Mousius commented on a change in pull request #8922: URL: https://github.com/apache/tvm/pull/8922#discussion_r723298267
########## File path: apps/microtvm/ethosu/convert_image.py ########## @@ -0,0 +1,88 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. + +import os +import pathlib +import re +import sys +from PIL import Image +import numpy as np + + +def create_header_file(name, section, npy_data, output_path): + """ + This function generates a header file containing the data from the numpy array provided. + """ + file_path = pathlib.Path(f"{output_path}/" + name).resolve() + + # Create header file with npy_data as a C array + raw_path = file_path.with_suffix(".h").resolve() + with open(raw_path, "w") as header_file: + header_file.write("#include <tvmgen_default.h>\n") + for tensor_name in npy_data.keys(): + sanitized_tensor_name = re.sub(r"\W+", "_", tensor_name) + header_file.write( + f"const size_t {sanitized_tensor_name}_len = {npy_data[tensor_name].size};\n" + ) + + # Convert numpy data type to C data type + if npy_data[tensor_name].dtype == np.uint8: + c_type = "uint8_t" + elif npy_data[tensor_name].dtype == np.int8: + c_type = "int8_t" + else: + raise RuntimeError(f"Data type {str(npy_data[tensor_name].dtype)} not supported") + + header_file.write( + f'{c_type} {sanitized_tensor_name}[] __attribute__((section("{section}"), aligned(16))) = "' + ) + + data_hexstr = npy_data[tensor_name].tobytes().hex() + for i in range(0, len(data_hexstr), 2): + header_file.write(f"\\x{data_hexstr[i:i+2]}") + header_file.write('";\n\n') Review comment: We're passing in dicts like `{"input": img_data.astype(np.uint8)}` and `{"output": np.zeros([1001], np.uint8)}`, which leads me to believe we can generate far less of this if we pass them separately and don't treat it as a `dict`: ```suggestion def create_header_file(name, section, tensor_name, tensor_data, output_path): """ This function generates a header file containing the data from the numpy array provided. """ file_path = pathlib.Path(f"{output_path}/" + name).resolve() # Create header file with npy_data as a C array raw_path = file_path.with_suffix(".h").resolve() with open(raw_path, "w") as header_file: header_file.write( "#include <tvmgen_default.h>\n", f"const size_t {tensor_name}_len = {npy_data[tensor_data].size};\n" f'uint8_t {tensor_name}[] __attribute__((section("{section}"), aligned(16))) = "' ) for i in range(0, len(data_hexstr), 2): header_file.write(f"\\x{data_hexstr[i:i+2]}") header_file.write('";\n\n') ``` This would parallel a developer who deploys this model, who would only process the data required as an input. ########## File path: apps/microtvm/ethosu/convert_image.py ########## @@ -0,0 +1,88 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. + +import os +import pathlib +import re +import sys +from PIL import Image +import numpy as np + + +def create_header_file(name, section, npy_data, output_path): + """ + This function generates a header file containing the data from the numpy array provided. + """ + file_path = pathlib.Path(f"{output_path}/" + name).resolve() + + # Create header file with npy_data as a C array + raw_path = file_path.with_suffix(".h").resolve() + with open(raw_path, "w") as header_file: + header_file.write("#include <tvmgen_default.h>\n") + for tensor_name in npy_data.keys(): + sanitized_tensor_name = re.sub(r"\W+", "_", tensor_name) + header_file.write( + f"const size_t {sanitized_tensor_name}_len = {npy_data[tensor_name].size};\n" + ) + + # Convert numpy data type to C data type + if npy_data[tensor_name].dtype == np.uint8: + c_type = "uint8_t" + elif npy_data[tensor_name].dtype == np.int8: + c_type = "int8_t" + else: + raise RuntimeError(f"Data type {str(npy_data[tensor_name].dtype)} not supported") + + header_file.write( + f'{c_type} {sanitized_tensor_name}[] __attribute__((section("{section}"), aligned(16))) = "' + ) + + data_hexstr = npy_data[tensor_name].tobytes().hex() + for i in range(0, len(data_hexstr), 2): + header_file.write(f"\\x{data_hexstr[i:i+2]}") + header_file.write('";\n\n') + + +def create_headers(image_name): + """ + This function generates C header files for the input and output arrays required to run inferences + """ + img_path = os.path.join("./", f"{image_name}") + + # Resize image to 224x224 + resized_image = Image.open(img_path).resize((224, 224)) + img_data = np.asarray(resized_image).astype("float32") + + # Convert input to NCHW + img_data = np.transpose(img_data, (2, 0, 1)) + + # Create input header file + input_data = {"input": img_data.astype(np.uint8)} + create_header_file("inputs", "ethosu_scratch", input_data, "./include") + + # Create output header file + output_data = {"output": np.zeros([1001], np.uint8)} + create_header_file( + "outputs", + "output_data_sec", + output_data, + "./include", + ) Review comment: Following from the above. ```suggestion input_data = img_data.astype(np.uint8) create_header_file("inputs", "ethosu_scratch", "input", input_data, "./include") # Create output header file output_data = np.zeros([1001], np.uint8) create_header_file( "outputs", "output_data_sec", "output", output_data, "./include", ) ``` -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
