Figure 22. Clicking On [Analyze] - ST AEKD-AICAR1 User Manual

Step 2. Click on [Analyze] in the newly opened window.
If the importing procedure for the LSTM AI-sensing node is correct, a new report is generated. This
new report shows the architecture of the neural network and the ROM and RAM memory usage.
Neural Network Tools for STM32 v1.4.0 (AI tools v5.2.0)
-- Importing model
-- Importing model - done (elapsed time 0.574s)
-- Rendering model
-- Rendering model - done (elapsed time 0.077s)
Creating report file C:\SPC5Studio-6.0\workspace_AI\SPC58ECxx_RLA AI Car Sensing
Node\components\spc5_ai_component_rla\cfg\sensing_node_network_analyze_report.txt
Exec/report summary (analyze dur=0.65s err=0)
------------------------------------------------------------------------------------------------------------------------
model file
type
c_name
compression
quantize
workspace dir
output dir
Node\components\spc5_ai_component_rla\cfg
model_name
model_hash
input
inputs (total)
output
outputs (total)
params #
macc
weights (ro)
activations (rw)
ram (total)
------------------------------------------------------------------------------------------------------------------------
id
------------------------------------------------------------------------------------------------------------------------
0
------------------------------------------------------------------------------------------------------------------------
1
------------------------------------------------------------------------------------------------------------------------
3
------------------------------------------------------------------------------------------------------------------------
4
------------------------------------------------------------------------------------------------------------------------
5
------------------------------------------------------------------------------------------------------------------------
model_car_sts p=24428(95.42 KBytes) macc=49668 rom=95.42 KBytes ram=3.06 KiB io_ram=616 B
Complexity per-layer - macc=49,668 rom=97,712
------------------------------------------------------------------------------------------------------------------------
id
------------------------------------------------------------------------------------------------------------------------
0
1
4
4
5
5
------------------------------------------------------------------------------------------------------------------------
UM3053 - Rev 1
: C:\SPC5Studio-6.0\workspace_AI\SPC58ECxx_RLA AI Car Sensing Node\source\model\model_car_sts.h5
: keras (keras_dump) - tf.keras 2.4.0
: sensing_node_network
: None
: None
: C:\SPC5Studio-6.0\workspace_AI\SPC58ECxx_RLA AI Car Sensing Node\components\stm32ai_ws
: C:\SPC5Studio-6.0\workspace_AI\SPC58ECxx_RLA AI Car Sensing
: model_car_sts
: 10794f1c230799b2a1cda67171827db2
: input_0 [150 items, 600 B, ai_float, FLOAT32, (50, 1, 3)]
: 600 B
: dense_1_nl [4 items, 16 B, ai_float, FLOAT32, (1, 1, 4)]
: 16 B
: 24,428 items (95.42 KiB)
: 49,668
: 97,712 B (95.42 KiB)
: 3,136 B (3.06 KiB)
: 3,752 B (3.66 KiB) = 3,136 + 600 + 16
layer (type) output shape
input_0 (Input) (50, 1, 3)
conv1d (Conv2D) (48, 1, 16)
conv1d_nl (Nonlinearity) (48, 1, 16)
conv1d_1 (Conv2D) (46, 1, 8)
conv1d_1_nl (Nonlinearity) (46, 1, 8)
flatten (Reshape) (368,)
dense (Dense) (64,)
dense_nl (Nonlinearity) (64,)
dense_1 (Dense) (4,)
dense_1_nl (Nonlinearity) (4,)
layer (type) macc
conv1d (Conv2D) ||||||||||
conv1d_1 (Conv2D) |||||||||||||||||||||||
dense (Dense) |||||||||||||||||||||||||||||||
dense_nl (Nonlinearity) |
dense_1 (Dense) |
dense_1_nl (Nonlinearity) |
Figure 22. Clicking on [Analyze]
param # connected to
160 input_0
conv1d
392 conv1d_nl
conv1d_1
conv1d_1_nl
23,616 flatten
dense
260 dense_nl
dense_1
15.5%
36.3%
47.4%
0.1%
0.5%
0.1%
UM3053
SPC5-STUDIO-AI plugin
macc rom
7,696 640
18,040 1,568
23,552 94,464
64
256 1,040
60
rom
|
|
||||||||||||||||||||||||||||||| 96.7%
|
|
|
0.7%
1.6%
0.0%
1.1%
0.0%
page 20/39