diff --git a/examples/CMakeLists.txt b/examples/CMakeLists.txt index 68ad89964..653abc73a 100644 --- a/examples/CMakeLists.txt +++ b/examples/CMakeLists.txt @@ -38,6 +38,7 @@ else() add_subdirectory(speculative) add_subdirectory(lookahead) add_subdirectory(lookup) + add_subdirectory(gguf) add_subdirectory(train-text-from-scratch) add_subdirectory(imatrix) if (LLAMA_BUILD_SERVER) diff --git a/gguf-py/examples/reader.py b/gguf-py/examples/reader.py new file mode 100644 index 000000000..62e0769da --- /dev/null +++ b/gguf-py/examples/reader.py @@ -0,0 +1,45 @@ +#!/usr/bin/env python3 +import sys +from pathlib import Path +from gguf.gguf_reader import GGUFReader + + +sys.path.insert(0, str(Path(__file__).parent.parent)) + + +def read_gguf_file(gguf_file_path): + """ + Reads and prints key-value pairs and tensor information from a GGUF file in an improved format. + + Parameters: + - gguf_file_path: Path to the GGUF file. + """ + + reader = GGUFReader(gguf_file_path) + + # List all key-value pairs in a columnized format + print("Key-Value Pairs:") + max_key_length = max(len(key) for key in reader.fields.keys()) + for key, field in reader.fields.items(): + value = field.parts[field.data[0]] + print(f"{key:{max_key_length}} : {value}") + print("----") + + # List all tensors + print("Tensors:") + tensor_info_format = "{:<30} | Shape: {:<15} | Size: {:<12} | Quantization: {}" + print(tensor_info_format.format("Tensor Name", "Shape", "Size", "Quantization")) + print("-" * 80) + for tensor in reader.tensors: + shape_str = "x".join(map(str, tensor.shape)) + size_str = str(tensor.n_elements) + quantization_str = tensor.tensor_type.name + print(tensor_info_format.format(tensor.name, shape_str, size_str, quantization_str)) + + +if __name__ == '__main__': + if len(sys.argv) < 2: + print("Usage: reader.py ") + sys.exit(1) + gguf_file_path = sys.argv[1] + read_gguf_file(gguf_file_path)