vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

fix: remove ollama patches

Browse source
This commit is contained in:
Joan Martinez 2024-06-05 09:15:36 +02:00
parent 3b44f8f658
commit 05659d3c7b
1 changed files with 20 additions and 20 deletions
Download patch file Download diff file Expand all files Collapse all files

40
llama.cpp
View file

@ -4653,8 +4653,16 @@ static void llm_load_vocab(
// for now, only BPE models have pre-tokenizers // for now, only BPE models have pre-tokenizers
if (vocab.type == LLAMA_VOCAB_TYPE_BPE) { if (vocab.type == LLAMA_VOCAB_TYPE_BPE) {
if ( if (tokenizer_pre.empty()) {
tokenizer_pre == "default") { LLAMA_LOG_WARN("%s: missing pre-tokenizer type, using: 'default'\n", __func__);
LLAMA_LOG_WARN("%s: \n", __func__);
LLAMA_LOG_WARN("%s: ************************************ \n", __func__);
LLAMA_LOG_WARN("%s: GENERATION QUALITY WILL BE DEGRADED! \n", __func__);
LLAMA_LOG_WARN("%s: CONSIDER REGENERATING THE MODEL \n", __func__);
LLAMA_LOG_WARN("%s: ************************************ \n", __func__);
LLAMA_LOG_WARN("%s: \n", __func__);
vocab.type_pre = LLAMA_VOCAB_PRE_TYPE_DEFAULT;
} else if (tokenizer_pre == "default") {
vocab.type_pre = LLAMA_VOCAB_PRE_TYPE_DEFAULT; vocab.type_pre = LLAMA_VOCAB_PRE_TYPE_DEFAULT;
} else if ( } else if (
tokenizer_pre == "llama3" || tokenizer_pre == "llama3" ||
@ -4706,8 +4714,7 @@ static void llm_load_vocab(
tokenizer_pre == "smaug-bpe") { tokenizer_pre == "smaug-bpe") {
vocab.type_pre = LLAMA_VOCAB_PRE_TYPE_SMAUG; vocab.type_pre = LLAMA_VOCAB_PRE_TYPE_SMAUG;
} else { } else {
LLAMA_LOG_WARN("%s: missing or unrecognized pre-tokenizer type, using: 'default'\n", __func__); throw std::runtime_error(format("unknown pre-tokenizer type: '%s'", tokenizer_pre.c_str()));
vocab.type_pre = LLAMA_VOCAB_PRE_TYPE_DEFAULT;
} }
} else { } else {
vocab.type_pre = LLAMA_VOCAB_PRE_TYPE_DEFAULT; vocab.type_pre = LLAMA_VOCAB_PRE_TYPE_DEFAULT;
@ -6623,7 +6630,7 @@ static int llama_model_load(const std::string & fname, llama_model & model, llam
} }
} catch (const std::exception & err) { } catch (const std::exception & err) {
LLAMA_LOG_ERROR("%s: error loading model: %s\n", __func__, err.what()); LLAMA_LOG_ERROR("%s: error loading model: %s\n", __func__, err.what());
throw; return -1;
} }
return 0; return 0;
@ -16246,23 +16253,16 @@ struct llama_model * llama_load_model_from_file(
} }
model->rpc_servers.push_back(servers); model->rpc_servers.push_back(servers);
} }
int status = llama_model_load(path_model, *model, params);
try { GGML_ASSERT(status <= 0);
int status = llama_model_load(path_model, *model, params); if (status < 0) {
GGML_ASSERT(status <= 0); if (status == -1) {
if (status < 0) { LLAMA_LOG_ERROR("%s: failed to load model\n", __func__);
if (status == -1) { } else if (status == -2) {
LLAMA_LOG_ERROR("%s: failed to load model\n", __func__); LLAMA_LOG_INFO("%s: cancelled model load\n", __func__);
} else if (status == -2) {
LLAMA_LOG_INFO("%s: cancelled model load\n", __func__);
}
delete model;
return nullptr;
} }
} catch (...) {
LLAMA_LOG_ERROR("%s: exception loading model\n", __func__);
delete model; delete model;
throw; return nullptr;
} }
return model; return model;