def convert_from_keras_model(model, backend, output_dir='my-hls-test', project_name='myproject',
fpga_part='xcku115-flvb2104-2-i', clock_period=5, hls_config={}):
config = create_config(output_dir=output_dir,
project_name=project_name, backend=backend.name, fpga_part=fpga_part, clock_period=clock_period)
config['KerasModel'] = model
model_config = hls_config.get('Model', None)
if model_config is not None:
if not all(k in model_config for k in ('Precision', 'ReuseFactor')):
raise Exception('Precision and ReuseFactor must be provided in the hls_config')
else:
model_config = {}
model_config['Precision'] = backend.get_pstring(16,6)
model_config['ReuseFactor'] = '1'
config['HLSConfig']['Model'] = model_config
if 'LayerName' in hls_config:
config['HLSConfig']['LayerName'] = hls_config['LayerName']
if 'LayerType' in hls_config:
config['HLSConfig']['LayerType'] = hls_config['LayerType']
if 'Optimizers' in hls_config:
config['HLSConfig']['Optimizers'] = hls_config['Optimizers']
return keras_to_hls(config)
def convert_from_keras_model(model,
output_dir='my-hls-test',
project_name='myproject',
input_data_tb=None,
output_data_tb=None,
backend='Vivado',
hls_config={},
**kwargs):
"""Convert to hls4ml model based on the provided configuration.
Args:
model: Keras model to convert
output_dir (str, optional): Output directory of the generated HLS
project. Defaults to 'my-hls-test'.
project_name (str, optional): Name of the HLS project.
Defaults to 'myproject'.
input_data_tb (str, optional): String representing the path of input data in .npy or .dat format that will be
used during csim and cosim.
output_data_tb (str, optional): String representing the path of output data in .npy or .dat format that will be
used during csim and cosim.
backend (str, optional): Name of the backend to use, e.g., 'Vivado'
or 'Quartus'.
board (str, optional): One of target boards specified in `supported_board.json` file. If set to `None` a default
device of a backend will be used. See documentation of the backend used.
part (str, optional): The FPGA part. If set to `None` a default part of a backend will be used.
See documentation of the backend used. Note that if `board` is specified, the part associated to that board
will overwrite any part passed as a parameter.
clock_period (int, optional): Clock period of the design.
Defaults to 5.
io_type (str, optional): Type of implementation used. One of
'io_parallel' or 'io_serial'. Defaults to 'io_parallel'.
hls_config (dict, optional): The HLS config.
kwargs** (dict, optional): Additional parameters that will be used to create the config of the specified backend
Raises:
Exception: If precision and reuse factor are not present in 'hls_config'
Returns:
ModelGraph: hls4ml model.
"""
config = create_config(output_dir=output_dir,
project_name=project_name,
backend=backend,
**kwargs)
config['KerasModel'] = model
config['InputData'] = input_data_tb
config['OutputPredictions'] = output_data_tb
config['HLSConfig'] = {}
model_config = hls_config.get('Model', None)
config['HLSConfig']['Model'] = _check_model_config(model_config)
_check_hls_config(config, hls_config)
return keras_to_hls(config)
def hls_model(settings):
io_type = settings[0]
strategy = settings[1]
config = yaml.safe_load(open('../../example-models/config-files/qkeras_mnist_cnn_config.yml').read())
config['KerasJson'] = '../../example-models/keras/qkeras_mnist_cnn.json'
config['KerasH5'] = '../../example-models/keras/qkeras_mnist_cnn_weights.h5'
config['OutputDir'] = 'hls4mlprj_cnn_mnist_{}_{}'.format(io_type, strategy)
config['IOType'] = io_type
config['HLSConfig']['Model']['Strategy'] = strategy
config['HLSConfig']['LayerName']['softmax']['Strategy'] = 'Stable'
hls_model = keras_to_hls(config)
hls_model.compile()
return hls_model
def hls_model(settings):
io_type = settings[0]
strategy = settings[1]
yml_path = example_model_path / 'config-files/qkeras_mnist_cnn_config.yml'
with yml_path.open('r') as f:
config = yaml.safe_load(f.read())
config['KerasJson'] = str(example_model_path / 'keras/qkeras_mnist_cnn.json')
config['KerasH5'] = str(example_model_path / 'keras/qkeras_mnist_cnn_weights.h5')
config['OutputDir'] = str(test_root_path / 'hls4mlprj_cnn_mnist_{}_{}'.format(io_type, strategy))
config['IOType'] = io_type
config['HLSConfig']['Model']['Strategy'] = strategy
config['HLSConfig']['LayerName']['softmax']['Strategy'] = 'Stable'
hls_model = keras_to_hls(config)
hls_model.compile()
return hls_model
def convert_from_yaml_config(yamlConfig):
model = None
if 'OnnxModel' in yamlConfig:
if __onnx_enabled__:
model = onnx_to_hls(yamlConfig)
else:
raise Exception("ONNX not found. Please install ONNX.")
elif 'PytorchModel' in yamlConfig:
if __pytorch_enabled__:
model = pytorch_to_hls(yamlConfig)
else:
raise Exception("PyTorch not found. Please install PyTorch.")
elif 'TensorFlowModel' in yamlConfig:
if __tensorflow_enabled__:
model = tf_to_hls(yamlConfig)
else:
raise Exception("TensorFlow not found. Please install TensorFlow.")
else:
model = keras_to_hls(yamlConfig)
return model
def hls_model(settings):
io_type = settings[0]
strategy = settings[1]
config = hls4ml.converters.create_config(
output_dir='hls4mlprj_conv1d_{}_{}'.format(io_type, strategy))
config['KerasJson'] = str(example_model_path / 'keras/KERAS_conv1d.json')
config['KerasH5'] = str(example_model_path /
'keras/KERAS_conv1d_weights.h5')
config['OutputDir'] = str(
test_root_path / 'hls4mlprj_conv1d_{}_{}'.format(io_type, strategy))
config['IOType'] = io_type
hls_config = {
'Model': {
'Strategy': strategy,
'ReuseFactor': 1,
'Precision': 'ap_fixed<16,3,AP_RND_CONV,AP_SAT>'
}
}
# Some model specific precision tuning
config['LayerName'] = {}
config['LayerName']['fc1_relu'] = {
'Precision': {
'weight': 'ap_fixed<16,3>',
'result': 'ap_fixed<16,6,AP_RND_CONV,AP_SAT>'
}
}
config['LayerName']['output_softmax'] = {
'Precision': {
'weight': 'ap_fixed<16,6>',
'result': 'ap_fixed<16,6,AP_RND_CONV,AP_SAT>'
}
}
config['LayerName']['output_softmax_softmax'] = {'Strategy': 'Stable'}
config['HLSConfig'] = hls_config
hls_model = keras_to_hls(config)
hls_model.compile()
return hls_model
def convert_from_config(config):
"""Convert to hls4ml model based on the provided configuration.
Arguments:
config: A string containing the path to the YAML configuration file on
the filesystem or a dict containig the parsed configuration.
Returns:
HLSModel: hls4ml model.
"""
if isinstance(config, str):
yamlConfig = parse_yaml_config(config)
else:
yamlConfig = config
model = None
if 'OnnxModel' in yamlConfig:
if __onnx_enabled__:
model = onnx_to_hls(yamlConfig)
else:
raise Exception("ONNX not found. Please install ONNX.")
elif 'PytorchModel' in yamlConfig:
if __pytorch_enabled__:
model = pytorch_to_hls(yamlConfig)
else:
raise Exception("PyTorch not found. Please install PyTorch.")
elif 'TensorFlowModel' in yamlConfig:
if __tensorflow_enabled__:
model = tf_to_hls(yamlConfig)
else:
raise Exception("TensorFlow not found. Please install TensorFlow.")
else:
model = keras_to_hls(yamlConfig)
return model
def convert_from_keras_model(model,
output_dir='my-hls-test',
project_name='myproject',
fpga_part='xcku115-flvb2104-2-i',
clock_period=5,
io_type='io_parallel',
hls_config={}):
"""Convert to hls4ml model based on the provided configuration.
Args:
model: Keras model to convert
output_dir (str, optional): Output directory of the generated HLS
project. Defaults to 'my-hls-test'.
project_name (str, optional): Name of the HLS project.
Defaults to 'myproject'.
fpga_part (str, optional): The target FPGA device.
Defaults to 'xcku115-flvb2104-2-i'.
clock_period (int, optional): Clock period of the design.
Defaults to 5.
io_type (str, optional): Type of implementation used. One of
'io_parallel' or 'io_serial'. Defaults to 'io_parallel'.
hls_config (dict, optional): The HLS config.
Raises:
Exception: If precision and reuse factor are not present in 'hls_config'
Returns:
HLSModel: hls4ml model.
"""
config = create_vivado_config(output_dir=output_dir,
project_name=project_name,
fpga_part=fpga_part,
clock_period=clock_period,
io_type=io_type)
config['KerasModel'] = model
model_config = hls_config.get('Model', None)
if model_config is not None:
if not all(k in model_config for k in ('Precision', 'ReuseFactor')):
raise Exception(
'Precision and ReuseFactor must be provided in the hls_config')
else:
model_config = {}
model_config['Precision'] = 'ap_fixed<16,6>'
model_config['ReuseFactor'] = '1'
config['HLSConfig']['Model'] = model_config
if 'LayerName' in hls_config:
config['HLSConfig']['LayerName'] = hls_config['LayerName']
if 'LayerType' in hls_config:
config['HLSConfig']['LayerType'] = hls_config['LayerType']
if 'Optimizers' in hls_config:
config['HLSConfig']['Optimizers'] = hls_config['Optimizers']
if 'SkipOptimizers' in hls_config:
config['HLSConfig']['SkipOptimizers'] = hls_config['SkipOptimizers']
return keras_to_hls(config)
