data: publish complete Calculet NPU research archive

source/repositories-complete/llama.cpp-exact/llama.cpp/gguf-py/examples/reader.py

@@ -0,0 +1,49 @@
+#!/usr/bin/env python3
+import logging
+import sys
+from pathlib import Path
+
+logger = logging.getLogger("reader")
+
+# Necessary to load the local gguf package
+sys.path.insert(0, str(Path(__file__).parent.parent))
+
+from gguf.gguf_reader import GGUFReader
+
+
+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:") # noqa: NP100
+    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}") # noqa: NP100
+    print("----") # noqa: NP100
+
+    # List all tensors
+    print("Tensors:") # noqa: NP100
+    tensor_info_format = "{:<30} | Shape: {:<15} | Size: {:<12} | Quantization: {}"
+    print(tensor_info_format.format("Tensor Name", "Shape", "Size", "Quantization")) # noqa: NP100
+    print("-" * 80) # noqa: NP100
+    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)) # noqa: NP100
+
+
+if __name__ == '__main__':
+    if len(sys.argv) < 2:
+        logger.info("Usage: reader.py <path_to_gguf_file>")
+        sys.exit(1)
+    gguf_file_path = sys.argv[1]
+    read_gguf_file(gguf_file_path)

source/repositories-complete/llama.cpp-exact/llama.cpp/gguf-py/examples/writer.py

@@ -0,0 +1,39 @@
+#!/usr/bin/env python3
+import sys
+from pathlib import Path
+
+import numpy as np
+
+# Necessary to load the local gguf package
+sys.path.insert(0, str(Path(__file__).parent.parent))
+
+from gguf import GGUFWriter  # noqa: E402
+
+
+# Example usage:
+def writer_example() -> None:
+    # Example usage with a file
+    gguf_writer = GGUFWriter("example.gguf", "llama")
+
+    gguf_writer.add_block_count(12)
+    gguf_writer.add_uint32("answer", 42)  # Write a 32-bit integer
+    gguf_writer.add_float32("answer_in_float", 42.0)  # Write a 32-bit float
+    gguf_writer.add_custom_alignment(64)
+
+    tensor1 = np.ones((32,), dtype=np.float32) * 100.0
+    tensor2 = np.ones((64,), dtype=np.float32) * 101.0
+    tensor3 = np.ones((96,), dtype=np.float32) * 102.0
+
+    gguf_writer.add_tensor("tensor1", tensor1)
+    gguf_writer.add_tensor("tensor2", tensor2)
+    gguf_writer.add_tensor("tensor3", tensor3)
+
+    gguf_writer.write_header_to_file()
+    gguf_writer.write_kv_data_to_file()
+    gguf_writer.write_tensors_to_file()
+
+    gguf_writer.close()
+
+
+if __name__ == '__main__':
+    writer_example()

source/repositories-complete/llama.cpp-exact/llama.cpp/gguf-py/examples/writer_cal_test.py

@@ -0,0 +1,84 @@
+#!/usr/bin/env python3
+import sys
+from pathlib import Path
+import json
+
+import numpy as np
+
+# Necessary to load the local gguf package
+sys.path.insert(0, str(Path(__file__).parent.parent))
+
+from gguf import GGUFWriter  # noqa: E402
+
+# Example usage:
+def writer_example() -> None:
+    # Example usage with a file
+    gguf_writer = GGUFWriter("cal_test.gguf", "cal_test")
+
+    tokens = [
+        "a",
+        "b",
+        "c",
+        "d",
+        "e",
+        "f",
+        "g",
+        "h",
+        "i",
+        "j",
+        "k",
+        "l",
+        "m",
+        "n",
+        "o",
+        "p",
+        "q",
+        "r",
+        "s",
+        "t",
+        "u",
+        "v",
+        "w",
+        "x",
+        "y",
+        "z",
+        "\u2581", #空格
+        "\n"
+    ]
+
+    vocab_size = len(tokens)
+    embedding_len = 256
+    ctx_len = 2048
+
+    gguf_writer.add_block_count(1)
+    gguf_writer.add_context_length(ctx_len)
+    gguf_writer.add_embedding_length(embedding_len)
+    gguf_writer.add_vocab_size(vocab_size)
+    
+    gguf_writer.add_head_count_kv(8)
+    gguf_writer.add_head_count(16)
+    gguf_writer.add_key_length(16)
+    gguf_writer.add_value_length(16)
+    gguf_writer.add_tokenizer_model("llama")
+    # 添加 tokenizer 列表到元数据
+    gguf_writer.add_token_list(tokens)
+    output_weight = np.ones((vocab_size, embedding_len), dtype=np.float32)
+
+    print(f"token 数量：{vocab_size}")
+
+    # token_embd_weight = np.ones((128000,embedding_len), dtype=np.float32)
+    # print(f"创建的张量形状: {output_weight.shape}")
+    # gguf_writer.add_tensor("output.weight", output_weight)
+
+    # attn_norm = np.ones((1))
+    # gguf_writer.add_tensor("blk.0.attn_norm.weight", attn_norm)
+
+    gguf_writer.write_header_to_file()
+    gguf_writer.write_kv_data_to_file()
+    gguf_writer.write_tensors_to_file()
+
+    gguf_writer.close()
+
+
+if __name__ == '__main__':
+    writer_example()