Finetune LORA

Makefile

@@ -1,5 +1,5 @@
 # Define the default target now so that it is always the first target
-BUILD_TARGETS = main quantize quantize-stats perplexity embedding vdot train-text-from-scratch convert-llama2c-to-ggml simple save-load-state server embd-input-test gguf llama-bench baby-llama beam-search speculative tests/test-c.o
+BUILD_TARGETS = main quantize quantize-stats perplexity embedding vdot train-text-from-scratch convert-llama2c-to-ggml simple save-load-state server embd-input-test gguf llama-bench baby-llama beam-search speculative finetune export-lora tests/test-c.o
 
 # Binaries only useful for tests
 TEST_TARGETS = tests/test-llama-grammar tests/test-grammar-parser tests/test-double-float tests/test-grad0 tests/test-opt tests/test-quantize-fns tests/test-quantize-perf tests/test-sampling tests/test-tokenizer-0-llama tests/test-tokenizer-0-falcon tests/test-tokenizer-1-llama
@@ -493,6 +493,9 @@ console.o: common/console.cpp common/console.h
 grammar-parser.o: common/grammar-parser.cpp common/grammar-parser.h
 	$(CXX) $(CXXFLAGS) -c $< -o $@
 
+train.o: common/train.cpp common/train.h
+	$(CXX) $(CXXFLAGS) -c $< -o $@
+
 libllama.so: llama.o ggml.o $(OBJS)
 	$(CXX) $(CXXFLAGS) -shared -fPIC -o $@ $^ $(LDFLAGS)
 
@@ -540,7 +543,7 @@ embd-input-test: $(LIB_PRE)embdinput$(DSO_EXT) examples/embd-input/embd-input-te
 gguf: examples/gguf/gguf.cpp ggml.o llama.o $(OBJS)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
 
-train-text-from-scratch: examples/train-text-from-scratch/train-text-from-scratch.cpp ggml.o llama.o common.o $(OBJS)
+train-text-from-scratch: examples/train-text-from-scratch/train-text-from-scratch.cpp ggml.o llama.o common.o train.o $(OBJS)
 	$(CXX) $(TTFS_CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
 
 convert-llama2c-to-ggml: examples/convert-llama2c-to-ggml/convert-llama2c-to-ggml.cpp ggml.o llama.o $(OBJS)
@@ -549,12 +552,18 @@ convert-llama2c-to-ggml: examples/convert-llama2c-to-ggml/convert-llama2c-to-ggm
 llama-bench: examples/llama-bench/llama-bench.cpp build-info.h ggml.o llama.o common.o $(OBJS)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
 
-baby-llama: examples/baby-llama/baby-llama.cpp ggml.o llama.o common.o $(OBJS)
+baby-llama: examples/baby-llama/baby-llama.cpp ggml.o llama.o common.o train.o $(OBJS)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
 
 beam-search: examples/beam-search/beam-search.cpp build-info.h ggml.o llama.o common.o $(OBJS)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
 
+finetune: examples/finetune/finetune.cpp build-info.h ggml.o llama.o common.o train.o $(OBJS)
+	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
+
+export-lora: examples/export-lora/export-lora.cpp build-info.h ggml.o common.o $(OBJS)
+	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
+
 speculative: examples/speculative/speculative.cpp build-info.h ggml.o llama.o common.o grammar-parser.o $(OBJS)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
 

common/CMakeLists.txt

@@ -9,6 +9,8 @@ add_library(${TARGET} OBJECT
     console.cpp
     grammar-parser.h
     grammar-parser.cpp
+    train.h
+    train.cpp
     )
 
 if (BUILD_SHARED_LIBS)

common/common.cpp

@@ -78,7 +78,7 @@ int32_t get_num_physical_cores() {
     return n_threads > 0 ? (n_threads <= 4 ? n_threads : n_threads / 2) : 4;
 }
 
-static void process_escapes(std::string& input) {
+void process_escapes(std::string& input) {
     std::size_t input_len = input.length();
     std::size_t output_idx = 0;
 
@@ -340,7 +340,19 @@ bool gpt_params_parse(int argc, char ** argv, gpt_params & params) {
                 invalid_param = true;
                 break;
             }
-            params.lora_adapter = argv[i];
+            params.lora_adapter.push_back({argv[i], 1.0f});
+            params.use_mmap = false;
+        } else if (arg == "--lora-scaled") {
+            if (++i >= argc) {
+                invalid_param = true;
+                break;
+            }
+            const char * lora_adapter = argv[i];
+            if (++i >= argc) {
+                invalid_param = true;
+                break;
+            }
+            params.lora_adapter.push_back({lora_adapter, std::stof(argv[i])});
             params.use_mmap = false;
         } else if (arg == "--lora-base") {
             if (++i >= argc) {
@@ -689,6 +701,7 @@ void gpt_print_usage(int /*argc*/, char ** argv, const gpt_params & params) {
     printf("  --verbose-prompt      print prompt before generation\n");
     fprintf(stderr, "  --simple-io           use basic IO for better compatibility in subprocesses and limited consoles\n");
     printf("  --lora FNAME          apply LoRA adapter (implies --no-mmap)\n");
+    printf("  --lora-scaled FNAME S apply LoRA adapter with user defined scaling S (implies --no-mmap)\n");
     printf("  --lora-base FNAME     optional model to use as a base for the layers modified by the LoRA adapter\n");
     printf("  -m FNAME, --model FNAME\n");
     printf("                        model path (default: %s)\n", params.model.c_str());
@@ -762,10 +775,15 @@ std::tuple<struct llama_model *, struct llama_context *> llama_init_from_gpt_par
         return std::make_tuple(nullptr, nullptr);
     }
 
-    if (!params.lora_adapter.empty()) {
+    for (unsigned int i = 0; i < params.lora_adapter.size(); ++i) {
+        const std::string& lora_adapter = std::get<0>(params.lora_adapter[i]);
+        float lora_scale = std::get<1>(params.lora_adapter[i]);
         int err = llama_model_apply_lora_from_file(model,
-                                             params.lora_adapter.c_str(),
-                                             params.lora_base.empty() ? NULL : params.lora_base.c_str(),
+                                             lora_adapter.c_str(),
+                                             lora_scale,
+                                             ((i > 0) || params.lora_base.empty())
+                                                ? NULL
+                                                : params.lora_base.c_str(),
                                              params.n_threads);
         if (err != 0) {
             fprintf(stderr, "%s: error: failed to apply lora adapter\n", __func__);
@@ -1211,7 +1229,20 @@ void dump_non_result_info_yaml(FILE * stream, const gpt_params & params, const l
         fprintf(stream, "  %d: %f", lb.first, lb.second);
     }
 
-    fprintf(stream, "lora: %s\n", params.lora_adapter.c_str());
+    fprintf(stream, "lora:\n");
+    for (std::tuple<std::string, float> la : params.lora_adapter) {
+        if (std::get<1>(la) != 1.0f) {
+            continue;
+        }
+        fprintf(stream, "  - %s\n", std::get<0>(la).c_str());
+    }
+    fprintf(stream, "lora_scaled:\n");
+    for (std::tuple<std::string, float> la : params.lora_adapter) {
+        if (std::get<1>(la) == 1.0f) {
+            continue;
+        }
+        fprintf(stream, "  - %s: %f\n", std::get<0>(la).c_str(), std::get<1>(la));
+    }
     fprintf(stream, "lora_base: %s\n", params.lora_base.c_str());
     fprintf(stream, "low_vram: %s # default: false\n", params.low_vram ? "true" : "false");
     fprintf(stream, "main_gpu: %d # default: 0\n", params.main_gpu);

common/common.h

@@ -84,8 +84,8 @@ struct gpt_params {
     std::vector<std::string> antiprompt; // string upon seeing which more user input is prompted
     std::string logdir            = "";  // directory in which to save YAML log files
 
-    std::string lora_adapter = "";  // lora adapter path
-    std::string lora_base    = "";  // base model path for the lora adapter
+    std::vector<std::tuple<std::string, float>> lora_adapter; // lora adapter path with user defined scale
+    std::string lora_base  = "";                              // base model path for the lora adapter
 
     int  ppl_stride        = 0;     // stride for perplexity calculations. If left at 0, the pre-existing approach will be used.
     int  ppl_output_type   = 0;     // = 0 -> ppl output is as usual, = 1 -> ppl output is num_tokens, ppl, one per line
@@ -127,6 +127,8 @@ void gpt_print_usage(int argc, char ** argv, const gpt_params & params);
 
 std::string gpt_random_prompt(std::mt19937 & rng);
 
+void process_escapes(std::string& input);
+
 //
 // Model utils
 //