Sfoglia categorie
Esplora
Fiverr Pro
Italiano
$
USD
I will design signal processing and deep learning models in python
Embedded Hardware, Flight Controls and ML Engineer
I bridge the gap between bare-metal firmware and compact hardware design. I specialize in developing custom multi-layer PCBs (ESP32, LoRa, cellular, GPS) and writing highly efficient C code for STM32 ...
Informazioni su questo servizio
Transform raw signal data into actionable intelligence.
I build custom signal processing pipelines and deep learning models for 1D/2D sensor, audio, and visual data.
Core Capabilities:
Signal Preprocessing: Feature extraction via PyWavelets (Continuous Wavelet Transforms) and OpenCV.
Deep Learning: Custom Convolutional Neural Networks (CNNs) built and trained in TensorFlow and Keras.
Pure Python Algorithms: Foundational matrix logic and algorithm implementation without heavy ML libraries for constrained hardware.
Deliverables:
Documented Python source code.
Preprocessed feature matrices.
Trained neural networks with complete evaluation metrics.
Plz message me before ordering to review your dataset structure and architecture requirements.
Linguaggio di programmazione:
Python
Framework:
Scikit-learn
•
DeepPy
•
keras
•
PyTorch
•
Panda
API:
Visione artificiale Microsoft AI
Strumenti:
opencv
•
OpenNN
•
tensorflow
•
SimpleCV
Il mio portfolio
FAQ
What types of raw signal data can you work with?
I work with 1D and 2D signals, including raw time-series sensor data, audio recordings (.wav), laser image arrays, and matrix datasets formatted in CSV, JSON, or NumPy arrays (.npy).
Can you implement algorithms in pure Python without heavy ML libraries?
Yes. If your project has hardware constraints or requires custom algorithmic design, I can build models, gradient descent routines, and matrix transformations using pure Python and NumPy.
What signal preprocessing techniques do you apply?
I utilize Continuous Wavelet Transforms (CWT via PyWavelets), Fourier Transforms (FFT/spectrograms), noise filtering, and OpenCV feature extraction to prepare raw signals for deep learning.
What files will I receive upon completion?
You will receive fully documented Python source code, preprocessed feature matrices, the trained neural network model files (.keras / .h5), and complete performance evaluation plots (loss curves, accuracy, confusion matrix).
