-    fprintf(stderr, "\n");
+    printf("\n");
 
     const auto t_main_end = ggml_time_us();