Skip to content

Commit

Permalink
Fixed error.
Browse files Browse the repository at this point in the history
  • Loading branch information
popovaan committed Dec 20, 2024
1 parent c7d54dd commit 51cb0a8
Showing 1 changed file with 6 additions and 6 deletions.
12 changes: 6 additions & 6 deletions samples/cpp/speculative_decoding_lm/speculative_decoding_lm.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -6,9 +6,9 @@
#include "openvino/genai/llm_pipeline.hpp"

int main(int argc, char* argv[]) try {
if (4 != argc) {
throw std::runtime_error(std::string{"Usage: "} + argv[0] + " <MODEL_DIR> <DRAFT_MODEL_DIR> '<PROMPT>'");
}
// if (4 != argc) {
// throw std::runtime_error(std::string{"Usage: "} + argv[0] + " <MODEL_DIR> <DRAFT_MODEL_DIR> '<PROMPT>'");
// }

ov::genai::GenerationConfig config;
config.max_new_tokens = 100;
Expand All @@ -18,9 +18,9 @@ int main(int argc, char* argv[]) try {
// add parameter to enable speculative decoding to generate candidates by draft_model while candidate probability is higher than `assistant_confidence_threshold`
// config.assistant_confidence_threshold = 0.4;

std::string main_model_path = argv[1];
std::string draft_model_path = argv[2];
std::string prompt = argv[3];
std::string main_model_path = "/home/panas/test_models/spec_dec/Llama-2-7b-chat-hf/";
std::string draft_model_path = "/home/panas/test_models/spec_dec/tiny-llama-1.1b-chat/";
std::string prompt = "What is openvino?";

// User can run main and draft model on different devices.
// Please, set device for main model in `LLMPipeline` constructor and in in `ov::genai::draft_model` for draft.
Expand Down

0 comments on commit 51cb0a8

Please sign in to comment.