root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-24 10:24:35 +00:00
Code Issues Actions 5 Packages Projects Releases Wiki Activity

ggml, common, examples, tests : fixed type arguments in printf (#5528)

Browse Source
This commit is contained in:
Herman Semenov 2024-02-18 16:20:12 +00:00 committed by GitHub
parent fc0c8d286a
commit 5d3de51f97
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
9 changed files with 45 additions and 45 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
common/common.cpp
View File

@ -1741,7 +1741,7 @@ void dump_non_result_info_yaml(FILE * stream, const gpt_params & params, const l
fprintf(stream, "rope_freq_base: %f # default: 10000.0\n", params.rope_freq_base); fprintf(stream, "rope_freq_base: %f # default: 10000.0\n", params.rope_freq_base);
fprintf(stream, "rope_freq_scale: %f # default: 1.0\n", params.rope_freq_scale); fprintf(stream, "rope_freq_scale: %f # default: 1.0\n", params.rope_freq_scale);
fprintf(stream, "seed: %d # default: -1 (random seed)\n", params.seed); fprintf(stream, "seed: %u # default: -1 (random seed)\n", params.seed);
fprintf(stream, "simple_io: %s # default: false\n", params.simple_io ? "true" : "false"); fprintf(stream, "simple_io: %s # default: false\n", params.simple_io ? "true" : "false");
fprintf(stream, "cont_batching: %s # default: false\n", params.cont_batching ? "true" : "false"); fprintf(stream, "cont_batching: %s # default: false\n", params.cont_batching ? "true" : "false");
fprintf(stream, "temp: %f # default: 0.8\n", sparams.temp); fprintf(stream, "temp: %f # default: 0.8\n", sparams.temp);
@ -1750,7 +1750,7 @@ void dump_non_result_info_yaml(FILE * stream, const gpt_params & params, const l
dump_vector_float_yaml(stream, "tensor_split", tensor_split_vector); dump_vector_float_yaml(stream, "tensor_split", tensor_split_vector);
fprintf(stream, "tfs: %f # default: 1.0\n", sparams.tfs_z); fprintf(stream, "tfs: %f # default: 1.0\n", sparams.tfs_z);
fprintf(stream, "threads: %d # default: %d\n", params.n_threads, std::thread::hardware_concurrency()); fprintf(stream, "threads: %d # default: %u\n", params.n_threads, std::thread::hardware_concurrency());
fprintf(stream, "top_k: %d # default: 40\n", sparams.top_k); fprintf(stream, "top_k: %d # default: 40\n", sparams.top_k);
fprintf(stream, "top_p: %f # default: 0.95\n", sparams.top_p); fprintf(stream, "top_p: %f # default: 0.95\n", sparams.top_p);
fprintf(stream, "min_p: %f # default: 0.0\n", sparams.min_p); fprintf(stream, "min_p: %f # default: 0.0\n", sparams.min_p);

2
examples/batched-bench/batched-bench.cpp
View File

@ -159,7 +159,7 @@ int main(int argc, char ** argv) {
} }
LOG_TEE("\n"); LOG_TEE("\n");
LOG_TEE("%s: n_kv_max = %d, is_pp_shared = %d, n_gpu_layers = %d, mmq = %d, n_threads = %d, n_threads_batch = %d\n", __func__, n_kv_max, is_pp_shared, n_gpu_layers, mmq, ctx_params.n_threads, ctx_params.n_threads_batch); LOG_TEE("%s: n_kv_max = %d, is_pp_shared = %d, n_gpu_layers = %d, mmq = %d, n_threads = %u, n_threads_batch = %u\n", __func__, n_kv_max, is_pp_shared, n_gpu_layers, mmq, ctx_params.n_threads, ctx_params.n_threads_batch);
LOG_TEE("\n"); LOG_TEE("\n");
LOG_TEE("|%6s | %6s | %4s | %6s | %8s | %8s | %8s | %8s | %8s | %8s |\n", "PP", "TG", "B", "N_KV", "T_PP s", "S_PP t/s", "T_TG s", "S_TG t/s", "T s", "S t/s"); LOG_TEE("|%6s | %6s | %4s | %6s | %8s | %8s | %8s | %8s | %8s | %8s |\n", "PP", "TG", "B", "N_KV", "T_PP s", "S_PP t/s", "T_TG s", "S_TG t/s", "T s", "S t/s");

2
examples/batched/batched.cpp
View File

@ -92,7 +92,7 @@ int main(int argc, char ** argv) {
const int n_ctx = llama_n_ctx(ctx); const int n_ctx = llama_n_ctx(ctx);
LOG_TEE("\n%s: n_len = %d, n_ctx = %d, n_batch = %d, n_parallel = %d, n_kv_req = %d\n", __func__, n_len, n_ctx, ctx_params.n_batch, n_parallel, n_kv_req); LOG_TEE("\n%s: n_len = %d, n_ctx = %d, n_batch = %u, n_parallel = %d, n_kv_req = %d\n", __func__, n_len, n_ctx, ctx_params.n_batch, n_parallel, n_kv_req);
// make sure the KV cache is big enough to hold all the prompt and generated tokens // make sure the KV cache is big enough to hold all the prompt and generated tokens
if (n_kv_req > n_ctx) { if (n_kv_req > n_ctx) {

38
examples/convert-llama2c-to-ggml/convert-llama2c-to-ggml.cpp
View File

@ -325,14 +325,14 @@ struct train_params {
}; };
static void print_params(struct my_llama_hparams * params) { static void print_params(struct my_llama_hparams * params) {
printf("%s: n_vocab: %d\n", __func__, params->n_vocab); printf("%s: n_vocab: %u\n", __func__, params->n_vocab);
printf("%s: n_ctx: %d\n", __func__, params->n_ctx); printf("%s: n_ctx: %u\n", __func__, params->n_ctx);
printf("%s: n_embd: %d\n", __func__, params->n_embd); printf("%s: n_embd: %u\n", __func__, params->n_embd);
printf("%s: n_mult: %d\n", __func__, params->n_mult); printf("%s: n_mult: %u\n", __func__, params->n_mult);
printf("%s: n_head: %d\n", __func__, params->n_head); printf("%s: n_head: %u\n", __func__, params->n_head);
printf("%s: n_ff: %d\n", __func__, params->n_ff); printf("%s: n_ff: %u\n", __func__, params->n_ff);
printf("%s: n_layer: %d\n", __func__, params->n_layer); printf("%s: n_layer: %u\n", __func__, params->n_layer);
printf("%s: n_rot: %d\n", __func__, params->n_rot); printf("%s: n_rot: %u\n", __func__, params->n_rot);
} }
static void init_model(struct my_llama_model * model) { static void init_model(struct my_llama_model * model) {
@ -350,25 +350,25 @@ static void init_model(struct my_llama_model * model) {
model->train_tokens = 0; model->train_tokens = 0;
model->tok_embeddings = ggml_new_tensor_2d(ctx, GGML_TYPE_F32, n_embd, n_vocab); model->tok_embeddings = ggml_new_tensor_2d(ctx, GGML_TYPE_F32, n_embd, n_vocab);
printf("[%s:GG] Allocating [%d] x [%d] = [%d] float space for model->tok_embeddings\n",__func__,n_embd , n_vocab, n_embd * n_vocab); printf("[%s:GG] Allocating [%u] x [%u] = [%u] float space for model->tok_embeddings\n",__func__,n_embd , n_vocab, n_embd * n_vocab);
model->norm = ggml_new_tensor_1d(ctx, GGML_TYPE_F32, n_embd); model->norm = ggml_new_tensor_1d(ctx, GGML_TYPE_F32, n_embd);
printf("[%s:GG] Allocating [%d] float space for model->norm\n",__func__,n_embd); printf("[%s:GG] Allocating [%u] float space for model->norm\n",__func__,n_embd);
model->output = ggml_new_tensor_2d(ctx, GGML_TYPE_F32, n_embd, n_vocab); model->output = ggml_new_tensor_2d(ctx, GGML_TYPE_F32, n_embd, n_vocab);
printf("[%s:GG] Allocating [%d] x[%d] = [%d] float space for model->output\n",__func__,n_embd, n_vocab, n_embd * n_vocab); printf("[%s:GG] Allocating [%u] x[%u] = [%u] float space for model->output\n",__func__,n_embd, n_vocab, n_embd * n_vocab);
// printing the per-layer allocations here so we dont print in the for loop. // printing the per-layer allocations here so we dont print in the for loop.
printf("[%s:GG] Allocating [%d] x[%d] = [%d] float space for layer.wq for [%d] layers\n",__func__, n_embd, n_embd, n_embd * n_embd, n_layer); printf("[%s:GG] Allocating [%u] x[%u] = [%u] float space for layer.wq for [%u] layers\n",__func__, n_embd, n_embd, n_embd * n_embd, n_layer);
printf("[%s:GG] Allocating [%d] x[%d] = [%d] float space for layer.wk for [%d] layers\n",__func__, n_embd, n_embd, n_embd * n_embd, n_layer); printf("[%s:GG] Allocating [%u] x[%u] = [%u] float space for layer.wk for [%u] layers\n",__func__, n_embd, n_embd, n_embd * n_embd, n_layer);
printf("[%s:GG] Allocating [%d] x[%d] = [%d] float space for layer.wv for [%d] layers\n",__func__, n_embd, n_embd, n_embd * n_embd, n_layer); printf("[%s:GG] Allocating [%u] x[%u] = [%u] float space for layer.wv for [%u] layers\n",__func__, n_embd, n_embd, n_embd * n_embd, n_layer);
printf("[%s:GG] Allocating [%d] x[%d] = [%d] float space for layer.wo for [%d] layers\n",__func__, n_embd, n_embd, n_embd * n_embd, n_layer); printf("[%s:GG] Allocating [%u] x[%u] = [%u] float space for layer.wo for [%u] layers\n",__func__, n_embd, n_embd, n_embd * n_embd, n_layer);
printf("[%s:GG] Allocating [%d] float space for layer.ffn_norm for [%d] layers\n",__func__,n_embd, n_layer); printf("[%s:GG] Allocating [%u] float space for layer.ffn_norm for [%u] layers\n",__func__,n_embd, n_layer);
printf("[%s:GG] Allocating [%d] x[%d] = [%d] float space for layer.w1 for [%d] layers\n",__func__, n_ff, n_embd, n_embd * n_ff, n_layer); printf("[%s:GG] Allocating [%u] x[%u] = [%u] float space for layer.w1 for [%u] layers\n",__func__, n_ff, n_embd, n_embd * n_ff, n_layer);
printf("[%s:GG] Allocating [%d] x[%d] = [%d] float space for layer.w2 for [%d] layers\n",__func__, n_embd, n_ff, n_ff * n_embd, n_layer); printf("[%s:GG] Allocating [%u] x[%u] = [%u] float space for layer.w2 for [%u] layers\n",__func__, n_embd, n_ff, n_ff * n_embd, n_layer);
printf("[%s:GG] Allocating [%d] x[%d] = [%d] float space for layer.w3 for [%d] layers\n",__func__, n_ff, n_embd, n_embd * n_ff, n_layer); printf("[%s:GG] Allocating [%u] x[%u] = [%u] float space for layer.w3 for [%u] layers\n",__func__, n_ff, n_embd, n_embd * n_ff, n_layer);
ggml_set_name(model->tok_embeddings, "tok_embeddings.weight"); ggml_set_name(model->tok_embeddings, "tok_embeddings.weight");
ggml_set_name(model->norm, "norm.weight"); ggml_set_name(model->norm, "norm.weight");

2
examples/perplexity/perplexity.cpp
View File

@ -1623,7 +1623,7 @@ static void kl_divergence(llama_context * ctx, const gpt_params & params) {
uint32_t n_ctx; uint32_t n_ctx;
in.read((char *)&n_ctx, sizeof(n_ctx)); in.read((char *)&n_ctx, sizeof(n_ctx));
if (n_ctx > llama_n_ctx(ctx)) { if (n_ctx > llama_n_ctx(ctx)) {
fprintf(stderr, "%s: %s has been computed with %d, while the current context is %d. Increase it with -c and retry\n", fprintf(stderr, "%s: %s has been computed with %u, while the current context is %d. Increase it with -c and retry\n",
__func__, params.logits_file.c_str(), n_ctx, params.n_ctx); __func__, params.logits_file.c_str(), n_ctx, params.n_ctx);
} }

14
examples/train-text-from-scratch/train-text-from-scratch.cpp
View File

@ -111,13 +111,13 @@ static const char * LLM_TENSOR_FFN_DOWN = "blk.%d.ffn_down";
static const char * LLM_TENSOR_FFN_UP = "blk.%d.ffn_up"; static const char * LLM_TENSOR_FFN_UP = "blk.%d.ffn_up";
static void print_params(struct my_llama_hparams * params) { static void print_params(struct my_llama_hparams * params) {
printf("%s: n_vocab: %d\n", __func__, params->n_vocab); printf("%s: n_vocab: %u\n", __func__, params->n_vocab);
printf("%s: n_ctx: %d\n", __func__, params->n_ctx); printf("%s: n_ctx: %u\n", __func__, params->n_ctx);
printf("%s: n_embd: %d\n", __func__, params->n_embd); printf("%s: n_embd: %u\n", __func__, params->n_embd);
printf("%s: n_head: %d\n", __func__, params->n_head); printf("%s: n_head: %u\n", __func__, params->n_head);
printf("%s: n_ff: %d\n", __func__, params->n_ff); printf("%s: n_ff: %u\n", __func__, params->n_ff);
printf("%s: n_layer: %d\n", __func__, params->n_layer); printf("%s: n_layer: %u\n", __func__, params->n_layer);
printf("%s: n_rot: %d\n", __func__, params->n_rot); printf("%s: n_rot: %u\n", __func__, params->n_rot);
} }
static void set_param_model(struct my_llama_model * model) { static void set_param_model(struct my_llama_model * model) {

4
ggml.c
View File

@ -17909,7 +17909,7 @@ struct ggml_cgraph * ggml_graph_import(const char * fname, struct ggml_context *
ptr += ggml_nbytes(tensor); ptr += ggml_nbytes(tensor);
fprintf(stderr, "%s: loaded leaf %d: '%16s', %9zu bytes\n", __func__, i, tensor->name, ggml_nbytes(tensor)); fprintf(stderr, "%s: loaded leaf %u: '%16s', %9zu bytes\n", __func__, i, tensor->name, ggml_nbytes(tensor));
} }
} }
@ -18012,7 +18012,7 @@ struct ggml_cgraph * ggml_graph_import(const char * fname, struct ggml_context *
result->nodes[i] = tensor; result->nodes[i] = tensor;
fprintf(stderr, "%s: loaded node %d: '%16s', %9zu bytes\n", __func__, i, tensor->name, ggml_nbytes(tensor)); fprintf(stderr, "%s: loaded node %u: '%16s', %9zu bytes\n", __func__, i, tensor->name, ggml_nbytes(tensor));
} }
} }
} }

20
tests/test-grammar-parser.cpp
View File

@ -38,8 +38,8 @@ term ::= [0-9]+)""";
// pretty print error message before asserting // pretty print error message before asserting
if (expected_pair.first != key || expected_pair.second != value) if (expected_pair.first != key || expected_pair.second != value)
{ {
fprintf(stderr, "expected_pair: %s, %d\n", expected_pair.first.c_str(), expected_pair.second); fprintf(stderr, "expected_pair: %s, %u\n", expected_pair.first.c_str(), expected_pair.second);
fprintf(stderr, "actual_pair: %s, %d\n", key.c_str(), value); fprintf(stderr, "actual_pair: %s, %u\n", key.c_str(), value);
fprintf(stderr, "expected_pair != actual_pair\n"); fprintf(stderr, "expected_pair != actual_pair\n");
} }
@ -96,9 +96,9 @@ term ::= [0-9]+)""";
// pretty print error message before asserting // pretty print error message before asserting
if (expected_element.type != element.type || expected_element.value != element.value) if (expected_element.type != element.type || expected_element.value != element.value)
{ {
fprintf(stderr, "index: %d\n", index); fprintf(stderr, "index: %u\n", index);
fprintf(stderr, "expected_element: %d, %d\n", expected_element.type, expected_element.value); fprintf(stderr, "expected_element: %d, %u\n", expected_element.type, expected_element.value);
fprintf(stderr, "actual_element: %d, %d\n", element.type, element.value); fprintf(stderr, "actual_element: %d, %u\n", element.type, element.value);
fprintf(stderr, "expected_element != actual_element\n"); fprintf(stderr, "expected_element != actual_element\n");
} }
@ -144,8 +144,8 @@ term ::= [0-9]+)""";
// pretty print error message before asserting // pretty print error message before asserting
if (expected_pair.first != key || expected_pair.second != value) if (expected_pair.first != key || expected_pair.second != value)
{ {
fprintf(stderr, "expected_pair: %s, %d\n", expected_pair.first.c_str(), expected_pair.second); fprintf(stderr, "expected_pair: %s, %u\n", expected_pair.first.c_str(), expected_pair.second);
fprintf(stderr, "actual_pair: %s, %d\n", key.c_str(), value); fprintf(stderr, "actual_pair: %s, %u\n", key.c_str(), value);
fprintf(stderr, "expected_pair != actual_pair\n"); fprintf(stderr, "expected_pair != actual_pair\n");
} }
@ -235,9 +235,9 @@ term ::= [0-9]+)""";
// pretty print error message before asserting // pretty print error message before asserting
if (expected_element.type != element.type || expected_element.value != element.value) if (expected_element.type != element.type || expected_element.value != element.value)
{ {
fprintf(stderr, "index: %d\n", index); fprintf(stderr, "index: %u\n", index);
fprintf(stderr, "expected_element: %d, %d\n", expected_element.type, expected_element.value); fprintf(stderr, "expected_element: %d, %u\n", expected_element.type, expected_element.value);
fprintf(stderr, "actual_element: %d, %d\n", element.type, element.value); fprintf(stderr, "actual_element: %d, %u\n", element.type, element.value);
fprintf(stderr, "expected_element != actual_element\n"); fprintf(stderr, "expected_element != actual_element\n");
} }

4
tests/test-llama-grammar.cpp
View File

@ -180,8 +180,8 @@ int main()
if (expected_element.type != element->type || expected_element.value != element->value) if (expected_element.type != element->type || expected_element.value != element->value)
{ {
fprintf(stderr, "index: %d\n", index); fprintf(stderr, "index: %d\n", index);
fprintf(stderr, "expected_element: %d, %d\n", expected_element.type, expected_element.value); fprintf(stderr, "expected_element: %d, %u\n", expected_element.type, expected_element.value);
fprintf(stderr, "actual_element: %d, %d\n", element->type, element->value); fprintf(stderr, "actual_element: %d, %u\n", element->type, element->value);
fprintf(stderr, "expected_element != actual_element\n"); fprintf(stderr, "expected_element != actual_element\n");
} }