diff --git a/.github/workflows/python-publish.yml b/.github/workflows/python-publish.yml
new file mode 100644
index 0000000..9087804
--- /dev/null
+++ b/.github/workflows/python-publish.yml
@@ -0,0 +1,31 @@
+# This workflow will upload a Python Package using Twine when a release is created
+# For more information see: https://help.github.com/en/actions/language-and-framework-guides/using-python-with-github-actions#publishing-to-package-registries
+
+name: Upload Python Package
+
+on:
+ release:
+ types: [created]
+
+jobs:
+ deploy:
+
+ runs-on: ubuntu-latest
+
+ steps:
+ - uses: actions/checkout@v2
+ - name: Set up Python
+ uses: actions/setup-python@v2
+ with:
+ python-version: '3.x'
+ - name: Install dependencies
+ run: |
+ python -m pip install --upgrade pip
+ pip install setuptools wheel twine
+ - name: Build and publish
+ env:
+ TWINE_USERNAME: __token__
+ TWINE_PASSWORD: ${{ secrets.PYPI_API_TOKEN }}
+ run: |
+ python setup.py sdist
+ twine upload dist/*
diff --git a/AUTHORS.rst b/AUTHORS.rst
new file mode 100644
index 0000000..ba06e5d
--- /dev/null
+++ b/AUTHORS.rst
@@ -0,0 +1,10 @@
+============
+Contributors
+============
+
+* Yaman Umuroglu (@maltanar) (maintainer)
+* Jakoba Petri-Koenig (@auphelia)
+* Lucian Petrica (@quetric)
+* Tobias Alonso (@Tobi-Alonso)
+* Hendrik Borras (@HenniOVP)
+* Felix Paul Jentzsch (@felixpj)
diff --git a/README.md b/README.md
index 5168620..8467a64 100644
--- a/README.md
+++ b/README.md
@@ -14,10 +14,13 @@ pre-built bitfiles, PYNQ Python drivers and Jupyter notebooks to get started,
and you can rebuild them from source.
Both PYNQ on Zynq and Alveo are supported.
+Need help with a problem in this repo, or got a question? Feel free to ask for help in the [FINN Gitter channel](https://gitter.im/xilinx-finn/community).
+
## Quickstart
*For Alveo we recommend setting up everything inside a virtualenv as described [here](https://pynq.readthedocs.io/en/v2.6.1/getting_started/alveo_getting_started.html?highlight=alveo#install-conda).*
+*For PYNQ boards, all commands below must be prefixed with `sudo` or by first going into `sudo su`.*
First, ensure that your `pip` and `setuptools` installations are up-to-date
on your PYNQ board or Alveo server:
@@ -62,11 +65,11 @@ dummy_out = accel.execute(dummy_in)
|----------------------------------------------------------------|-------------------------|------------------------------------------------------------|------------------|
| CIFAR-10 | CNV (VGG-11-like) | several variants: 1/2-bit weights/activations | all |
| MNIST | 3-layer fully-connected | several variants: 1/2-bit weights/activations | all |
-| ImageNet | MobileNet-v1 | 4-bit weights and activations 8-bit first layer weights | Alveo U250 |
+| ImageNet | MobileNet-v1 | 4-bit weights and activations 8-bit first layer weights | Alveo U250 ZCU104 |
## Supported Boards
-*Note that the larger NNs are only available on Alveo boards.*
+*Note that the larger NNs are only available on Alveo or selected Zynq boards.*
`finn-examples` provides pre-built FPGA bitfiles for the following boards:
diff --git a/build/README.md b/build/README.md
index cd699c9..e284deb 100644
--- a/build/README.md
+++ b/build/README.md
@@ -27,8 +27,7 @@ Please see the READMEs under the respective subfolders here for instructions on
All examples in this repo use the same Python PYNQ driver, located under
`finn_examples/driver.py` in the repo. This driver can support any FINN-generated
-accelerator that doesn't use external weights, the only thing that needs to be
-specified is the configuration for the input and output tensors in the `io_shape_dict`. Have a look at `finn_examples/models.py` to see how this is done for the example models in this repo:
+accelerator, the only thing that needs to be specified is the configuration for the input and output tensors in the `io_shape_dict`. Have a look at `finn_examples/models.py` to see how this is done for the example models in this repo:
```python
_cifar10_cnv_io_shape_dict = {
diff --git a/build/get-finn.sh b/build/get-finn.sh
index 3f5cf53..016a69c 100755
--- a/build/get-finn.sh
+++ b/build/get-finn.sh
@@ -30,7 +30,7 @@
# URL for git repo to be cloned
REPO_URL=https://github.com/Xilinx/finn
# commit hash for repo
-REPO_COMMIT=4fee6ffd8e13f91314ec9086e9ce9b2ea9de15c7
+REPO_COMMIT=e5da788bdc74fc9c234bb0176521ad51e830c22e
# directory (under the same folder as this script) to clone to
REPO_DIR=finn
diff --git a/build/mobilenet-v1/README.md b/build/mobilenet-v1/README.md
index b929679..38b69a2 100644
--- a/build/mobilenet-v1/README.md
+++ b/build/mobilenet-v1/README.md
@@ -17,7 +17,8 @@ It requires about 2 MB of weight storage and 1.1 GMACs per inference, yielding
Due to the depthwise separable convolutions in MobileNet-v1,
we use a specialized build script that replaces a few of the standard steps
in FINN with custom ones.
-**MobileNet-v1 is currently only supported on Alveo U250.**
+**MobileNet-v1 is currently only supported on Alveo U250 and ZCU104.**
+We also provide a folding configuration for the **ZCU102**, but there is no pre-built Pynq image available for this board.
0. Ensure you have performed the *Setup* steps in the top-level README for setting up the FINN requirements and environment variables.
diff --git a/build/mobilenet-v1/build.py b/build/mobilenet-v1/build.py
index 5d67f1c..fc4c48c 100644
--- a/build/mobilenet-v1/build.py
+++ b/build/mobilenet-v1/build.py
@@ -28,53 +28,147 @@
import finn.builder.build_dataflow as build
import finn.builder.build_dataflow_config as build_cfg
+from finn.util.basic import alveo_default_platform
+import os
+import shutil
# custom steps for mobilenetv1
from custom_steps import (
step_mobilenet_streamline,
step_mobilenet_convert_to_hls_layers,
+ step_mobilenet_convert_to_hls_layers_separate_th,
step_mobilenet_lower_convs,
+ step_mobilenet_slr_floorplan,
)
model_name = "mobilenetv1-w4a4"
-board = "U250"
-vitis_platform = "xilinx_u250_xdma_201830_2"
-synth_clk_period_ns = 3.0
-mobilenet_build_steps = [
- step_mobilenet_streamline,
- step_mobilenet_lower_convs,
- step_mobilenet_convert_to_hls_layers,
- "step_create_dataflow_partition",
- "step_apply_folding_config",
- "step_generate_estimate_reports",
- "step_hls_ipgen",
- "step_set_fifo_depths",
- "step_create_stitched_ip",
- "step_make_pynq_driver",
- "step_synthesize_bitfile",
- "step_deployment_package",
-]
-
-
-cfg = build_cfg.DataflowBuildConfig(
- steps=mobilenet_build_steps,
- output_dir="output_%s_%s" % (model_name, board),
- folding_config_file="folding_config/%s_folding_config.json" % board,
- synth_clk_period_ns=synth_clk_period_ns,
- board=board,
- shell_flow_type=build_cfg.ShellFlowType.VITIS_ALVEO,
- # folding config comes with FIFO depths already
- auto_fifo_depths=False,
- vitis_platform=vitis_platform,
- # enable extra performance optimizations (physopt)
- vitis_opt_strategy=build_cfg.VitisOptStrategyCfg.PERFORMANCE_BEST,
- generate_outputs=[
- build_cfg.DataflowOutputType.PYNQ_DRIVER,
- build_cfg.DataflowOutputType.ESTIMATE_REPORTS,
- build_cfg.DataflowOutputType.BITFILE,
- build_cfg.DataflowOutputType.DEPLOYMENT_PACKAGE,
- ],
-)
-model_file = "models/%s_pre_post_tidy.onnx" % model_name
-build.build_dataflow_cfg(model_file, cfg)
+# which platforms to build the networks for
+zynq_platforms = ["ZCU102", "ZCU104"]
+#alveo_platforms = ["U50", "U200", "U250", "U280"]
+alveo_platforms = ["U250"]
+platforms_to_build = zynq_platforms + alveo_platforms
+
+
+# determine which shell flow to use for a given platform
+def platform_to_shell(platform):
+ if platform in zynq_platforms:
+ return build_cfg.ShellFlowType.VIVADO_ZYNQ
+ elif platform in alveo_platforms:
+ return build_cfg.ShellFlowType.VITIS_ALVEO
+ else:
+ raise Exception("Unknown platform, can't determine ShellFlowType")
+
+
+# select target clock frequency
+def select_clk_period(platform):
+ if platform in zynq_platforms:
+ return 5.4
+ elif platform in alveo_platforms:
+ return 3.0
+
+
+# select build steps (ZCU104/102 folding config is based on separate thresholding nodes)
+def select_build_steps(platform):
+ if platform in zynq_platforms:
+ return [
+ step_mobilenet_streamline,
+ step_mobilenet_lower_convs,
+ step_mobilenet_convert_to_hls_layers_separate_th,
+ "step_create_dataflow_partition",
+ "step_apply_folding_config",
+ "step_generate_estimate_reports",
+ "step_hls_codegen",
+ "step_hls_ipgen",
+ "step_set_fifo_depths",
+ "step_create_stitched_ip",
+ "step_synthesize_bitfile",
+ "step_make_pynq_driver",
+ "step_deployment_package",
+ ]
+ elif platform in alveo_platforms:
+ return [
+ step_mobilenet_streamline,
+ step_mobilenet_lower_convs,
+ step_mobilenet_convert_to_hls_layers,
+ "step_create_dataflow_partition",
+ "step_apply_folding_config",
+ "step_generate_estimate_reports",
+ "step_hls_codegen",
+ "step_hls_ipgen",
+ "step_set_fifo_depths",
+ step_mobilenet_slr_floorplan,
+ "step_synthesize_bitfile",
+ "step_make_pynq_driver",
+ "step_deployment_package",
+ ]
+
+
+# create a release dir, used for finn-examples release packaging
+os.makedirs("release", exist_ok=True)
+
+
+for platform_name in platforms_to_build:
+ shell_flow_type = platform_to_shell(platform_name)
+ if shell_flow_type == build_cfg.ShellFlowType.VITIS_ALVEO:
+ vitis_platform = alveo_default_platform[platform_name]
+ # for Alveo, use the Vitis platform name as the release name
+ # e.g. xilinx_u250_xdma_201830_2
+ release_platform_name = vitis_platform
+ else:
+ vitis_platform = None
+ # for Zynq, use the board name as the release name
+ # e.g. ZCU104
+ release_platform_name = platform_name
+ platform_dir = "release/%s" % release_platform_name
+ os.makedirs(platform_dir, exist_ok=True)
+
+ cfg = build_cfg.DataflowBuildConfig(
+ steps=select_build_steps(platform_name),
+ output_dir="output_%s_%s" % (model_name, release_platform_name),
+ folding_config_file="folding_config/%s_folding_config.json" % platform_name,
+ synth_clk_period_ns=select_clk_period(platform_name),
+ board=platform_name,
+ shell_flow_type=shell_flow_type,
+ vitis_platform=vitis_platform,
+ # folding config comes with FIFO depths already
+ auto_fifo_depths=False,
+ # enable extra performance optimizations (physopt)
+ vitis_opt_strategy=build_cfg.VitisOptStrategyCfg.PERFORMANCE_BEST,
+ generate_outputs=[
+ build_cfg.DataflowOutputType.PYNQ_DRIVER,
+ build_cfg.DataflowOutputType.ESTIMATE_REPORTS,
+ build_cfg.DataflowOutputType.BITFILE,
+ build_cfg.DataflowOutputType.DEPLOYMENT_PACKAGE,
+ ],
+ )
+ model_file = "models/%s_pre_post_tidy.onnx" % model_name
+ build.build_dataflow_cfg(model_file, cfg)
+
+ # copy bitfiles and runtime weights into release dir if found
+ bitfile_gen_dir = cfg.output_dir + "/bitfile"
+ files_to_check_and_copy = [
+ "finn-accel.bit",
+ "finn-accel.hwh",
+ "finn-accel.xclbin",
+ ]
+ for f in files_to_check_and_copy:
+ src_file = bitfile_gen_dir + "/" + f
+ dst_file = platform_dir + "/" + f.replace("finn-accel", model_name)
+ if os.path.isfile(src_file):
+ shutil.copy(src_file, dst_file)
+
+ weight_gen_dir = cfg.output_dir + "/driver/runtime_weights"
+ weight_dst_dir = platform_dir + "/%s_runtime_weights" % model_name
+ if os.path.isdir(weight_gen_dir):
+ weight_files = os.listdir(weight_gen_dir)
+ if weight_files:
+ shutil.copytree(weight_gen_dir, weight_dst_dir)
+
+ # create zipfile for all examples for this platform
+ shutil.make_archive(
+ "release/" + release_platform_name,
+ "zip",
+ root_dir="release",
+ base_dir=release_platform_name,
+ )
diff --git a/build/mobilenet-v1/custom_steps.py b/build/mobilenet-v1/custom_steps.py
index a18faee..9f30597 100644
--- a/build/mobilenet-v1/custom_steps.py
+++ b/build/mobilenet-v1/custom_steps.py
@@ -26,7 +26,10 @@
# OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
from finn.core.modelwrapper import ModelWrapper
-from finn.builder.build_dataflow_config import DataflowBuildConfig
+from finn.builder.build_dataflow_config import (
+ DataflowBuildConfig,
+ ShellFlowType,
+)
from finn.transformation.streamline import Streamline
from finn.transformation.double_to_single_float import DoubleToSingleFloat
import finn.transformation.streamline.absorb as absorb
@@ -39,6 +42,7 @@
from finn.transformation.general import (
GiveReadableTensorNames,
GiveUniqueNodeNames,
+ ApplyConfig,
)
import finn.transformation.fpgadataflow.convert_to_hls_layers as to_hls
from finn.transformation.infer_shapes import InferShapes
@@ -94,3 +98,36 @@ def step_mobilenet_convert_to_hls_layers(model: ModelWrapper, cfg: DataflowBuild
model = model.transform(GiveUniqueNodeNames())
model = model.transform(GiveReadableTensorNames())
return model
+
+
+def step_mobilenet_slr_floorplan(model: ModelWrapper, cfg: DataflowBuildConfig):
+ if cfg.shell_flow_type == ShellFlowType.VITIS_ALVEO:
+ try:
+ from finn.analysis.partitioning import partition
+ # apply partitioning of the model, restricting the first and last layers to SLR0
+ default_slr = 0
+ abs_anchors = [(0,[default_slr]),(-1,[default_slr])]
+ floorplan = partition(model, cfg.synth_clk_period_ns, cfg.board, abs_anchors=abs_anchors, multivariant=False)[0]
+ # apply floorplan to model
+ model = model.transform(ApplyConfig(floorplan))
+ print("SLR floorplanning applied")
+ except:
+ print("No SLR floorplanning applied")
+ return model
+
+
+def step_mobilenet_convert_to_hls_layers_separate_th(
+ model: ModelWrapper, cfg: DataflowBuildConfig
+):
+ mem_mode = cfg.default_mem_mode.value
+ model = model.transform(to_hls.InferPool_Batch())
+ model = model.transform(to_hls.InferConvInpGen())
+ model = model.transform(to_hls.InferThresholdingLayer())
+ model = model.transform(to_hls.InferVVAU())
+ model = model.transform(to_hls.InferQuantizedStreamingFCLayer(mem_mode))
+ model = model.transform(to_hls.InferChannelwiseLinearLayer())
+ model = model.transform(to_hls.InferLabelSelectLayer())
+ model = model.transform(InferShapes())
+ model = model.transform(GiveUniqueNodeNames())
+ model = model.transform(GiveReadableTensorNames())
+ return model
diff --git a/build/mobilenet-v1/folding_config/U200_folding_config.json b/build/mobilenet-v1/folding_config/U200_folding_config.json
new file mode 100644
index 0000000..f5ccf9b
--- /dev/null
+++ b/build/mobilenet-v1/folding_config/U200_folding_config.json
@@ -0,0 +1,499 @@
+{
+ "Defaults": {},
+ "StreamingFIFO_0": {
+ "ram_style": "ultra",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_0": {
+ "SIMD": 3,
+ "ram_style": "distributed"
+ },
+ "StreamingFCLayer_Batch_0": {
+ "PE": 32,
+ "SIMD": 3,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "dsp"
+ },
+ "FMPadding_Batch_0": {
+ "SIMD": 32
+ },
+ "StreamingFIFO_3": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "ConvolutionInputGenerator_1": {
+ "SIMD": 32,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_0": {
+ "PE": 32,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_0": {
+ "impl_style": "hls"
+ },
+ "StreamingFCLayer_Batch_1": {
+ "PE": 16,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_1": {
+ "impl_style": "hls"
+ },
+ "FMPadding_Batch_1": {
+ "SIMD": 32
+ },
+ "StreamingFIFO_9": {
+ "ram_style": "ultra",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_2": {
+ "SIMD": 32,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_1": {
+ "PE": 32,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_2": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_12": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_2": {
+ "PE": 16,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_3": {
+ "impl_style": "hls"
+ },
+ "FMPadding_Batch_2": {
+ "SIMD": 64
+ },
+ "StreamingFIFO_15": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "ConvolutionInputGenerator_3": {
+ "SIMD": 64,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_2": {
+ "PE": 64,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_4": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_18": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_3": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_5": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_20": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_3": {
+ "SIMD": 16
+ },
+ "StreamingFIFO_21": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_4": {
+ "SIMD": 16,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_3": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingFIFO_23": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_4": {
+ "PE": 16,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_6": {
+ "impl_style": "hls"
+ },
+ "FMPadding_Batch_4": {
+ "SIMD": 32
+ },
+ "StreamingFIFO_26": {
+ "ram_style": "ultra",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_5": {
+ "SIMD": 32,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_4": {
+ "PE": 32,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_7": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_29": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_5": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_8": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_31": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_5": {
+ "SIMD": 8
+ },
+ "StreamingFIFO_32": {
+ "ram_style": "ultra",
+ "depth": 2048,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_6": {
+ "SIMD": 8,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_5": {
+ "PE": 8,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_9": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_35": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_6": {
+ "PE": 16,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "FMPadding_Batch_6": {
+ "SIMD": 16
+ },
+ "StreamingFIFO_37": {
+ "ram_style": "ultra",
+ "depth": 2048,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_7": {
+ "SIMD": 16,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_6": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingFIFO_39": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_7": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_10": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_41": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_7": {
+ "SIMD": 16
+ },
+ "StreamingFIFO_42": {
+ "ram_style": "ultra",
+ "depth": 2048,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_8": {
+ "SIMD": 16,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_7": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingFIFO_44": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_8": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_11": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_46": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_8": {
+ "SIMD": 16
+ },
+ "StreamingFIFO_47": {
+ "ram_style": "ultra",
+ "depth": 2048,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_9": {
+ "SIMD": 16,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_8": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingFIFO_49": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_9": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_12": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_51": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_9": {
+ "SIMD": 16
+ },
+ "StreamingFIFO_52": {
+ "ram_style": "ultra",
+ "depth": 2048,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_10": {
+ "SIMD": 16,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_9": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingFIFO_54": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_10": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_13": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_56": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_10": {
+ "SIMD": 16
+ },
+ "StreamingFIFO_57": {
+ "ram_style": "ultra",
+ "depth": 2048,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_11": {
+ "SIMD": 16,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_10": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingFIFO_59": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_11": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_14": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_61": {
+ "ram_style": "auto",
+ "depth": 128,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_11": {
+ "SIMD": 4
+ },
+ "StreamingFIFO_62": {
+ "ram_style": "ultra",
+ "depth": 4096,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_12": {
+ "SIMD": 4,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_11": {
+ "PE": 4,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_15": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_65": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_12": {
+ "PE": 16,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_16": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_67": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_12": {
+ "SIMD": 8
+ },
+ "StreamingFIFO_68": {
+ "ram_style": "ultra",
+ "depth": 4096,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_13": {
+ "SIMD": 8,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_12": {
+ "PE": 8,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_17": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_71": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_13": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_18": {
+ "impl_style": "hls"
+ },
+ "ConvolutionInputGenerator_14": {
+ "SIMD": 4,
+ "ram_style": "distributed"
+ },
+ "Pool_Batch_0": {
+ "PE": 4
+ },
+ "StreamingFCLayer_Batch_14": {
+ "PE": 4,
+ "SIMD": 4,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_19": {
+ "impl_style": "hls"
+ },
+ "ChannelwiseOp_Batch_0": {
+ "PE": 1,
+ "ram_style": "distributed"
+ },
+ "LabelSelect_Batch_0": {
+ "PE": 1
+ }
+}
diff --git a/build/mobilenet-v1/folding_config/U280_folding_config.json b/build/mobilenet-v1/folding_config/U280_folding_config.json
new file mode 100644
index 0000000..f5ccf9b
--- /dev/null
+++ b/build/mobilenet-v1/folding_config/U280_folding_config.json
@@ -0,0 +1,499 @@
+{
+ "Defaults": {},
+ "StreamingFIFO_0": {
+ "ram_style": "ultra",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_0": {
+ "SIMD": 3,
+ "ram_style": "distributed"
+ },
+ "StreamingFCLayer_Batch_0": {
+ "PE": 32,
+ "SIMD": 3,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "dsp"
+ },
+ "FMPadding_Batch_0": {
+ "SIMD": 32
+ },
+ "StreamingFIFO_3": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "ConvolutionInputGenerator_1": {
+ "SIMD": 32,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_0": {
+ "PE": 32,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_0": {
+ "impl_style": "hls"
+ },
+ "StreamingFCLayer_Batch_1": {
+ "PE": 16,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_1": {
+ "impl_style": "hls"
+ },
+ "FMPadding_Batch_1": {
+ "SIMD": 32
+ },
+ "StreamingFIFO_9": {
+ "ram_style": "ultra",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_2": {
+ "SIMD": 32,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_1": {
+ "PE": 32,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_2": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_12": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_2": {
+ "PE": 16,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_3": {
+ "impl_style": "hls"
+ },
+ "FMPadding_Batch_2": {
+ "SIMD": 64
+ },
+ "StreamingFIFO_15": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "ConvolutionInputGenerator_3": {
+ "SIMD": 64,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_2": {
+ "PE": 64,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_4": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_18": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_3": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_5": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_20": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_3": {
+ "SIMD": 16
+ },
+ "StreamingFIFO_21": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_4": {
+ "SIMD": 16,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_3": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingFIFO_23": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_4": {
+ "PE": 16,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_6": {
+ "impl_style": "hls"
+ },
+ "FMPadding_Batch_4": {
+ "SIMD": 32
+ },
+ "StreamingFIFO_26": {
+ "ram_style": "ultra",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_5": {
+ "SIMD": 32,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_4": {
+ "PE": 32,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_7": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_29": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_5": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_8": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_31": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_5": {
+ "SIMD": 8
+ },
+ "StreamingFIFO_32": {
+ "ram_style": "ultra",
+ "depth": 2048,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_6": {
+ "SIMD": 8,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_5": {
+ "PE": 8,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_9": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_35": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_6": {
+ "PE": 16,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "FMPadding_Batch_6": {
+ "SIMD": 16
+ },
+ "StreamingFIFO_37": {
+ "ram_style": "ultra",
+ "depth": 2048,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_7": {
+ "SIMD": 16,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_6": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingFIFO_39": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_7": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_10": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_41": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_7": {
+ "SIMD": 16
+ },
+ "StreamingFIFO_42": {
+ "ram_style": "ultra",
+ "depth": 2048,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_8": {
+ "SIMD": 16,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_7": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingFIFO_44": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_8": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_11": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_46": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_8": {
+ "SIMD": 16
+ },
+ "StreamingFIFO_47": {
+ "ram_style": "ultra",
+ "depth": 2048,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_9": {
+ "SIMD": 16,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_8": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingFIFO_49": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_9": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_12": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_51": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_9": {
+ "SIMD": 16
+ },
+ "StreamingFIFO_52": {
+ "ram_style": "ultra",
+ "depth": 2048,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_10": {
+ "SIMD": 16,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_9": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingFIFO_54": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_10": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_13": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_56": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_10": {
+ "SIMD": 16
+ },
+ "StreamingFIFO_57": {
+ "ram_style": "ultra",
+ "depth": 2048,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_11": {
+ "SIMD": 16,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_10": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingFIFO_59": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_11": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_14": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_61": {
+ "ram_style": "auto",
+ "depth": 128,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_11": {
+ "SIMD": 4
+ },
+ "StreamingFIFO_62": {
+ "ram_style": "ultra",
+ "depth": 4096,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_12": {
+ "SIMD": 4,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_11": {
+ "PE": 4,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_15": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_65": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_12": {
+ "PE": 16,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_16": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_67": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_12": {
+ "SIMD": 8
+ },
+ "StreamingFIFO_68": {
+ "ram_style": "ultra",
+ "depth": 4096,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_13": {
+ "SIMD": 8,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_12": {
+ "PE": 8,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_17": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_71": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_13": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_18": {
+ "impl_style": "hls"
+ },
+ "ConvolutionInputGenerator_14": {
+ "SIMD": 4,
+ "ram_style": "distributed"
+ },
+ "Pool_Batch_0": {
+ "PE": 4
+ },
+ "StreamingFCLayer_Batch_14": {
+ "PE": 4,
+ "SIMD": 4,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_19": {
+ "impl_style": "hls"
+ },
+ "ChannelwiseOp_Batch_0": {
+ "PE": 1,
+ "ram_style": "distributed"
+ },
+ "LabelSelect_Batch_0": {
+ "PE": 1
+ }
+}
diff --git a/build/mobilenet-v1/folding_config/U50_folding_config.json b/build/mobilenet-v1/folding_config/U50_folding_config.json
new file mode 100644
index 0000000..f5ccf9b
--- /dev/null
+++ b/build/mobilenet-v1/folding_config/U50_folding_config.json
@@ -0,0 +1,499 @@
+{
+ "Defaults": {},
+ "StreamingFIFO_0": {
+ "ram_style": "ultra",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_0": {
+ "SIMD": 3,
+ "ram_style": "distributed"
+ },
+ "StreamingFCLayer_Batch_0": {
+ "PE": 32,
+ "SIMD": 3,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "dsp"
+ },
+ "FMPadding_Batch_0": {
+ "SIMD": 32
+ },
+ "StreamingFIFO_3": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "ConvolutionInputGenerator_1": {
+ "SIMD": 32,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_0": {
+ "PE": 32,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_0": {
+ "impl_style": "hls"
+ },
+ "StreamingFCLayer_Batch_1": {
+ "PE": 16,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_1": {
+ "impl_style": "hls"
+ },
+ "FMPadding_Batch_1": {
+ "SIMD": 32
+ },
+ "StreamingFIFO_9": {
+ "ram_style": "ultra",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_2": {
+ "SIMD": 32,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_1": {
+ "PE": 32,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_2": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_12": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_2": {
+ "PE": 16,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_3": {
+ "impl_style": "hls"
+ },
+ "FMPadding_Batch_2": {
+ "SIMD": 64
+ },
+ "StreamingFIFO_15": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "ConvolutionInputGenerator_3": {
+ "SIMD": 64,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_2": {
+ "PE": 64,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_4": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_18": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_3": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_5": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_20": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_3": {
+ "SIMD": 16
+ },
+ "StreamingFIFO_21": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_4": {
+ "SIMD": 16,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_3": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingFIFO_23": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_4": {
+ "PE": 16,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_6": {
+ "impl_style": "hls"
+ },
+ "FMPadding_Batch_4": {
+ "SIMD": 32
+ },
+ "StreamingFIFO_26": {
+ "ram_style": "ultra",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_5": {
+ "SIMD": 32,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_4": {
+ "PE": 32,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_7": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_29": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_5": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_8": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_31": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_5": {
+ "SIMD": 8
+ },
+ "StreamingFIFO_32": {
+ "ram_style": "ultra",
+ "depth": 2048,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_6": {
+ "SIMD": 8,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_5": {
+ "PE": 8,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_9": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_35": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_6": {
+ "PE": 16,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "FMPadding_Batch_6": {
+ "SIMD": 16
+ },
+ "StreamingFIFO_37": {
+ "ram_style": "ultra",
+ "depth": 2048,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_7": {
+ "SIMD": 16,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_6": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingFIFO_39": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_7": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_10": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_41": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_7": {
+ "SIMD": 16
+ },
+ "StreamingFIFO_42": {
+ "ram_style": "ultra",
+ "depth": 2048,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_8": {
+ "SIMD": 16,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_7": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingFIFO_44": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_8": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_11": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_46": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_8": {
+ "SIMD": 16
+ },
+ "StreamingFIFO_47": {
+ "ram_style": "ultra",
+ "depth": 2048,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_9": {
+ "SIMD": 16,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_8": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingFIFO_49": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_9": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_12": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_51": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_9": {
+ "SIMD": 16
+ },
+ "StreamingFIFO_52": {
+ "ram_style": "ultra",
+ "depth": 2048,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_10": {
+ "SIMD": 16,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_9": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingFIFO_54": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_10": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_13": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_56": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_10": {
+ "SIMD": 16
+ },
+ "StreamingFIFO_57": {
+ "ram_style": "ultra",
+ "depth": 2048,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_11": {
+ "SIMD": 16,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_10": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingFIFO_59": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_11": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_14": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_61": {
+ "ram_style": "auto",
+ "depth": 128,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_11": {
+ "SIMD": 4
+ },
+ "StreamingFIFO_62": {
+ "ram_style": "ultra",
+ "depth": 4096,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_12": {
+ "SIMD": 4,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_11": {
+ "PE": 4,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_15": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_65": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_12": {
+ "PE": 16,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_16": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_67": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_12": {
+ "SIMD": 8
+ },
+ "StreamingFIFO_68": {
+ "ram_style": "ultra",
+ "depth": 4096,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_13": {
+ "SIMD": 8,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_12": {
+ "PE": 8,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_17": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_71": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "StreamingFCLayer_Batch_13": {
+ "PE": 32,
+ "SIMD": 16,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_18": {
+ "impl_style": "hls"
+ },
+ "ConvolutionInputGenerator_14": {
+ "SIMD": 4,
+ "ram_style": "distributed"
+ },
+ "Pool_Batch_0": {
+ "PE": 4
+ },
+ "StreamingFCLayer_Batch_14": {
+ "PE": 4,
+ "SIMD": 4,
+ "ram_style": "block",
+ "mem_mode": "decoupled",
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_19": {
+ "impl_style": "hls"
+ },
+ "ChannelwiseOp_Batch_0": {
+ "PE": 1,
+ "ram_style": "distributed"
+ },
+ "LabelSelect_Batch_0": {
+ "PE": 1
+ }
+}
diff --git a/build/mobilenet-v1/folding_config/ZCU102_folding_config.json b/build/mobilenet-v1/folding_config/ZCU102_folding_config.json
new file mode 100755
index 0000000..02d6d6e
--- /dev/null
+++ b/build/mobilenet-v1/folding_config/ZCU102_folding_config.json
@@ -0,0 +1,816 @@
+{
+ "Defaults": {},
+ "StreamingFIFO_0": {
+ "ram_style": "block",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_0": {
+ "SIMD": 1,
+ "ram_style": "distributed"
+ },
+ "StreamingDataWidthConverter_Batch_0": {
+ "impl_style": "hls"
+ },
+ "StreamingFCLayer_Batch_0": {
+ "PE": 16,
+ "SIMD": 3,
+ "ram_style": "auto",
+ "resType": "dsp",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_3": {
+ "ram_style": "auto",
+ "depth": 64,
+ "impl_style": "rtl"
+ },
+ "StreamingDataWidthConverter_Batch_1": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_0": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_2": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_6": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_0": {
+ "SIMD": 2
+ },
+ "StreamingDataWidthConverter_Batch_3": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_8": {
+ "ram_style": "block",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_1": {
+ "SIMD": 16,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_0": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingFIFO_10": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "StreamingDataWidthConverter_Batch_4": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_1": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_5": {
+ "impl_style": "hls"
+ },
+ "StreamingFCLayer_Batch_1": {
+ "PE": 8,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_6": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_2": {
+ "PE": 2,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_7": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_17": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_1": {
+ "SIMD": 4
+ },
+ "StreamingDataWidthConverter_Batch_8": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_19": {
+ "ram_style": "block",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_2": {
+ "SIMD": 8,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_1": {
+ "PE": 8,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_9": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_3": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_10": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_24": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_2": {
+ "PE": 16,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_11": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_4": {
+ "PE": 2,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_27": {
+ "ram_style": "auto",
+ "depth": 128,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_2": {
+ "SIMD": 2
+ },
+ "StreamingDataWidthConverter_Batch_12": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_29": {
+ "ram_style": "block",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_3": {
+ "SIMD": 16,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_2": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_13": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_5": {
+ "PE": 2,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_14": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_34": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_3": {
+ "PE": 32,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_15": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_6": {
+ "PE": 2,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_37": {
+ "ram_style": "auto",
+ "depth": 128,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_3": {
+ "SIMD": 2
+ },
+ "StreamingDataWidthConverter_Batch_16": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_39": {
+ "ram_style": "block",
+ "depth": 4096,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_4": {
+ "SIMD": 4,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_3": {
+ "PE": 4,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_17": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_7": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_18": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_44": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_4": {
+ "PE": 16,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_19": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_8": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_47": {
+ "ram_style": "block",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "FMPadding_Batch_4": {
+ "SIMD": 1
+ },
+ "StreamingDataWidthConverter_Batch_20": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_49": {
+ "ram_style": "block",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_5": {
+ "SIMD": 8,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_4": {
+ "PE": 8,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_21": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_9": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_22": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_54": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_5": {
+ "PE": 32,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_23": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_10": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_57": {
+ "ram_style": "block",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "FMPadding_Batch_5": {
+ "SIMD": 1
+ },
+ "StreamingDataWidthConverter_Batch_24": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_59": {
+ "ram_style": "block",
+ "depth": 8192,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_6": {
+ "SIMD": 2,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_5": {
+ "PE": 2,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_25": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_11": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_26": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_64": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_6": {
+ "PE": 16,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_27": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_12": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_67": {
+ "ram_style": "block",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "FMPadding_Batch_6": {
+ "SIMD": 1
+ },
+ "StreamingDataWidthConverter_Batch_28": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_69": {
+ "ram_style": "block",
+ "depth": 4096,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_7": {
+ "SIMD": 4,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_6": {
+ "PE": 4,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_29": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_13": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_30": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_74": {
+ "ram_style": "auto",
+ "depth": 64,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_7": {
+ "PE": 32,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_31": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_14": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_77": {
+ "ram_style": "block",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "FMPadding_Batch_7": {
+ "SIMD": 1
+ },
+ "StreamingDataWidthConverter_Batch_32": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_79": {
+ "ram_style": "block",
+ "depth": 4096,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_8": {
+ "SIMD": 4,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_7": {
+ "PE": 4,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_33": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_15": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_34": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_84": {
+ "ram_style": "auto",
+ "depth": 64,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_8": {
+ "PE": 32,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_35": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_16": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_87": {
+ "ram_style": "block",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "FMPadding_Batch_8": {
+ "SIMD": 1
+ },
+ "StreamingDataWidthConverter_Batch_36": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_89": {
+ "ram_style": "block",
+ "depth": 4096,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_9": {
+ "SIMD": 4,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_8": {
+ "PE": 4,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_37": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_17": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_38": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_94": {
+ "ram_style": "auto",
+ "depth": 64,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_9": {
+ "PE": 32,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_39": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_18": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_97": {
+ "ram_style": "block",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "FMPadding_Batch_9": {
+ "SIMD": 1
+ },
+ "StreamingDataWidthConverter_Batch_40": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_99": {
+ "ram_style": "block",
+ "depth": 4096,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_10": {
+ "SIMD": 4,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_9": {
+ "PE": 4,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_41": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_19": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_42": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_104": {
+ "ram_style": "auto",
+ "depth": 64,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_10": {
+ "PE": 32,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_43": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_20": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_107": {
+ "ram_style": "block",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "FMPadding_Batch_10": {
+ "SIMD": 1
+ },
+ "StreamingDataWidthConverter_Batch_44": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_109": {
+ "ram_style": "block",
+ "depth": 4096,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_11": {
+ "SIMD": 4,
+ "ram_style": "block"
+ },
+ "Vector_Vector_Activate_Batch_10": {
+ "PE": 4,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_45": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_21": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_46": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_114": {
+ "ram_style": "auto",
+ "depth": 64,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_11": {
+ "PE": 32,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_47": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_22": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_117": {
+ "ram_style": "block",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "FMPadding_Batch_11": {
+ "SIMD": 1
+ },
+ "StreamingFIFO_118": {
+ "ram_style": "block",
+ "depth": 16384,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_12": {
+ "SIMD": 1,
+ "ram_style": "block"
+ },
+ "Vector_Vector_Activate_Batch_11": {
+ "PE": 1,
+ "resType": "lut"
+ },
+ "Thresholding_Batch_23": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_48": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_122": {
+ "ram_style": "auto",
+ "depth": 64,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_12": {
+ "PE": 16,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_49": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_24": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_125": {
+ "ram_style": "block",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "FMPadding_Batch_12": {
+ "SIMD": 1
+ },
+ "StreamingDataWidthConverter_Batch_50": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_127": {
+ "ram_style": "block",
+ "depth": 16384,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_13": {
+ "SIMD": 2,
+ "ram_style": "block"
+ },
+ "Vector_Vector_Activate_Batch_12": {
+ "PE": 2,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_51": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_25": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_52": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_132": {
+ "ram_style": "auto",
+ "depth": 128,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_13": {
+ "PE": 32,
+ "SIMD": 8,
+ "ram_style": "block",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_53": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_26": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "ConvolutionInputGenerator_14": {
+ "SIMD": 1,
+ "ram_style": "block"
+ },
+ "Pool_Batch_0": {
+ "PE": 1
+ },
+ "StreamingDataWidthConverter_Batch_54": {
+ "impl_style": "hls"
+ },
+ "StreamingFCLayer_Batch_14": {
+ "PE": 1,
+ "SIMD": 16,
+ "ram_style": "block",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "ChannelwiseOp_Batch_0": {
+ "PE": 1,
+ "ram_style": "distributed"
+ },
+ "LabelSelect_Batch_0": {
+ "PE": 1
+ }
+}
diff --git a/build/mobilenet-v1/folding_config/ZCU104_folding_config.json b/build/mobilenet-v1/folding_config/ZCU104_folding_config.json
new file mode 100755
index 0000000..b441206
--- /dev/null
+++ b/build/mobilenet-v1/folding_config/ZCU104_folding_config.json
@@ -0,0 +1,816 @@
+{
+ "Defaults": {},
+ "StreamingFIFO_0": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_0": {
+ "SIMD": 1,
+ "ram_style": "distributed"
+ },
+ "StreamingDataWidthConverter_Batch_0": {
+ "impl_style": "hls"
+ },
+ "StreamingFCLayer_Batch_0": {
+ "PE": 16,
+ "SIMD": 3,
+ "ram_style": "auto",
+ "resType": "dsp",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_3": {
+ "ram_style": "auto",
+ "depth": 64,
+ "impl_style": "rtl"
+ },
+ "StreamingDataWidthConverter_Batch_1": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_0": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_2": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_6": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_0": {
+ "SIMD": 2
+ },
+ "StreamingDataWidthConverter_Batch_3": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_8": {
+ "ram_style": "ultra",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_1": {
+ "SIMD": 16,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_0": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingFIFO_10": {
+ "ram_style": "auto",
+ "depth": 256,
+ "impl_style": "rtl"
+ },
+ "StreamingDataWidthConverter_Batch_4": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_1": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_5": {
+ "impl_style": "hls"
+ },
+ "StreamingFCLayer_Batch_1": {
+ "PE": 8,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_6": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_2": {
+ "PE": 2,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_7": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_17": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_1": {
+ "SIMD": 4
+ },
+ "StreamingDataWidthConverter_Batch_8": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_19": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_2": {
+ "SIMD": 8,
+ "ram_style": "distributed"
+ },
+ "Vector_Vector_Activate_Batch_1": {
+ "PE": 8,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_9": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_3": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_10": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_24": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_2": {
+ "PE": 16,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_11": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_4": {
+ "PE": 2,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_27": {
+ "ram_style": "auto",
+ "depth": 128,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_2": {
+ "SIMD": 2
+ },
+ "StreamingDataWidthConverter_Batch_12": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_29": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_3": {
+ "SIMD": 16,
+ "ram_style": "block"
+ },
+ "Vector_Vector_Activate_Batch_2": {
+ "PE": 16,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_13": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_5": {
+ "PE": 2,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_14": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_34": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_3": {
+ "PE": 32,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_15": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_6": {
+ "PE": 2,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_37": {
+ "ram_style": "auto",
+ "depth": 128,
+ "impl_style": "rtl"
+ },
+ "FMPadding_Batch_3": {
+ "SIMD": 2
+ },
+ "StreamingDataWidthConverter_Batch_16": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_39": {
+ "ram_style": "ultra",
+ "depth": 4096,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_4": {
+ "SIMD": 4,
+ "ram_style": "block"
+ },
+ "Vector_Vector_Activate_Batch_3": {
+ "PE": 4,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_17": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_7": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_18": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_44": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_4": {
+ "PE": 16,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_19": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_8": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_47": {
+ "ram_style": "ultra",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "FMPadding_Batch_4": {
+ "SIMD": 1
+ },
+ "StreamingDataWidthConverter_Batch_20": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_49": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_5": {
+ "SIMD": 8,
+ "ram_style": "block"
+ },
+ "Vector_Vector_Activate_Batch_4": {
+ "PE": 8,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_21": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_9": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_22": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_54": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_5": {
+ "PE": 32,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_23": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_10": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_57": {
+ "ram_style": "ultra",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "FMPadding_Batch_5": {
+ "SIMD": 1
+ },
+ "StreamingDataWidthConverter_Batch_24": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_59": {
+ "ram_style": "ultra",
+ "depth": 8192,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_6": {
+ "SIMD": 2,
+ "ram_style": "block"
+ },
+ "Vector_Vector_Activate_Batch_5": {
+ "PE": 2,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_25": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_11": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_26": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_64": {
+ "ram_style": "auto",
+ "depth": 32,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_6": {
+ "PE": 16,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_27": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_12": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_67": {
+ "ram_style": "ultra",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "FMPadding_Batch_6": {
+ "SIMD": 1
+ },
+ "StreamingDataWidthConverter_Batch_28": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_69": {
+ "ram_style": "ultra",
+ "depth": 4096,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_7": {
+ "SIMD": 4,
+ "ram_style": "block"
+ },
+ "Vector_Vector_Activate_Batch_6": {
+ "PE": 4,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_29": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_13": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_30": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_74": {
+ "ram_style": "auto",
+ "depth": 64,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_7": {
+ "PE": 32,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_31": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_14": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_77": {
+ "ram_style": "ultra",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "FMPadding_Batch_7": {
+ "SIMD": 1
+ },
+ "StreamingDataWidthConverter_Batch_32": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_79": {
+ "ram_style": "ultra",
+ "depth": 4096,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_8": {
+ "SIMD": 4,
+ "ram_style": "block"
+ },
+ "Vector_Vector_Activate_Batch_7": {
+ "PE": 4,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_33": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_15": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_34": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_84": {
+ "ram_style": "auto",
+ "depth": 64,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_8": {
+ "PE": 32,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_35": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_16": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_87": {
+ "ram_style": "ultra",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "FMPadding_Batch_8": {
+ "SIMD": 1
+ },
+ "StreamingDataWidthConverter_Batch_36": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_89": {
+ "ram_style": "ultra",
+ "depth": 4096,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_9": {
+ "SIMD": 4,
+ "ram_style": "block"
+ },
+ "Vector_Vector_Activate_Batch_8": {
+ "PE": 4,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_37": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_17": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_38": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_94": {
+ "ram_style": "auto",
+ "depth": 64,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_9": {
+ "PE": 32,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_39": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_18": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_97": {
+ "ram_style": "ultra",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "FMPadding_Batch_9": {
+ "SIMD": 1
+ },
+ "StreamingDataWidthConverter_Batch_40": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_99": {
+ "ram_style": "ultra",
+ "depth": 4096,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_10": {
+ "SIMD": 4,
+ "ram_style": "block"
+ },
+ "Vector_Vector_Activate_Batch_9": {
+ "PE": 4,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_41": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_19": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_42": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_104": {
+ "ram_style": "auto",
+ "depth": 64,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_10": {
+ "PE": 32,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_43": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_20": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_107": {
+ "ram_style": "ultra",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "FMPadding_Batch_10": {
+ "SIMD": 1
+ },
+ "StreamingDataWidthConverter_Batch_44": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_109": {
+ "ram_style": "ultra",
+ "depth": 4096,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_11": {
+ "SIMD": 4,
+ "ram_style": "block"
+ },
+ "Vector_Vector_Activate_Batch_10": {
+ "PE": 4,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_45": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_21": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_46": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_114": {
+ "ram_style": "auto",
+ "depth": 64,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_11": {
+ "PE": 32,
+ "SIMD": 8,
+ "ram_style": "auto",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_47": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_22": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_117": {
+ "ram_style": "ultra",
+ "depth": 512,
+ "impl_style": "vivado"
+ },
+ "FMPadding_Batch_11": {
+ "SIMD": 1
+ },
+ "StreamingFIFO_118": {
+ "ram_style": "ultra",
+ "depth": 16384,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_12": {
+ "SIMD": 1,
+ "ram_style": "block"
+ },
+ "Vector_Vector_Activate_Batch_11": {
+ "PE": 1,
+ "resType": "lut"
+ },
+ "Thresholding_Batch_23": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_48": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_122": {
+ "ram_style": "auto",
+ "depth": 64,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_12": {
+ "PE": 16,
+ "SIMD": 8,
+ "ram_style": "ultra",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 1
+ },
+ "StreamingDataWidthConverter_Batch_49": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_24": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingFIFO_125": {
+ "ram_style": "ultra",
+ "depth": 1024,
+ "impl_style": "vivado"
+ },
+ "FMPadding_Batch_12": {
+ "SIMD": 1
+ },
+ "StreamingDataWidthConverter_Batch_50": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_127": {
+ "ram_style": "ultra",
+ "depth": 16384,
+ "impl_style": "vivado"
+ },
+ "ConvolutionInputGenerator_13": {
+ "SIMD": 2,
+ "ram_style": "block"
+ },
+ "Vector_Vector_Activate_Batch_12": {
+ "PE": 2,
+ "resType": "lut"
+ },
+ "StreamingDataWidthConverter_Batch_51": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_25": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "StreamingDataWidthConverter_Batch_52": {
+ "impl_style": "hls"
+ },
+ "StreamingFIFO_132": {
+ "ram_style": "auto",
+ "depth": 128,
+ "impl_style": "rtl"
+ },
+ "StreamingFCLayer_Batch_13": {
+ "PE": 32,
+ "SIMD": 8,
+ "ram_style": "ultra",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 1
+ },
+ "StreamingDataWidthConverter_Batch_53": {
+ "impl_style": "hls"
+ },
+ "Thresholding_Batch_26": {
+ "PE": 1,
+ "ram_style": "distributed",
+ "mem_mode": "const",
+ "runtime_writeable_weights": 0
+ },
+ "ConvolutionInputGenerator_14": {
+ "SIMD": 1,
+ "ram_style": "block"
+ },
+ "Pool_Batch_0": {
+ "PE": 1
+ },
+ "StreamingDataWidthConverter_Batch_54": {
+ "impl_style": "hls"
+ },
+ "StreamingFCLayer_Batch_14": {
+ "PE": 1,
+ "SIMD": 16,
+ "ram_style": "ultra",
+ "resType": "lut",
+ "mem_mode": "decoupled",
+ "runtime_writeable_weights": 1
+ },
+ "ChannelwiseOp_Batch_0": {
+ "PE": 1,
+ "ram_style": "distributed"
+ },
+ "LabelSelect_Batch_0": {
+ "PE": 1
+ }
+}
diff --git a/finn_examples/bitfiles/bitfiles.zip.link b/finn_examples/bitfiles/bitfiles.zip.link
index b9e0520..6c05a3a 100644
--- a/finn_examples/bitfiles/bitfiles.zip.link
+++ b/finn_examples/bitfiles/bitfiles.zip.link
@@ -12,11 +12,11 @@
"md5sum": "59598d7f36ffdc74a0a0262f5b67423c"
},
"ZCU104": {
- "url": "https://github.com/Xilinx/finn-examples/releases/download/v0.0.1a/ZCU104.zip",
- "md5sum": "cdc1b757a059b0bb2b7270b3081ae52e"
+ "url": "https://github.com/Xilinx/finn-examples/releases/download/mnv1-zcu104/ZCU104.zip",
+ "md5sum": "1ed10d74e85eec70fd094b2947b5b8e3"
},
"xilinx_u250_xdma_201830_2": {
- "url": "https://github.com/Xilinx/finn-examples/releases/download/v0.0.1a/xilinx_u250_xdma_201830_2.zip",
- "md5sum": "5e8f3625fcf14aaa4fc7416fd9f15450"
+ "url": "https://github.com/Xilinx/finn-examples/releases/download/mnv1-u250-partitioned/xilinx_u250_xdma_201830_2.zip",
+ "md5sum": "d8c7d67c688f3471b6e2c53762b8b258"
}
}
diff --git a/finn_examples/driver.py b/finn_examples/driver.py
index 24f9f11..4dd5a08 100644
--- a/finn_examples/driver.py
+++ b/finn_examples/driver.py
@@ -32,6 +32,8 @@
from pynq import Overlay, allocate
from pynq.ps import Clocks
+from finn.core.datatype import DataType
+from finn.util.basic import gen_finn_dt_tensor
from finn.util.data_packing import (
finnpy_to_packed_bytearray,
packed_bytearray_to_finnpy,
@@ -84,25 +86,78 @@ def __init__(
self.batch_size = batch_size
self.fclk_mhz = fclk_mhz
if self.platform == "alveo":
- self.idma = self.idma0
+ if "input_dma_name" in io_shape_dict.keys():
+ self.idma = getattr(self, io_shape_dict["input_dma_name"])
+ else:
+ self.idma = self.idma0
self.odma = self.odma0
self.odma_handle = None
elif self.platform == "zynq-iodma":
- self.idma = self.idma0
+ if "input_dma_name" in io_shape_dict.keys():
+ self.idma = getattr(self, io_shape_dict["input_dma_name"])
+ else:
+ self.idma = self.idma0
self.odma = self.odma0
# set the clock frequency as specified by user during transformations
if self.fclk_mhz > 0:
Clocks.fclk0_mhz = self.fclk_mhz
else:
raise ValueError("Supported platforms are zynq-iodma alveo")
- # load any runtime weights
+ # load any external + runtime weights
+ self.load_external_weights()
self.load_runtime_weights()
+ def load_external_weights(self):
+ """Load any existing external (DRAM) weights from the specified dir into the
+ appropriate layer of the accelerator. Note that this must be enabled
+ during the accelerator build process. The weights directory
+ is specified as the class member ``runtime_weight_dir``. External (DRAM)
+ weights are one .npy file per layer.
+ """
+
+ self.external_weights = []
+ w_filenames = []
+ if not os.path.isdir(self.runtime_weight_dir):
+ return
+ for (dirpath, dirnames, filenames) in os.walk(self.runtime_weight_dir):
+ w_filenames.extend(filenames)
+
+ tmp_weight_dict = {}
+
+ for w_filename in w_filenames:
+ if w_filename.endswith(".npy"):
+ weight_tensor = np.load(self.runtime_weight_dir + "/" + w_filename)
+ else:
+ continue
+
+ idma_name = w_filename.split(".")[0]
+ tmp_weight_dict[idma_name] = weight_tensor
+
+ for idma_name in tmp_weight_dict.keys():
+ if idma_name in self.ip_dict.keys():
+ iwdma = getattr(self, idma_name)
+ weight_tensor = tmp_weight_dict[idma_name]
+ weight_buf = allocate(weight_tensor.shape, dtype=np.uint8)
+ weight_buf[:] = weight_tensor
+ # weight_buf.sync_to_device()
+ weight_buf.flush()
+
+ self.external_weights += [(iwdma, weight_buf, idma_name)]
+
+ if "number_of_external_weights" in self._io_shape_dict:
+ hw_ext_weights = self._io_shape_dict["number_of_external_weights"]
+ assert len(self.external_weights) == hw_ext_weights, (
+ "Number of hardware external weights and number of external "
+ + "weight tensors available do not match. \n"
+ + "Is runtime_weight_dir pointing to the correct folder?"
+ )
+
def load_runtime_weights(self, flush_accel=True, verify=True):
- """Load any existing runtime weights from the specified dir into the
+ """Load any existing runtime-writable weights from the specified dir into the
appropriate layer of the accelerator. Note that this must be enabled
during the accelerator build process. The runtime weights directory
- is specified as the class member ``runtime_weight_dir``.
+ is specified as the class member ``runtime_weight_dir``. Runtime-writable
+ weights are provided as one .dat file per layer.
Parameters
----------
@@ -122,18 +177,25 @@ def load_runtime_weights(self, flush_accel=True, verify=True):
if w_filename.endswith(".dat"):
with open(self.runtime_weight_dir + "/" + w_filename, "r") as f:
dat = f.read()
+ else:
+ continue
layer_w = np.fromiter(
[int(x, 16) for x in dat.strip().split()], dtype=np.uint32
)
- layer_ind = int(w_filename.split("_")[0])
- rt_weight_dict[layer_ind] = layer_w
- for layer_ind in rt_weight_dict.keys():
- cand_if_name = "StreamingDataflowPartition_1/s_axilite_%d" % layer_ind
+ sdp_ind = int(w_filename.split("_")[0])
+ layer_ind = int(w_filename.split("_")[1])
+ rt_weight_dict[(sdp_ind, layer_ind)] = layer_w
+ for sdp_ind, layer_ind in rt_weight_dict.keys():
+ cand_if_name = "StreamingDataflowPartition_%d/s_axilite_%d" % (
+ sdp_ind,
+ layer_ind,
+ )
if cand_if_name in self.ip_dict.keys():
layer_mmio = getattr(
- self.StreamingDataflowPartition_1, "s_axilite_%d" % layer_ind
+ getattr(self, "StreamingDataflowPartition_%d" % sdp_ind),
+ "s_axilite_%d" % layer_ind,
).mmio
- layer_w = rt_weight_dict[layer_ind]
+ layer_w = rt_weight_dict[(sdp_ind, layer_ind)]
layer_mmio.write_mm(0, layer_w.tobytes())
if verify:
new_w = np.copy(layer_mmio.array[: layer_w.shape[0]])
@@ -278,6 +340,10 @@ def execute_on_buffers(self, asynch=False, batch_size=None):
if self.platform == "zynq-iodma":
assert self.odma.read(0x00) & 0x4 != 0, "Output DMA is not idle"
# manually launch IODMAs since signatures are missing
+ for iwdma, iwbuf, iwdma_name in self.external_weights:
+ iwdma.write(0x10, iwbuf.device_address)
+ iwdma.write(0x1C, batch_size)
+ iwdma.write(0x00, 1)
self.idma.write(0x10, self.ibuf_packed_device.device_address)
self.idma.write(0x1C, batch_size)
self.odma.write(0x10, self.obuf_packed_device.device_address)
@@ -287,6 +353,8 @@ def execute_on_buffers(self, asynch=False, batch_size=None):
elif self.platform == "alveo":
assert self.odma_handle is None, "Output DMA is already running"
self.idma.start(self.ibuf_packed_device, batch_size)
+ for iwdma, iwbuf, iwdma_name in self.external_weights:
+ iwdma.start(iwbuf, batch_size)
self.odma_handle = self.odma.start(self.obuf_packed_device, batch_size)
else:
raise Exception("Unrecognized platform: %s" % self.platform)
@@ -338,46 +406,55 @@ def throughput_test(self):
res["DRAM_out_bandwidth[Mb/s]"] = (
np.prod(self.oshape_packed) * 0.000001 / runtime
)
- if self.platform != "alveo":
+ for iwdma, iwbuf, iwdma_name in self.external_weights:
+ res["DRAM_extw_%s_bandwidth[Mb/s]" % iwdma_name] = (
+ self.batch_size * np.prod(iwbuf.shape) * 0.000001 / runtime
+ )
+ if self.platform == "zynq-iodma":
res["fclk[mhz]"] = Clocks.fclk0_mhz
- else:
- res["fclk[mhz]"] = self.fclk_mhz
+ elif self.platform == "alveo":
+ res["fclk[mhz]"] = self.clock_dict["clock0"]["frequency"]
res["batch_size"] = self.batch_size
# also benchmark driver-related overheads
- input_npy = np.zeros(self.ishape_normal, dtype=self.idt.to_numpy_dt())
+ input_npy = gen_finn_dt_tensor(self.idt, self.ishape_normal)
+ # provide as int8/uint8 to support fast packing path where possible
+ if self.idt == DataType.UINT8:
+ input_npy = input_npy.astype(np.uint8)
+ elif self.idt == DataType.INT8:
+ input_npy = input_npy.astype(np.int8)
start = time.time()
ibuf_folded = self.fold_input(input_npy)
end = time.time()
runtime = end - start
- res["fold_input[ms]"] = runtime
+ res["fold_input[ms]"] = runtime * 1000
start = time.time()
ibuf_packed = self.pack_input(ibuf_folded)
end = time.time()
runtime = end - start
- res["pack_input[ms]"] = runtime
+ res["pack_input[ms]"] = runtime * 1000
start = time.time()
self.copy_input_data_to_device(ibuf_packed)
end = time.time()
runtime = end - start
- res["copy_input_data_to_device[ms]"] = runtime
+ res["copy_input_data_to_device[ms]"] = runtime * 1000
start = time.time()
self.copy_output_data_from_device(self.obuf_packed)
end = time.time()
runtime = end - start
- res["copy_output_data_from_device[ms]"] = runtime
+ res["copy_output_data_from_device[ms]"] = runtime * 1000
start = time.time()
obuf_folded = self.unpack_output(self.obuf_packed)
end = time.time()
runtime = end - start
- res["unpack_output[ms]"] = runtime
+ res["unpack_output[ms]"] = runtime * 1000
start = time.time()
self.unfold_output(obuf_folded)
end = time.time()
runtime = end - start
- res["unfold_output[ms]"] = runtime
+ res["unfold_output[ms]"] = runtime * 1000
return res
diff --git a/finn_examples/models.py b/finn_examples/models.py
index fc5da8e..6e6e147 100644
--- a/finn_examples/models.py
+++ b/finn_examples/models.py
@@ -104,6 +104,28 @@ def find_bitfile(model_name, target_platform):
)
+def find_runtime_weights(model_name, target_platform):
+ weight_dir = "%s_runtime_weights" % (model_name)
+ weight_dir_candidates = [
+ pk.resource_filename(
+ "finn_examples", "bitfiles/%s/%s" % (target_platform, weight_dir)
+ ),
+ pk.resource_filename(
+ "finn_examples",
+ "bitfiles/bitfiles.zip.d/%s/%s" % (target_platform, weight_dir),
+ ),
+ ]
+ for candidate in weight_dir_candidates:
+ if os.path.isdir(candidate):
+ weight_files = os.listdir(candidate)
+ if weight_files:
+ return candidate
+ raise Exception(
+ "Runtime weights for model = %s target platform = %s not found. Looked in: %s"
+ % (model_name, target_platform, str(weight_dir_candidates))
+ )
+
+
def get_driver_mode():
driver_modes = {"edge": "zynq-iodma", "pcie": "alveo"}
return driver_modes[get_edge_or_pcie()]
@@ -170,4 +192,16 @@ def mobilenetv1_w4a4_imagenet(target_platform=None):
driver_mode = get_driver_mode()
model_name = "mobilenetv1-w4a4"
filename = find_bitfile(model_name, target_platform)
- return FINNExampleOverlay(filename, driver_mode, _imagenet_top5inds_io_shape_dict)
+ if target_platform in ["ZCU104"]:
+ runtime_weight_dir = find_runtime_weights(model_name, target_platform)
+ else:
+ runtime_weight_dir = None
+ # target 185 MHz for Zynq (this is ignored for Alveo)
+ fclk_mhz = 185.0
+ return FINNExampleOverlay(
+ filename,
+ driver_mode,
+ _imagenet_top5inds_io_shape_dict,
+ runtime_weight_dir=runtime_weight_dir,
+ fclk_mhz=fclk_mhz,
+ )
diff --git a/finn_examples/notebooks/2_imagenet_with_mobilenet_v1.ipynb b/finn_examples/notebooks/2_imagenet_with_mobilenet_v1.ipynb
old mode 100644
new mode 100755
index 327d701..4574435
--- a/finn_examples/notebooks/2_imagenet_with_mobilenet_v1.ipynb
+++ b/finn_examples/notebooks/2_imagenet_with_mobilenet_v1.ipynb
@@ -69,7 +69,9 @@
"metadata": {},
"outputs": [],
"source": [
- "accel = models.mobilenetv1_w4a4_imagenet()"
+ "accel = models.mobilenetv1_w4a4_imagenet()\n",
+ "#some systems might require a manual platform setting:\n",
+ "#accel = models.mobilenetv1_w4a4_imagenet(\"ZCU102\")"
]
},
{
@@ -91,87 +93,94 @@
"print(\"Expected output shape and datatype: %s %s\" % (str(accel.oshape_normal), str(accel.odt)))"
]
},
- {
- "cell_type": "code",
- "execution_count": 4,
- "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "WARNING: pip is being invoked by an old script wrapper. This will fail in a future version of pip.\n",
- "Please see https://github.com/pypa/pip/issues/5599 for advice on fixing the underlying issue.\n",
- "To avoid this problem you can invoke Python with '-m pip' instead of running pip directly.\n",
- "Requirement already satisfied: torchvision in /scratch/users/yamanu/conda/finn-examples/lib/python3.7/site-packages (0.8.2)\n",
- "Requirement already satisfied: pillow>=4.1.1 in /scratch/users/yamanu/conda/finn-examples/lib/python3.7/site-packages (from torchvision) (7.0.0)\n",
- "Requirement already satisfied: numpy in /scratch/users/yamanu/conda/finn-examples/lib/python3.7/site-packages (from torchvision) (1.18.1)\n",
- "Requirement already satisfied: torch==1.7.1 in /scratch/users/yamanu/conda/finn-examples/lib/python3.7/site-packages (from torchvision) (1.7.1)\n",
- "Requirement already satisfied: typing-extensions in /scratch/users/yamanu/conda/finn-examples/lib/python3.7/site-packages (from torch==1.7.1->torchvision) (3.7.4.3)\n"
- ]
- }
- ],
- "source": [
- "! pip install torchvision"
- ]
- },
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "# Load the ImageNet validation dataset"
+ "# Prepare loading of ImageNet validation dataset"
]
},
{
"cell_type": "code",
- "execution_count": 5,
+ "execution_count": 4,
"metadata": {},
"outputs": [
{
- "data": {
- "text/plain": [
- "'/proj/xlabs_t3/users/ml-workspace/datasets/imagenet/raw-images/imagenet_symlink/val'"
- ]
- },
- "execution_count": 5,
- "metadata": {},
- "output_type": "execute_result"
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "/proj/xlabs_t3/users/ml-workspace/datasets/imagenet/raw-images/imagenet_symlink/val\n"
+ ]
}
],
"source": [
- "import torchvision.transforms as transforms\n",
- "import torchvision.datasets as datasets\n",
- "import torch\n",
"import numpy as np\n",
+ "from PIL import Image\n",
+ "from dataset_loading import FileQueue, ImgQueue\n",
"import os\n",
"\n",
- "os.environ[\"IMAGENET_VAL_PATH\"]"
+ "# 2 ways to provide the data:\n",
+ "# without a label file: expect images in 1000 sorted subfolders\n",
+ "# with a label file: expect images directly in val directory\n",
+ "val_dir = os.environ[\"IMAGENET_VAL_PATH\"]\n",
+ "label_file = None\n",
+ "print(val_dir)"
]
},
{
"cell_type": "code",
- "execution_count": 6,
- "metadata": {},
- "outputs": [],
- "source": [
- "valdir = os.environ[\"IMAGENET_VAL_PATH\"]\n",
- "batch_size = 1\n",
- "val_loader = torch.utils.data.DataLoader(\n",
- " datasets.ImageFolder(valdir, transforms.Compose([\n",
- " transforms.Resize(256),\n",
- " transforms.CenterCrop(224),\n",
- " transforms.Lambda(lambda x: np.array(x, dtype=np.uint8))\n",
- " ])),\n",
- " batch_size=batch_size, shuffle=False, num_workers=0, pin_memory=True)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 7,
+ "execution_count": 5,
"metadata": {},
"outputs": [],
"source": [
- "test_single_x, test_single_y = val_loader.sampler.data_source.__getitem__(0)"
+ "def img_resize(img, size):\n",
+ " w, h = img.size\n",
+ " if (w <= h and w == size) or (h <= w and h == size):\n",
+ " return img\n",
+ " if w < h:\n",
+ " ow = size\n",
+ " oh = int(size * h / w)\n",
+ " return img.resize((ow, oh), Image.BILINEAR)\n",
+ " else:\n",
+ " oh = size\n",
+ " ow = int(size * w / h)\n",
+ " return img.resize((ow, oh), Image.BILINEAR)\n",
+ "\n",
+ "def img_center_crop(img, size):\n",
+ " crop_height, crop_width = (size, size)\n",
+ " image_width, image_height = img.size\n",
+ " crop_top = int(round((image_height - crop_height) / 2.))\n",
+ " crop_left = int(round((image_width - crop_width) / 2.))\n",
+ " return img.crop((crop_left, crop_top, crop_left + crop_width, crop_top + crop_height))\n",
+ "\n",
+ "def pre_process(img_np):\n",
+ " img = Image.fromarray(img_np.astype(np.uint8))\n",
+ " img = img_resize(img, 256)\n",
+ " img = img_center_crop(img, 224)\n",
+ " img = np.array(img, dtype=np.uint8)\n",
+ " return img\n",
+ "\n",
+ "def setup_dataloader(val_path, label_file_path = None, batch_size=100, n_images = 50000):\n",
+ " if label_file_path is None:\n",
+ " val_folders = [ f.name for f in os.scandir(val_path) if f.is_dir() ]\n",
+ " assert len(val_folders) == 1000, \"Expected 1000 subfolders in ILSVRC2012 val\"\n",
+ " files = []\n",
+ " labels = []\n",
+ " for idx, folder in enumerate(val_folders):\n",
+ " current_files = sorted(os.listdir(os.path.join(val_path, folder)))\n",
+ " current_files = [os.path.join(folder, file) for file in current_files]\n",
+ " files.extend(current_files)\n",
+ " labels.extend([idx]*len(current_files))\n",
+ " files = files[:n_images]\n",
+ " else:\n",
+ " files = ['ILSVRC2012_val_{:08d}.JPEG'.format(i) for i in range(1,n_images+1)]\n",
+ " labels = np.loadtxt(label_file_path, dtype=int, usecols=1)\n",
+ "\n",
+ " file_queue = FileQueue()\n",
+ " file_queue.load_epochs(list(zip(files,labels)), shuffle=False)\n",
+ " img_queue = ImgQueue(maxsize=batch_size)\n",
+ " img_queue.start_loaders(file_queue, num_threads=4, img_dir=val_path, transform=pre_process)\n",
+ " return img_queue"
]
},
{
@@ -183,12 +192,12 @@
},
{
"cell_type": "code",
- "execution_count": 8,
+ "execution_count": 6,
"metadata": {},
"outputs": [
{
"data": {
- "image/png": "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\n",
+ "image/png": "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\n",
"text/plain": [
""
]
@@ -200,15 +209,20 @@
}
],
"source": [
+ "%matplotlib inline\n",
"from matplotlib import pyplot as plt\n",
"\n",
- "plt.imshow(test_single_x.reshape(224,224,3))\n",
+ "img_queue = setup_dataloader(val_dir, label_file, 1, 1)\n",
+ "\n",
+ "test_single_x, test_single_y = img_queue.get()\n",
+ "\n",
+ "plt.imshow(test_single_x)\n",
"plt.show()"
]
},
{
"cell_type": "code",
- "execution_count": 9,
+ "execution_count": 7,
"metadata": {},
"outputs": [
{
@@ -217,7 +231,7 @@
"0"
]
},
- "execution_count": 9,
+ "execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
@@ -228,7 +242,7 @@
},
{
"cell_type": "code",
- "execution_count": 10,
+ "execution_count": 8,
"metadata": {},
"outputs": [
{
@@ -246,7 +260,7 @@
},
{
"cell_type": "code",
- "execution_count": 11,
+ "execution_count": 9,
"metadata": {},
"outputs": [],
"source": [
@@ -255,14 +269,14 @@
},
{
"cell_type": "code",
- "execution_count": 12,
+ "execution_count": 10,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
- "Top-5 classes predicted by the accelerator: [[[[391. 48. 0. 39. 395.]]]]\n"
+ "Top-5 classes predicted by the accelerator: [[[[391. 0. 395. 394. 48.]]]]\n"
]
}
],
@@ -272,14 +286,14 @@
},
{
"cell_type": "code",
- "execution_count": 13,
+ "execution_count": 11,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
- "2.71 ms ± 22.8 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)\n"
+ "2.18 ms ± 6.29 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)\n"
]
}
],
@@ -297,102 +311,561 @@
},
{
"cell_type": "code",
- "execution_count": 14,
+ "execution_count": 12,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
- "Accelerator buffer shapes are (1000, 224, 224, 1, 3) for input, (1000, 1, 1, 1, 10) for output\n"
+ "Accelerator buffer shapes are (100, 224, 224, 1, 3) for input, (100, 1, 1, 1, 10) for output\n"
]
}
],
"source": [
- "import numpy as np\n",
- "\n",
- "batch_size = 1000\n",
+ "batch_size = 100\n",
"accel.batch_size = batch_size\n",
- "print(\"Accelerator buffer shapes are %s for input, %s for output\" % (str(accel.ishape_packed), str(accel.oshape_packed)) )\n",
- "obuf_packed = np.empty_like(accel.obuf_packed_device)\n",
- "val_loader = torch.utils.data.DataLoader(\n",
- " datasets.ImageFolder(valdir, transforms.Compose([\n",
- " transforms.Resize(256),\n",
- " transforms.CenterCrop(224),\n",
- " transforms.Lambda(lambda x: np.array(x, dtype=np.uint8))\n",
- " ])),\n",
- " batch_size=batch_size, shuffle=False, num_workers=16, pin_memory=True)"
+ "print(\"Accelerator buffer shapes are %s for input, %s for output\" % (str(accel.ishape_packed), str(accel.oshape_packed)) )"
]
},
{
"cell_type": "code",
- "execution_count": 15,
- "metadata": {},
+ "execution_count": 13,
+ "metadata": {
+ "scrolled": true
+ },
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
- "batch 1 : total OK 844 NOK 156\n",
- "batch 2 : total OK 1589 NOK 411\n",
- "batch 3 : total OK 2296 NOK 704\n",
- "batch 4 : total OK 2967 NOK 1033\n",
- "batch 5 : total OK 3842 NOK 1158\n",
- "batch 6 : total OK 4622 NOK 1378\n",
- "batch 7 : total OK 5437 NOK 1563\n",
- "batch 8 : total OK 6247 NOK 1753\n",
- "batch 9 : total OK 6949 NOK 2051\n",
- "batch 10 : total OK 7675 NOK 2325\n",
- "batch 11 : total OK 8445 NOK 2555\n",
- "batch 12 : total OK 9172 NOK 2828\n",
- "batch 13 : total OK 9935 NOK 3065\n",
- "batch 14 : total OK 10671 NOK 3329\n",
- "batch 15 : total OK 11444 NOK 3556\n",
- "batch 16 : total OK 12173 NOK 3827\n",
- "batch 17 : total OK 13030 NOK 3970\n",
- "batch 18 : total OK 13765 NOK 4235\n",
- "batch 19 : total OK 14550 NOK 4450\n",
- "batch 20 : total OK 15292 NOK 4708\n",
- "batch 21 : total OK 15973 NOK 5027\n",
- "batch 22 : total OK 16643 NOK 5357\n",
- "batch 23 : total OK 17294 NOK 5706\n",
- "batch 24 : total OK 17922 NOK 6078\n",
- "batch 25 : total OK 18528 NOK 6472\n",
- "batch 26 : total OK 19139 NOK 6861\n",
- "batch 27 : total OK 19806 NOK 7194\n",
- "batch 28 : total OK 20487 NOK 7513\n",
- "batch 29 : total OK 21251 NOK 7749\n",
- "batch 30 : total OK 21891 NOK 8109\n",
- "batch 31 : total OK 22590 NOK 8410\n",
- "batch 32 : total OK 23150 NOK 8850\n",
- "batch 33 : total OK 23804 NOK 9196\n",
- "batch 34 : total OK 24450 NOK 9550\n",
- "batch 35 : total OK 25115 NOK 9885\n",
- "batch 36 : total OK 25799 NOK 10201\n",
- "batch 37 : total OK 26470 NOK 10530\n",
- "batch 38 : total OK 27106 NOK 10894\n",
- "batch 39 : total OK 27777 NOK 11223\n",
- "batch 40 : total OK 28422 NOK 11578\n",
- "batch 41 : total OK 29092 NOK 11908\n",
- "batch 42 : total OK 29712 NOK 12288\n",
- "batch 43 : total OK 30363 NOK 12637\n",
- "batch 44 : total OK 31033 NOK 12967\n",
- "batch 45 : total OK 31664 NOK 13336\n",
- "batch 46 : total OK 32305 NOK 13695\n",
- "batch 47 : total OK 33019 NOK 13981\n",
- "batch 48 : total OK 33809 NOK 14191\n",
- "batch 49 : total OK 34391 NOK 14609\n",
- "batch 50 : total OK 35206 NOK 14794\n"
+ "batch 1 : total OK 88 NOK 12\n",
+ "batch 2 : total OK 164 NOK 36\n",
+ "batch 3 : total OK 241 NOK 59\n",
+ "batch 4 : total OK 311 NOK 89\n",
+ "batch 5 : total OK 403 NOK 97\n",
+ "batch 6 : total OK 491 NOK 109\n",
+ "batch 7 : total OK 580 NOK 120\n",
+ "batch 8 : total OK 670 NOK 130\n",
+ "batch 9 : total OK 757 NOK 143\n",
+ "batch 10 : total OK 846 NOK 154\n",
+ "batch 11 : total OK 927 NOK 173\n",
+ "batch 12 : total OK 1018 NOK 182\n",
+ "batch 13 : total OK 1110 NOK 190\n",
+ "batch 14 : total OK 1180 NOK 220\n",
+ "batch 15 : total OK 1262 NOK 238\n",
+ "batch 16 : total OK 1337 NOK 263\n",
+ "batch 17 : total OK 1395 NOK 305\n",
+ "batch 18 : total OK 1454 NOK 346\n",
+ "batch 19 : total OK 1521 NOK 379\n",
+ "batch 20 : total OK 1590 NOK 410\n",
+ "batch 21 : total OK 1650 NOK 450\n",
+ "batch 22 : total OK 1728 NOK 472\n",
+ "batch 23 : total OK 1798 NOK 502\n",
+ "batch 24 : total OK 1861 NOK 539\n",
+ "batch 25 : total OK 1933 NOK 567\n",
+ "batch 26 : total OK 2019 NOK 581\n",
+ "batch 27 : total OK 2093 NOK 607\n",
+ "batch 28 : total OK 2145 NOK 655\n",
+ "batch 29 : total OK 2225 NOK 675\n",
+ "batch 30 : total OK 2294 NOK 706\n",
+ "batch 31 : total OK 2352 NOK 748\n",
+ "batch 32 : total OK 2412 NOK 788\n",
+ "batch 33 : total OK 2474 NOK 826\n",
+ "batch 34 : total OK 2529 NOK 871\n",
+ "batch 35 : total OK 2593 NOK 907\n",
+ "batch 36 : total OK 2677 NOK 923\n",
+ "batch 37 : total OK 2750 NOK 950\n",
+ "batch 38 : total OK 2817 NOK 983\n",
+ "batch 39 : total OK 2895 NOK 1005\n",
+ "batch 40 : total OK 2965 NOK 1035\n",
+ "batch 41 : total OK 3051 NOK 1049\n",
+ "batch 42 : total OK 3138 NOK 1062\n",
+ "batch 43 : total OK 3226 NOK 1074\n",
+ "batch 44 : total OK 3310 NOK 1090\n",
+ "batch 45 : total OK 3402 NOK 1098\n",
+ "batch 46 : total OK 3491 NOK 1109\n",
+ "batch 47 : total OK 3580 NOK 1120\n",
+ "batch 48 : total OK 3672 NOK 1128\n",
+ "batch 49 : total OK 3756 NOK 1144\n",
+ "batch 50 : total OK 3836 NOK 1164\n",
+ "batch 51 : total OK 3915 NOK 1185\n",
+ "batch 52 : total OK 3997 NOK 1203\n",
+ "batch 53 : total OK 4079 NOK 1221\n",
+ "batch 54 : total OK 4154 NOK 1246\n",
+ "batch 55 : total OK 4230 NOK 1270\n",
+ "batch 56 : total OK 4309 NOK 1291\n",
+ "batch 57 : total OK 4383 NOK 1317\n",
+ "batch 58 : total OK 4458 NOK 1342\n",
+ "batch 59 : total OK 4538 NOK 1362\n",
+ "batch 60 : total OK 4606 NOK 1394\n",
+ "batch 61 : total OK 4677 NOK 1423\n",
+ "batch 62 : total OK 4756 NOK 1444\n",
+ "batch 63 : total OK 4817 NOK 1483\n",
+ "batch 64 : total OK 4891 NOK 1509\n",
+ "batch 65 : total OK 4978 NOK 1522\n",
+ "batch 66 : total OK 5067 NOK 1533\n",
+ "batch 67 : total OK 5152 NOK 1548\n",
+ "batch 68 : total OK 5235 NOK 1565\n",
+ "batch 69 : total OK 5326 NOK 1574\n",
+ "batch 70 : total OK 5418 NOK 1582\n",
+ "batch 71 : total OK 5503 NOK 1597\n",
+ "batch 72 : total OK 5589 NOK 1611\n",
+ "batch 73 : total OK 5678 NOK 1622\n",
+ "batch 74 : total OK 5763 NOK 1637\n",
+ "batch 75 : total OK 5853 NOK 1647\n",
+ "batch 76 : total OK 5923 NOK 1677\n",
+ "batch 77 : total OK 6000 NOK 1700\n",
+ "batch 78 : total OK 6081 NOK 1719\n",
+ "batch 79 : total OK 6172 NOK 1728\n",
+ "batch 80 : total OK 6231 NOK 1769\n",
+ "batch 81 : total OK 6314 NOK 1786\n",
+ "batch 82 : total OK 6374 NOK 1826\n",
+ "batch 83 : total OK 6441 NOK 1859\n",
+ "batch 84 : total OK 6490 NOK 1910\n",
+ "batch 85 : total OK 6570 NOK 1930\n",
+ "batch 86 : total OK 6638 NOK 1962\n",
+ "batch 87 : total OK 6709 NOK 1991\n",
+ "batch 88 : total OK 6779 NOK 2021\n",
+ "batch 89 : total OK 6858 NOK 2042\n",
+ "batch 90 : total OK 6934 NOK 2066\n",
+ "batch 91 : total OK 7007 NOK 2093\n",
+ "batch 92 : total OK 7086 NOK 2114\n",
+ "batch 93 : total OK 7158 NOK 2142\n",
+ "batch 94 : total OK 7224 NOK 2176\n",
+ "batch 95 : total OK 7290 NOK 2210\n",
+ "batch 96 : total OK 7368 NOK 2232\n",
+ "batch 97 : total OK 7426 NOK 2274\n",
+ "batch 98 : total OK 7510 NOK 2290\n",
+ "batch 99 : total OK 7581 NOK 2319\n",
+ "batch 100 : total OK 7661 NOK 2339\n",
+ "batch 101 : total OK 7720 NOK 2380\n",
+ "batch 102 : total OK 7805 NOK 2395\n",
+ "batch 103 : total OK 7874 NOK 2426\n",
+ "batch 104 : total OK 7957 NOK 2443\n",
+ "batch 105 : total OK 8038 NOK 2462\n",
+ "batch 106 : total OK 8111 NOK 2489\n",
+ "batch 107 : total OK 8185 NOK 2515\n",
+ "batch 108 : total OK 8269 NOK 2531\n",
+ "batch 109 : total OK 8358 NOK 2542\n",
+ "batch 110 : total OK 8439 NOK 2561\n",
+ "batch 111 : total OK 8514 NOK 2586\n",
+ "batch 112 : total OK 8585 NOK 2615\n",
+ "batch 113 : total OK 8666 NOK 2634\n",
+ "batch 114 : total OK 8722 NOK 2678\n",
+ "batch 115 : total OK 8808 NOK 2692\n",
+ "batch 116 : total OK 8868 NOK 2732\n",
+ "batch 117 : total OK 8939 NOK 2761\n",
+ "batch 118 : total OK 9022 NOK 2778\n",
+ "batch 119 : total OK 9094 NOK 2806\n",
+ "batch 120 : total OK 9165 NOK 2835\n",
+ "batch 121 : total OK 9213 NOK 2887\n",
+ "batch 122 : total OK 9287 NOK 2913\n",
+ "batch 123 : total OK 9376 NOK 2924\n",
+ "batch 124 : total OK 9446 NOK 2954\n",
+ "batch 125 : total OK 9510 NOK 2990\n",
+ "batch 126 : total OK 9579 NOK 3021\n",
+ "batch 127 : total OK 9659 NOK 3041\n",
+ "batch 128 : total OK 9757 NOK 3043\n",
+ "batch 129 : total OK 9832 NOK 3068\n",
+ "batch 130 : total OK 9926 NOK 3074\n",
+ "batch 131 : total OK 10010 NOK 3090\n",
+ "batch 132 : total OK 10097 NOK 3103\n",
+ "batch 133 : total OK 10153 NOK 3147\n",
+ "batch 134 : total OK 10212 NOK 3188\n",
+ "batch 135 : total OK 10292 NOK 3208\n",
+ "batch 136 : total OK 10353 NOK 3247\n",
+ "batch 137 : total OK 10423 NOK 3277\n",
+ "batch 138 : total OK 10516 NOK 3284\n",
+ "batch 139 : total OK 10588 NOK 3312\n",
+ "batch 140 : total OK 10666 NOK 3334\n",
+ "batch 141 : total OK 10731 NOK 3369\n",
+ "batch 142 : total OK 10790 NOK 3410\n",
+ "batch 143 : total OK 10860 NOK 3440\n",
+ "batch 144 : total OK 10938 NOK 3462\n",
+ "batch 145 : total OK 11025 NOK 3475\n",
+ "batch 146 : total OK 11106 NOK 3494\n",
+ "batch 147 : total OK 11197 NOK 3503\n",
+ "batch 148 : total OK 11283 NOK 3517\n",
+ "batch 149 : total OK 11363 NOK 3537\n",
+ "batch 150 : total OK 11439 NOK 3561\n",
+ "batch 151 : total OK 11522 NOK 3578\n",
+ "batch 152 : total OK 11585 NOK 3615\n",
+ "batch 153 : total OK 11658 NOK 3642\n",
+ "batch 154 : total OK 11745 NOK 3655\n",
+ "batch 155 : total OK 11824 NOK 3676\n",
+ "batch 156 : total OK 11889 NOK 3711\n",
+ "batch 157 : total OK 11950 NOK 3750\n",
+ "batch 158 : total OK 12010 NOK 3790\n",
+ "batch 159 : total OK 12098 NOK 3802\n",
+ "batch 160 : total OK 12177 NOK 3823\n",
+ "batch 161 : total OK 12270 NOK 3830\n",
+ "batch 162 : total OK 12359 NOK 3841\n",
+ "batch 163 : total OK 12450 NOK 3850\n",
+ "batch 164 : total OK 12540 NOK 3860\n",
+ "batch 165 : total OK 12613 NOK 3887\n",
+ "batch 166 : total OK 12695 NOK 3905\n",
+ "batch 167 : total OK 12789 NOK 3911\n",
+ "batch 168 : total OK 12871 NOK 3929\n",
+ "batch 169 : total OK 12955 NOK 3945\n",
+ "batch 170 : total OK 13047 NOK 3953\n",
+ "batch 171 : total OK 13119 NOK 3981\n",
+ "batch 172 : total OK 13202 NOK 3998\n",
+ "batch 173 : total OK 13276 NOK 4024\n",
+ "batch 174 : total OK 13363 NOK 4037\n",
+ "batch 175 : total OK 13422 NOK 4078\n",
+ "batch 176 : total OK 13514 NOK 4086\n",
+ "batch 177 : total OK 13579 NOK 4121\n",
+ "batch 178 : total OK 13668 NOK 4132\n",
+ "batch 179 : total OK 13728 NOK 4172\n",
+ "batch 180 : total OK 13785 NOK 4215\n",
+ "batch 181 : total OK 13866 NOK 4234\n",
+ "batch 182 : total OK 13947 NOK 4253\n",
+ "batch 183 : total OK 14037 NOK 4263\n",
+ "batch 184 : total OK 14115 NOK 4285\n",
+ "batch 185 : total OK 14188 NOK 4312\n",
+ "batch 186 : total OK 14261 NOK 4339\n",
+ "batch 187 : total OK 14335 NOK 4365\n",
+ "batch 188 : total OK 14401 NOK 4399\n",
+ "batch 189 : total OK 14489 NOK 4411\n",
+ "batch 190 : total OK 14564 NOK 4436\n",
+ "batch 191 : total OK 14610 NOK 4490\n",
+ "batch 192 : total OK 14680 NOK 4520\n",
+ "batch 193 : total OK 14747 NOK 4553\n",
+ "batch 194 : total OK 14828 NOK 4572\n",
+ "batch 195 : total OK 14914 NOK 4586\n",
+ "batch 196 : total OK 14979 NOK 4621\n",
+ "batch 197 : total OK 15067 NOK 4633\n",
+ "batch 198 : total OK 15134 NOK 4666\n",
+ "batch 199 : total OK 15225 NOK 4675\n",
+ "batch 200 : total OK 15306 NOK 4694\n",
+ "batch 201 : total OK 15371 NOK 4729\n",
+ "batch 202 : total OK 15434 NOK 4766\n",
+ "batch 203 : total OK 15516 NOK 4784\n",
+ "batch 204 : total OK 15595 NOK 4805\n",
+ "batch 205 : total OK 15664 NOK 4836\n",
+ "batch 206 : total OK 15742 NOK 4858\n",
+ "batch 207 : total OK 15797 NOK 4903\n",
+ "batch 208 : total OK 15834 NOK 4966\n",
+ "batch 209 : total OK 15920 NOK 4980\n",
+ "batch 210 : total OK 15973 NOK 5027\n",
+ "batch 211 : total OK 16048 NOK 5052\n",
+ "batch 212 : total OK 16108 NOK 5092\n",
+ "batch 213 : total OK 16183 NOK 5117\n",
+ "batch 214 : total OK 16258 NOK 5142\n",
+ "batch 215 : total OK 16328 NOK 5172\n",
+ "batch 216 : total OK 16410 NOK 5190\n",
+ "batch 217 : total OK 16476 NOK 5224\n",
+ "batch 218 : total OK 16520 NOK 5280\n",
+ "batch 219 : total OK 16587 NOK 5313\n",
+ "batch 220 : total OK 16649 NOK 5351\n",
+ "batch 221 : total OK 16719 NOK 5381\n",
+ "batch 222 : total OK 16768 NOK 5432\n",
+ "batch 223 : total OK 16839 NOK 5461\n",
+ "batch 224 : total OK 16893 NOK 5507\n",
+ "batch 225 : total OK 16971 NOK 5529\n",
+ "batch 226 : total OK 17056 NOK 5544\n",
+ "batch 227 : total OK 17121 NOK 5579\n",
+ "batch 228 : total OK 17175 NOK 5625\n",
+ "batch 229 : total OK 17241 NOK 5659\n",
+ "batch 230 : total OK 17313 NOK 5687\n",
+ "batch 231 : total OK 17368 NOK 5732\n",
+ "batch 232 : total OK 17422 NOK 5778\n",
+ "batch 233 : total OK 17462 NOK 5838\n",
+ "batch 234 : total OK 17545 NOK 5855\n",
+ "batch 235 : total OK 17596 NOK 5904\n",
+ "batch 236 : total OK 17663 NOK 5937\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "batch 237 : total OK 17729 NOK 5971\n",
+ "batch 238 : total OK 17807 NOK 5993\n",
+ "batch 239 : total OK 17882 NOK 6018\n",
+ "batch 240 : total OK 17926 NOK 6074\n",
+ "batch 241 : total OK 17989 NOK 6111\n",
+ "batch 242 : total OK 18044 NOK 6156\n",
+ "batch 243 : total OK 18096 NOK 6204\n",
+ "batch 244 : total OK 18172 NOK 6228\n",
+ "batch 245 : total OK 18220 NOK 6280\n",
+ "batch 246 : total OK 18297 NOK 6303\n",
+ "batch 247 : total OK 18342 NOK 6358\n",
+ "batch 248 : total OK 18412 NOK 6388\n",
+ "batch 249 : total OK 18491 NOK 6409\n",
+ "batch 250 : total OK 18536 NOK 6464\n",
+ "batch 251 : total OK 18592 NOK 6508\n",
+ "batch 252 : total OK 18644 NOK 6556\n",
+ "batch 253 : total OK 18693 NOK 6607\n",
+ "batch 254 : total OK 18760 NOK 6640\n",
+ "batch 255 : total OK 18824 NOK 6676\n",
+ "batch 256 : total OK 18902 NOK 6698\n",
+ "batch 257 : total OK 18960 NOK 6740\n",
+ "batch 258 : total OK 19022 NOK 6778\n",
+ "batch 259 : total OK 19081 NOK 6819\n",
+ "batch 260 : total OK 19153 NOK 6847\n",
+ "batch 261 : total OK 19230 NOK 6870\n",
+ "batch 262 : total OK 19290 NOK 6910\n",
+ "batch 263 : total OK 19351 NOK 6949\n",
+ "batch 264 : total OK 19407 NOK 6993\n",
+ "batch 265 : total OK 19483 NOK 7017\n",
+ "batch 266 : total OK 19540 NOK 7060\n",
+ "batch 267 : total OK 19619 NOK 7081\n",
+ "batch 268 : total OK 19693 NOK 7107\n",
+ "batch 269 : total OK 19766 NOK 7134\n",
+ "batch 270 : total OK 19831 NOK 7169\n",
+ "batch 271 : total OK 19897 NOK 7203\n",
+ "batch 272 : total OK 19943 NOK 7257\n",
+ "batch 273 : total OK 20018 NOK 7282\n",
+ "batch 274 : total OK 20100 NOK 7300\n",
+ "batch 275 : total OK 20167 NOK 7333\n",
+ "batch 276 : total OK 20241 NOK 7359\n",
+ "batch 277 : total OK 20320 NOK 7380\n",
+ "batch 278 : total OK 20406 NOK 7394\n",
+ "batch 279 : total OK 20463 NOK 7437\n",
+ "batch 280 : total OK 20511 NOK 7489\n",
+ "batch 281 : total OK 20595 NOK 7505\n",
+ "batch 282 : total OK 20665 NOK 7535\n",
+ "batch 283 : total OK 20750 NOK 7550\n",
+ "batch 284 : total OK 20805 NOK 7595\n",
+ "batch 285 : total OK 20885 NOK 7615\n",
+ "batch 286 : total OK 20962 NOK 7638\n",
+ "batch 287 : total OK 21041 NOK 7659\n",
+ "batch 288 : total OK 21124 NOK 7676\n",
+ "batch 289 : total OK 21208 NOK 7692\n",
+ "batch 290 : total OK 21273 NOK 7727\n",
+ "batch 291 : total OK 21352 NOK 7748\n",
+ "batch 292 : total OK 21424 NOK 7776\n",
+ "batch 293 : total OK 21461 NOK 7839\n",
+ "batch 294 : total OK 21523 NOK 7877\n",
+ "batch 295 : total OK 21577 NOK 7923\n",
+ "batch 296 : total OK 21635 NOK 7965\n",
+ "batch 297 : total OK 21707 NOK 7993\n",
+ "batch 298 : total OK 21788 NOK 8012\n",
+ "batch 299 : total OK 21841 NOK 8059\n",
+ "batch 300 : total OK 21905 NOK 8095\n",
+ "batch 301 : total OK 21944 NOK 8156\n",
+ "batch 302 : total OK 22022 NOK 8178\n",
+ "batch 303 : total OK 22104 NOK 8196\n",
+ "batch 304 : total OK 22188 NOK 8212\n",
+ "batch 305 : total OK 22259 NOK 8241\n",
+ "batch 306 : total OK 22339 NOK 8261\n",
+ "batch 307 : total OK 22420 NOK 8280\n",
+ "batch 308 : total OK 22494 NOK 8306\n",
+ "batch 309 : total OK 22575 NOK 8325\n",
+ "batch 310 : total OK 22610 NOK 8390\n",
+ "batch 311 : total OK 22658 NOK 8442\n",
+ "batch 312 : total OK 22694 NOK 8506\n",
+ "batch 313 : total OK 22768 NOK 8532\n",
+ "batch 314 : total OK 22829 NOK 8571\n",
+ "batch 315 : total OK 22907 NOK 8593\n",
+ "batch 316 : total OK 22976 NOK 8624\n",
+ "batch 317 : total OK 23012 NOK 8688\n",
+ "batch 318 : total OK 23069 NOK 8731\n",
+ "batch 319 : total OK 23138 NOK 8762\n",
+ "batch 320 : total OK 23166 NOK 8834\n",
+ "batch 321 : total OK 23243 NOK 8857\n",
+ "batch 322 : total OK 23312 NOK 8888\n",
+ "batch 323 : total OK 23395 NOK 8905\n",
+ "batch 324 : total OK 23467 NOK 8933\n",
+ "batch 325 : total OK 23534 NOK 8966\n",
+ "batch 326 : total OK 23583 NOK 9017\n",
+ "batch 327 : total OK 23643 NOK 9057\n",
+ "batch 328 : total OK 23704 NOK 9096\n",
+ "batch 329 : total OK 23741 NOK 9159\n",
+ "batch 330 : total OK 23813 NOK 9187\n",
+ "batch 331 : total OK 23885 NOK 9215\n",
+ "batch 332 : total OK 23935 NOK 9265\n",
+ "batch 333 : total OK 23976 NOK 9324\n",
+ "batch 334 : total OK 24035 NOK 9365\n",
+ "batch 335 : total OK 24124 NOK 9376\n",
+ "batch 336 : total OK 24201 NOK 9399\n",
+ "batch 337 : total OK 24266 NOK 9434\n",
+ "batch 338 : total OK 24327 NOK 9473\n",
+ "batch 339 : total OK 24383 NOK 9517\n",
+ "batch 340 : total OK 24458 NOK 9542\n",
+ "batch 341 : total OK 24507 NOK 9593\n",
+ "batch 342 : total OK 24579 NOK 9621\n",
+ "batch 343 : total OK 24664 NOK 9636\n",
+ "batch 344 : total OK 24731 NOK 9669\n",
+ "batch 345 : total OK 24792 NOK 9708\n",
+ "batch 346 : total OK 24852 NOK 9748\n",
+ "batch 347 : total OK 24907 NOK 9793\n",
+ "batch 348 : total OK 24978 NOK 9822\n",
+ "batch 349 : total OK 25041 NOK 9859\n",
+ "batch 350 : total OK 25113 NOK 9887\n",
+ "batch 351 : total OK 25184 NOK 9916\n",
+ "batch 352 : total OK 25247 NOK 9953\n",
+ "batch 353 : total OK 25315 NOK 9985\n",
+ "batch 354 : total OK 25385 NOK 10015\n",
+ "batch 355 : total OK 25454 NOK 10046\n",
+ "batch 356 : total OK 25505 NOK 10095\n",
+ "batch 357 : total OK 25580 NOK 10120\n",
+ "batch 358 : total OK 25655 NOK 10145\n",
+ "batch 359 : total OK 25727 NOK 10173\n",
+ "batch 360 : total OK 25794 NOK 10206\n",
+ "batch 361 : total OK 25863 NOK 10237\n",
+ "batch 362 : total OK 25946 NOK 10254\n",
+ "batch 363 : total OK 26008 NOK 10292\n",
+ "batch 364 : total OK 26088 NOK 10312\n",
+ "batch 365 : total OK 26129 NOK 10371\n",
+ "batch 366 : total OK 26183 NOK 10417\n",
+ "batch 367 : total OK 26240 NOK 10460\n",
+ "batch 368 : total OK 26308 NOK 10492\n",
+ "batch 369 : total OK 26383 NOK 10517\n",
+ "batch 370 : total OK 26468 NOK 10532\n",
+ "batch 371 : total OK 26518 NOK 10582\n",
+ "batch 372 : total OK 26578 NOK 10622\n",
+ "batch 373 : total OK 26630 NOK 10670\n",
+ "batch 374 : total OK 26701 NOK 10699\n",
+ "batch 375 : total OK 26753 NOK 10747\n",
+ "batch 376 : total OK 26820 NOK 10780\n",
+ "batch 377 : total OK 26891 NOK 10809\n",
+ "batch 378 : total OK 26961 NOK 10839\n",
+ "batch 379 : total OK 27037 NOK 10863\n",
+ "batch 380 : total OK 27103 NOK 10897\n",
+ "batch 381 : total OK 27175 NOK 10925\n",
+ "batch 382 : total OK 27242 NOK 10958\n",
+ "batch 383 : total OK 27297 NOK 11003\n",
+ "batch 384 : total OK 27366 NOK 11034\n",
+ "batch 385 : total OK 27442 NOK 11058\n",
+ "batch 386 : total OK 27524 NOK 11076\n",
+ "batch 387 : total OK 27575 NOK 11125\n",
+ "batch 388 : total OK 27634 NOK 11166\n",
+ "batch 389 : total OK 27703 NOK 11197\n",
+ "batch 390 : total OK 27776 NOK 11224\n",
+ "batch 391 : total OK 27860 NOK 11240\n",
+ "batch 392 : total OK 27916 NOK 11284\n",
+ "batch 393 : total OK 27970 NOK 11330\n",
+ "batch 394 : total OK 28032 NOK 11368\n",
+ "batch 395 : total OK 28106 NOK 11394\n",
+ "batch 396 : total OK 28171 NOK 11429\n",
+ "batch 397 : total OK 28233 NOK 11467\n",
+ "batch 398 : total OK 28302 NOK 11498\n",
+ "batch 399 : total OK 28368 NOK 11532\n",
+ "batch 400 : total OK 28427 NOK 11573\n",
+ "batch 401 : total OK 28518 NOK 11582\n",
+ "batch 402 : total OK 28605 NOK 11595\n",
+ "batch 403 : total OK 28677 NOK 11623\n",
+ "batch 404 : total OK 28741 NOK 11659\n",
+ "batch 405 : total OK 28797 NOK 11703\n",
+ "batch 406 : total OK 28842 NOK 11758\n",
+ "batch 407 : total OK 28897 NOK 11803\n",
+ "batch 408 : total OK 28975 NOK 11825\n",
+ "batch 409 : total OK 29047 NOK 11853\n",
+ "batch 410 : total OK 29101 NOK 11899\n",
+ "batch 411 : total OK 29193 NOK 11907\n",
+ "batch 412 : total OK 29264 NOK 11936\n",
+ "batch 413 : total OK 29319 NOK 11981\n",
+ "batch 414 : total OK 29367 NOK 12033\n",
+ "batch 415 : total OK 29439 NOK 12061\n",
+ "batch 416 : total OK 29507 NOK 12093\n",
+ "batch 417 : total OK 29584 NOK 12116\n",
+ "batch 418 : total OK 29639 NOK 12161\n",
+ "batch 419 : total OK 29663 NOK 12237\n",
+ "batch 420 : total OK 29707 NOK 12293\n",
+ "batch 421 : total OK 29759 NOK 12341\n",
+ "batch 422 : total OK 29828 NOK 12372\n",
+ "batch 423 : total OK 29885 NOK 12415\n",
+ "batch 424 : total OK 29953 NOK 12447\n",
+ "batch 425 : total OK 30012 NOK 12488\n",
+ "batch 426 : total OK 30079 NOK 12521\n",
+ "batch 427 : total OK 30161 NOK 12539\n",
+ "batch 428 : total OK 30231 NOK 12569\n",
+ "batch 429 : total OK 30294 NOK 12606\n",
+ "batch 430 : total OK 30356 NOK 12644\n",
+ "batch 431 : total OK 30407 NOK 12693\n",
+ "batch 432 : total OK 30480 NOK 12720\n",
+ "batch 433 : total OK 30545 NOK 12755\n",
+ "batch 434 : total OK 30620 NOK 12780\n",
+ "batch 435 : total OK 30672 NOK 12828\n",
+ "batch 436 : total OK 30746 NOK 12854\n",
+ "batch 437 : total OK 30822 NOK 12878\n",
+ "batch 438 : total OK 30900 NOK 12900\n",
+ "batch 439 : total OK 30962 NOK 12938\n",
+ "batch 440 : total OK 31025 NOK 12975\n",
+ "batch 441 : total OK 31093 NOK 13007\n",
+ "batch 442 : total OK 31147 NOK 13053\n",
+ "batch 443 : total OK 31187 NOK 13113\n",
+ "batch 444 : total OK 31259 NOK 13141\n",
+ "batch 445 : total OK 31329 NOK 13171\n",
+ "batch 446 : total OK 31408 NOK 13192\n",
+ "batch 447 : total OK 31460 NOK 13240\n",
+ "batch 448 : total OK 31535 NOK 13265\n",
+ "batch 449 : total OK 31611 NOK 13289\n",
+ "batch 450 : total OK 31651 NOK 13349\n",
+ "batch 451 : total OK 31724 NOK 13376\n",
+ "batch 452 : total OK 31798 NOK 13402\n",
+ "batch 453 : total OK 31854 NOK 13446\n",
+ "batch 454 : total OK 31887 NOK 13513\n",
+ "batch 455 : total OK 31936 NOK 13564\n",
+ "batch 456 : total OK 31980 NOK 13620\n",
+ "batch 457 : total OK 32055 NOK 13645\n",
+ "batch 458 : total OK 32133 NOK 13667\n",
+ "batch 459 : total OK 32215 NOK 13685\n",
+ "batch 460 : total OK 32295 NOK 13705\n",
+ "batch 461 : total OK 32357 NOK 13743\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "batch 462 : total OK 32421 NOK 13779\n",
+ "batch 463 : total OK 32487 NOK 13813\n",
+ "batch 464 : total OK 32574 NOK 13826\n",
+ "batch 465 : total OK 32643 NOK 13857\n",
+ "batch 466 : total OK 32703 NOK 13897\n",
+ "batch 467 : total OK 32777 NOK 13923\n",
+ "batch 468 : total OK 32843 NOK 13957\n",
+ "batch 469 : total OK 32932 NOK 13968\n",
+ "batch 470 : total OK 33008 NOK 13992\n",
+ "batch 471 : total OK 33090 NOK 14010\n",
+ "batch 472 : total OK 33159 NOK 14041\n",
+ "batch 473 : total OK 33240 NOK 14060\n",
+ "batch 474 : total OK 33304 NOK 14096\n",
+ "batch 475 : total OK 33384 NOK 14116\n",
+ "batch 476 : total OK 33461 NOK 14139\n",
+ "batch 477 : total OK 33544 NOK 14156\n",
+ "batch 478 : total OK 33631 NOK 14169\n",
+ "batch 479 : total OK 33716 NOK 14184\n",
+ "batch 480 : total OK 33797 NOK 14203\n",
+ "batch 481 : total OK 33835 NOK 14265\n",
+ "batch 482 : total OK 33903 NOK 14297\n",
+ "batch 483 : total OK 33972 NOK 14328\n",
+ "batch 484 : total OK 34032 NOK 14368\n",
+ "batch 485 : total OK 34072 NOK 14428\n",
+ "batch 486 : total OK 34135 NOK 14465\n",
+ "batch 487 : total OK 34198 NOK 14502\n",
+ "batch 488 : total OK 34271 NOK 14529\n",
+ "batch 489 : total OK 34327 NOK 14573\n",
+ "batch 490 : total OK 34381 NOK 14619\n",
+ "batch 491 : total OK 34461 NOK 14639\n",
+ "batch 492 : total OK 34544 NOK 14656\n",
+ "batch 493 : total OK 34640 NOK 14660\n",
+ "batch 494 : total OK 34711 NOK 14689\n",
+ "batch 495 : total OK 34804 NOK 14696\n",
+ "batch 496 : total OK 34896 NOK 14704\n",
+ "batch 497 : total OK 34986 NOK 14714\n",
+ "batch 498 : total OK 35080 NOK 14720\n",
+ "batch 499 : total OK 35153 NOK 14747\n",
+ "batch 500 : total OK 35196 NOK 14804\n"
]
}
],
"source": [
+ "img_queue = setup_dataloader(val_dir, label_file, batch_size)\n",
+ "\n",
"ok = 0\n",
"nok = 0\n",
"i = 0\n",
- "for (imgs, lbls) in val_loader:\n",
- " ibuf_normal = imgs.numpy().reshape(accel.ishape_normal)\n",
- " exp = lbls.numpy()\n",
+ "while not img_queue.last_batch:\n",
+ " imgs, lbls = img_queue.get_batch(batch_size, timeout=None)\n",
+ " imgs = np.array(imgs)\n",
+ " exp = np.array(lbls)\n",
+ " \n",
+ " ibuf_normal = imgs.reshape(accel.ishape_normal)\n",
" obuf_normal = accel.execute(ibuf_normal)\n",
" obuf_normal = obuf_normal.reshape(batch_size, -1)[:,0]\n",
" ret = np.bincount(obuf_normal.flatten() == exp.flatten())\n",
@@ -404,14 +877,14 @@
},
{
"cell_type": "code",
- "execution_count": 16,
+ "execution_count": 14,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
- "Final top-1 accuracy: 70.412%\n"
+ "Final top-1 accuracy: 70.392%\n"
]
}
],
@@ -430,27 +903,27 @@
},
{
"cell_type": "code",
- "execution_count": 17,
+ "execution_count": 15,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
- "{'runtime[ms]': 551.5413284301758,\n",
- " 'throughput[images/s]': 1813.1007568304074,\n",
- " 'DRAM_in_bandwidth[Mb/s]': 272.92243072416755,\n",
- " 'DRAM_out_bandwidth[Mb/s]': 0.018131007568304075,\n",
- " 'fclk[mhz]': 206,\n",
- " 'batch_size': 1000,\n",
- " 'fold_input[ms]': 1.0013580322265625e-05,\n",
- " 'pack_input[ms]': 1.52587890625e-05,\n",
- " 'copy_input_data_to_device[ms]': 0.07888936996459961,\n",
- " 'copy_output_data_from_device[ms]': 0.00015473365783691406,\n",
- " 'unpack_output[ms]': 0.040567874908447266,\n",
- " 'unfold_output[ms]': 6.4373016357421875e-06}"
+ "{'runtime[ms]': 50.49920082092285,\n",
+ " 'throughput[images/s]': 1980.2293575815947,\n",
+ " 'DRAM_in_bandwidth[Mb/s]': 298.0799647380423,\n",
+ " 'DRAM_out_bandwidth[Mb/s]': 0.01980229357581595,\n",
+ " 'fclk[mhz]': 100.0,\n",
+ " 'batch_size': 100,\n",
+ " 'fold_input[ms]': 1.5020370483398438e-05,\n",
+ " 'pack_input[ms]': 2.4080276489257812e-05,\n",
+ " 'copy_input_data_to_device[ms]': 0.006676673889160156,\n",
+ " 'copy_output_data_from_device[ms]': 0.00022292137145996094,\n",
+ " 'unpack_output[ms]': 0.004586219787597656,\n",
+ " 'unfold_output[ms]': 6.9141387939453125e-06}"
]
},
- "execution_count": 17,
+ "execution_count": 15,
"metadata": {},
"output_type": "execute_result"
}
diff --git a/setup.py b/setup.py
index 7edd82b..1eb0930 100644
--- a/setup.py
+++ b/setup.py
@@ -1,4 +1,4 @@
-# Copyright (C) 2020 Xilinx, Inc
+# Copyright (C) 2020-2021 Xilinx, Inc
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
@@ -20,7 +20,7 @@
from pynq.utils import build_py as _build_py
__author__ = "Yaman Umuroglu"
-__copyright__ = "Copyright 2020, Xilinx"
+__copyright__ = "Copyright 2020-2021, Xilinx"
__email__ = "yamanu@xilinx.com"
@@ -90,7 +90,7 @@ def extend_package(path):
setup(
name=module_name,
- version="0.0.1b",
+ version="0.0.2b",
description="FINN Examples on PYNQ for Zynq and Alveo",
long_description=long_description,
long_description_content_type="text/markdown",
@@ -108,8 +108,8 @@ def extend_package(path):
setup_requires=["pynq>=2.5.1"],
install_requires=[
"pynq>=2.5.1",
- "finn-base==0.0.1b0",
- "finn-dataset_loading==0.0.4", # noqa
+ "finn-base==0.0.2b0",
+ "finn-dataset_loading==0.0.5", # noqa
],
extras_require={
':python_version<"3.6"': ["matplotlib<3.1", "ipython==7.9"],