Λογότυπο Freelancer
Είσοδος Εγγραφή
Ανάρτηση Εργασίας
ΕΞΕΡΕΎΝΗΣΕ
Amazon Web Services
Computer Vision
Machine Learning (ML)
Programming
Python
Εικόνα εξωφύλλου προφίλ
Τώρα ακολουθείς τον/την
Σφάλμα στην ακολουθία χρήστη.
Αυτός ο χρήστης δεν επιτρέπει σε χρήστες να τον ακολουθούν.
Έχεις ήδη ακολουθήσει αυτό τον χρήστη.
Το πλάνο συνδρομή σου επιτρέπει μόνο 0 ακόλουθους. Κάνε αναβάθμιση εδώ.
Δεν ακολουθείς τον/την
Σφάλμα κατάργησης ακόλουθου χρήστη.
Έχεις προτείνει επιτυχώς τον/την
Σφάλμα πρότασης χρήστη.
Κάτι πήγε στραβά. Παρακαλούμε ανανέωσε τη σελίδα και δοκίμασε ξανά.
Επιτυχώς επαληθευμένο Email.
Avatar Χρήστη

Ahmad Wafri C.

@ahmadwafri

verified-v2.svg
5,0
7
3,6
3,6
57%

Top AI & Full-Stack

$30 USD / ώρα
・
Σημαία της
Malaysia (9:22 μ.μ.)
・
Έγινε μέλος στις Ιουνίου 7, 2023
$30 USD / ώρα
・
I'm working on AI technologies and web development such like NLP, ASR, CV, full stack and so on for long years. I've expertized to several AI model building/deploying with extensible, scalable functionalities. Top skills are ChatBot, NER, RE, QA and Stock AI, Speech Analysis, Computer Vision AI, full stack. For 13 years Iive spent programming (8 as primary income) Iive been cultivating persistence, performance, and a love for challenges, which allows me to consistently deliver complex tasks that others might give up. Of course, programming is always a learning experience and I am to this day doing that.
Επαληθεύσεις
Εντός χρόνου
100%
Εντός προϋπολογισμού
100%
Βαθμός αποδοχής
70%
Freelancer Python Developers Malaysia
Προβολή όλων Python Developers σε Malaysia

Πορτφόλιο

Πορτφόλιο

12028006
In this notebook, I built and train a Multi-Step / Multi-Output Regression Model powered by TensorFlow & Keras in order to predict Bitcoin's future trend. Even though a hyperparameter tuning mechanism is out of the scope of this notebook, I have tried a series of networks such as: Dense, Convolutional with and without Max Pooling, Long Short-Term Memory (LSTM), Gated Recurrent Unit, etc. The metrics I was using to evaluate how well our model generalizes when predicting on unseen data are MAE (Mean Absolute Error) and MSE (Mean Squared Error). The Neural Network that scores the lowest loss is the one that should be deployed to production. It is also important to mention that this model will be trained based on the moving averages as the Bitcoin's price is full of noise, and I don't want our model to learn useless patterns
Cryptocurrency price prediction using LSTMs
In this notebook, I built and train a Multi-Step / Multi-Output Regression Model powered by TensorFlow & Keras in order to predict Bitcoin's future trend. Even though a hyperparameter tuning mechanism is out of the scope of this notebook, I have tried a series of networks such as: Dense, Convolutional with and without Max Pooling, Long Short-Term Memory (LSTM), Gated Recurrent Unit, etc. The metrics I was using to evaluate how well our model generalizes when predicting on unseen data are MAE (Mean Absolute Error) and MSE (Mean Squared Error). The Neural Network that scores the lowest loss is the one that should be deployed to production. It is also important to mention that this model will be trained based on the moving averages as the Bitcoin's price is full of noise, and I don't want our model to learn useless patterns
Cryptocurrency price prediction using LSTMs
In this notebook, I built and train a Multi-Step / Multi-Output Regression Model powered by TensorFlow & Keras in order to predict Bitcoin's future trend. Even though a hyperparameter tuning mechanism is out of the scope of this notebook, I have tried a series of networks such as: Dense, Convolutional with and without Max Pooling, Long Short-Term Memory (LSTM), Gated Recurrent Unit, etc. The metrics I was using to evaluate how well our model generalizes when predicting on unseen data are MAE (Mean Absolute Error) and MSE (Mean Squared Error). The Neural Network that scores the lowest loss is the one that should be deployed to production. It is also important to mention that this model will be trained based on the moving averages as the Bitcoin's price is full of noise, and I don't want our model to learn useless patterns
Cryptocurrency price prediction using LSTMs
In this notebook, I built and train a Multi-Step / Multi-Output Regression Model powered by TensorFlow & Keras in order to predict Bitcoin's future trend. Even though a hyperparameter tuning mechanism is out of the scope of this notebook, I have tried a series of networks such as: Dense, Convolutional with and without Max Pooling, Long Short-Term Memory (LSTM), Gated Recurrent Unit, etc. The metrics I was using to evaluate how well our model generalizes when predicting on unseen data are MAE (Mean Absolute Error) and MSE (Mean Squared Error). The Neural Network that scores the lowest loss is the one that should be deployed to production. It is also important to mention that this model will be trained based on the moving averages as the Bitcoin's price is full of noise, and I don't want our model to learn useless patterns
Cryptocurrency price prediction using LSTMs
11156463
- Utilize several cloud API for STT (Azure, AWS, GCP, SpeechBrain, etc) - Custom service based on trained model.
Speech to Text service
11156454
Chatbot can read various document type and answer to the question based on understanding from document content. It is capable to read docx, pdf, csv and webpage as of now.
Document readable Chatbot development
11156448
The service running on AWS will be high scalable, extendable, efficient by following this pipeline.
AWS service architecture
12028006
In this notebook, I built and train a Multi-Step / Multi-Output Regression Model powered by TensorFlow & Keras in order to predict Bitcoin's future trend. Even though a hyperparameter tuning mechanism is out of the scope of this notebook, I have tried a series of networks such as: Dense, Convolutional with and without Max Pooling, Long Short-Term Memory (LSTM), Gated Recurrent Unit, etc. The metrics I was using to evaluate how well our model generalizes when predicting on unseen data are MAE (Mean Absolute Error) and MSE (Mean Squared Error). The Neural Network that scores the lowest loss is the one that should be deployed to production. It is also important to mention that this model will be trained based on the moving averages as the Bitcoin's price is full of noise, and I don't want our model to learn useless patterns
Cryptocurrency price prediction using LSTMs
In this notebook, I built and train a Multi-Step / Multi-Output Regression Model powered by TensorFlow & Keras in order to predict Bitcoin's future trend. Even though a hyperparameter tuning mechanism is out of the scope of this notebook, I have tried a series of networks such as: Dense, Convolutional with and without Max Pooling, Long Short-Term Memory (LSTM), Gated Recurrent Unit, etc. The metrics I was using to evaluate how well our model generalizes when predicting on unseen data are MAE (Mean Absolute Error) and MSE (Mean Squared Error). The Neural Network that scores the lowest loss is the one that should be deployed to production. It is also important to mention that this model will be trained based on the moving averages as the Bitcoin's price is full of noise, and I don't want our model to learn useless patterns
Cryptocurrency price prediction using LSTMs
In this notebook, I built and train a Multi-Step / Multi-Output Regression Model powered by TensorFlow & Keras in order to predict Bitcoin's future trend. Even though a hyperparameter tuning mechanism is out of the scope of this notebook, I have tried a series of networks such as: Dense, Convolutional with and without Max Pooling, Long Short-Term Memory (LSTM), Gated Recurrent Unit, etc. The metrics I was using to evaluate how well our model generalizes when predicting on unseen data are MAE (Mean Absolute Error) and MSE (Mean Squared Error). The Neural Network that scores the lowest loss is the one that should be deployed to production. It is also important to mention that this model will be trained based on the moving averages as the Bitcoin's price is full of noise, and I don't want our model to learn useless patterns
Cryptocurrency price prediction using LSTMs
In this notebook, I built and train a Multi-Step / Multi-Output Regression Model powered by TensorFlow & Keras in order to predict Bitcoin's future trend. Even though a hyperparameter tuning mechanism is out of the scope of this notebook, I have tried a series of networks such as: Dense, Convolutional with and without Max Pooling, Long Short-Term Memory (LSTM), Gated Recurrent Unit, etc. The metrics I was using to evaluate how well our model generalizes when predicting on unseen data are MAE (Mean Absolute Error) and MSE (Mean Squared Error). The Neural Network that scores the lowest loss is the one that should be deployed to production. It is also important to mention that this model will be trained based on the moving averages as the Bitcoin's price is full of noise, and I don't want our model to learn useless patterns
Cryptocurrency price prediction using LSTMs
11156463
- Utilize several cloud API for STT (Azure, AWS, GCP, SpeechBrain, etc) - Custom service based on trained model.
Speech to Text service
11156454
Chatbot can read various document type and answer to the question based on understanding from document content. It is capable to read docx, pdf, csv and webpage as of now.
Document readable Chatbot development
11156448
The service running on AWS will be high scalable, extendable, efficient by following this pipeline.
AWS service architecture
Οι αλλαγές έχουν αποθηκευτεί
5.0 · 7 Reviews
Αξιολογήσεις
M
تطوير موقع تجارة شحن الألعاب والتطبيقات • $100 USD
لا يستطيع العمل الا بارسال الاموال اليه مقدما حقا اظن انه محتا
Mohamed A.
@mohameda193
•
Σημαία της
Minya, Egypt
•
5 μήνες πριν
Κοινοποίηση
Avatar Χρήστη
Bootstrap Website Content & Design Expert • $30 USD
Worst freelancer ever! wasted me a week of time, he cannot do anything! probably just take the project and give to another freelancer to do, it's just a simple html page, and replace some contents in it, he cannot complete and deliver the work, keep letting me wait, I should aware that's a negative sign, dont trust this guy, he will literally waste all your time until deadline, avoid!!!
Tom T.
@saasprivacy
•
Σημαία της
Central, Hong Kong
•
5 μήνες πριν
Κοινοποίηση
Avatar Χρήστη
Image Score Derivation and Selection Tool • $50 SGD
Ahmad has completed the project within timeline and met the requirements, his skill set on Python is extremely good.
Abdul Raheem G.
@etlfreelancer
•
Σημαία της
singapore, Singapore
•
6 μήνες πριν
Κοινοποίηση
D
first topic • $50 USD
The project is not completed according to the dispute
Diane A.
@Diane40
•
Σημαία της
Oak Point, United States
•
7 μήνες πριν
Κοινοποίηση
S
Create a specific Digit Recognition to detect date and time • ₹26.000 INR
He is good and have knowledge about the work and technologies he is working. He helped me with image processing, Open CV for time detection task in the image.
Sukumar Reddy M.
@sukumarreddy6
•
Σημαία της
Anantapur, India
•
2 χρόνια πριν
Κοινοποίηση
Η πρόσκληση έχει σταλεί επιτυχώς!
Ευχαριστούμε! Σου έχουμε στείλει ένα email με ένα σύνδεσμο για να διεκδικήσεις τη δωρεάν πίστωση σου.
Κάτι πήγε στραβά κατά την προσπάθεια αποστολής του email σου. Παρακαλούμε δοκίμασε ξανά.
Λογότυπο Freelancer Greece / ελληνικά
Βοήθεια & Υποστήριξη
 Freelancer
ΚατηγορίεςΕργασίεςΔιαγωνισμοί FreelancersEnterpriseΑνάπτυξη ΑΙΣυνδρομήΠρόγραμμα Προτιμώμενου FreelancerΔιαχείριση ΕργασιώνΤοπικές Δουλειές Photo AnywhereΈκθεσηAPI για DevelopersΛάβε ΕπαλήθευσηDesktop Εφαρμογή
 Σχετικά
Σχετικά με εμάςΠως ΛειτουργείΑσφάλειαΕπενδυτήςΧάρτης ΣελίδαςΙστορίεςΝέαΌμαδαΒραβείαΑνακοινώσεις ΤύπουΚαριέρες
 Όροι
Πολιτική Προστασίας Προσωπικών ΔεδομένωνΌροι και ΠροϋποθέσειςΠολιτική Πνευματικών Δικαιωμάτων Κώδικας ΕπικοινωνίαςΧρεώσεις και Προμήθειες
Συνεργάτες
Escrow.comLoadshiftWarrior Forum
 Εφαρμογές
λογότυπο Apple App StoreΛογότυπο Google Play
Η αντιγραφή στο πληκτρολόγιο απέτυχε, παρακαλούμε δοκίμασε ξανά αφού τροποποιήσεις τις άδειες σου.
Αντιγράφηκε στο πληκτρολόγιο.
Εγγεγραμμένοι Χρήστες Συνολικές Αναρτημένες Δουλειές
Freelancer ® is a registered Trademark of Freelancer Technology Pty Limited (ACN 142 189 759)
Copyright © 2025 Freelancer Technology Pty Limited (ACN 142 189 759)
Φόρτωση προεπισκόπησης
Δόθηκε πρόσβαση για Geolocation.
Η σύνδεση σου έχει λήξει και τώρα έχεις αποσυνδεθεί. Παρακαλούμε συνδέσου ξανά.