6 Open Source Machine Learning Frameworks and Tools
Open Source tools are an excellent choice for getting started with Machine learning. This article covers some of the top ML frameworks and tools.
Hi, This is a simple Image Classification / transformation project. The following is the requirement 1. The user will upload an image which will have a border(s) (the border color will be specified by the user) 2. The AI should remove everything in the image except the border 3. The AI should fill the border with black pixels or circles as shown in the example(User specifies the shape, size color of the border/shapes) Please see the example of the attachment
Need help with the debugging of a python AI application that is using a model trained on english words, then it predicts the words being said using a wav file. The model is training properly, but the prediction output is not. Seems like there is a syntax error. Libraries used: Keras, Librosa, Matplotlib, Numpy, OS, SKlearn.
Preprocessing of mammogram images using deep CNN and then the classification of train data with SVM, KNN, GA To calculate 1. Accuracy of whole images 2. Sensitivity 3. ROC curve 4. Accuracy of images wise 5. specificity 6. F-measure 7. recall We find, Accuracy upto 96%, Sensitivity , Specificity, F1-measure, Recall, and Confusion Matrix, and clssification of images into Milagnant an... Specificity, F1-measure, Recall, and Confusion Matrix, and clssification of images into Milagnant and Benign. After That we will visualize our models with plotting ROC cur, Comparison of accuracies with all algorithms, confusion matrix, WE have already a private data set including 152 malignant and 363 Benign images with all dimensions CC,RML,LML,RC Project should be run on tensorflow 2.3 and keras w...
By using visual studio code, You are expected to use TensorFlow and Keras to perform a deep learning process for creating an image classifier to distinguish between dog and cat. You should follow fchollet’s code to create the classifier and test this classifier with one test photo of dog and one test photo cat. Create a screenshot to show the classifying results of these two test photos. The dog and cat in the test photos should not be in your original training dataset, and you may use your own pets or your neighbor’s pets.
An encode-decoder enabled RNN that can generate improved runtime code. The RNN will be trained on number of coding samples from and and then later will be tested...folder stores the polynomial time coding sample and theoutput folder will store the corresponding linear time coding model training: The RNN model can be trained on the 80% of the stored RNN model with encoder-decoder () or seq2seq( ) feature needs to be applied for better : The above two steps can be tested, integrated for final analysis.
...closing at 80%+ accuracy. The only thing that matters is the color of the next candlestick, not the size of the price move. This is because the trading style is for "binary options" (which you can learn about from Google). Here are some reference projects I have found: Please feel free to ask me any questions. Though perhaps an asset, no prior knowledge of financial markets or trading is required, as the scope should be straightforward
I'm training a video classification model using TensorFlow and Keras. Currently my data pipeline is a custom generator than reads the raw video file, do some preprocessing and feed the processed data into the model batch by batch. However, this is really slow and not scalable to large datasets. I'm currently running the model on my university's high performance computing server, which allows me to request up to 28 cpu cores (with up to 128 gb of memory each) and 4 gpu cores per job. However, my custom generator, which processes videos sequentially are not utilizing these resources for speedup and is pretty much stuck at the I/O disk reading and video pre-processing part. I'm unfamiliar with pipeline but heard that it can prefetch data. I also heard about paral...
Using the Data Generators in Keras code here is the link (~shervine/blog/keras-how-to-generate-data-on-the-fly) work with this code, and albumentation augmentation code to start the training of CV model
Hello, I need a Python programmer to help me with an issue. I want to train a deep neural network in R using the keras and tensorflow packages. However, two keras commands do not work, namely: to_categorical() and keras_model_sequential(). I mainly receive two types of errors: Installation of TensorFlow not found OR Python module was not found I have tried to solve this in many ways. I have installed Python on my computer (Windows 7) with Miniconda, and the two packages (I watched Youtube tutorials on this topic). I also installed and activated the keras, reticulate and tensorflow packages in R. I have googled my error messages and looked in various forums for a solution. Nothing seems to work. Someone advised me to downgrade Miniconda from the latest version to...
looking to do hyperparameter optimization using three algorithms: 1- grid search, 2- Bayesian optimization 3- genetic algorithms. Dataset: 1- Santander Customer Transaction Prediction Dataset ( from Kaggle) this has to be done using KERAS (can be change) to build NN The input layer dimension fixed to 200, same as dataset features number and activation function to “relu”, while output layer size was one and “sigmoid” as activation function. looking to measure: 1-Accuracy 2-Loss 3-Mean Squared Error (MSE) 4-Mean Squared Log Error (MSLE) 5-Area Under the Curve (AUC) 6-Confusion matrix 7- Wall time search space: No Hidden layer Optimizers Loss Functions Activation Dropout V. Split 1 100 SGD Mean Squared Error elu 0.0 0.20 2 150 RMSprop Mean Absolute Error expo...
I need to measure the dimensions of a 3D model feature using camera vision in the OpenCV tool and use image processing of...measure the dimensions of a 3D model feature using camera vision in the OpenCV tool and use image processing of that model to gain a better quality of an image. -I need to implement a Convolutional neural network with Keras to predict the futures of the models/objects to gain the Performance measure on multiclass classification accuracy, for train and validation of the object or model. in summary - detect the feature shapes such as (hole, pocket, rectangular and text with grooving in another part. -measuring the dimensions of each feature shape or area. -apply the CNN with Keras for both images to predict the feature shapes and text(UTHM FKMP).
the project translob on github is incomplete, needs assembling and a missing ipython notebook to run the experiment in the paper. fragment of files here it is loosely based on deeplob also on github here which has a working model and notebook. the project is to recreate the missing notebook in the translob github so that the results in the paper in that repository can be correctly replicated. the FI2010 dataset I can provide, it is also available for free on kaggle and other places
I need the implementation of a paper which is based on a CNN model to segment blood in brain CT images. The method is attached as a pdf file. There are 36 CT images along with their labels. The paper uses 25,000 unlabe...which is based on a CNN model to segment blood in brain CT images. The method is attached as a pdf file. There are 36 CT images along with their labels. The paper uses 25,000 unlabelled images to help the supervised training; since I do not have this amount of data, I'm going to use 20 images for training, treat the 10 images as unlabelled, and 6 images as testing. My preferred DL framework is Keras, however, pytorch and tensorflow would be acceptable as well The project needs to be done preferably in one week (max:two weeks) Please let me know if you have...
I have a ubuntu server running 18.04 and have used conda to install tensorflow gpu. But after installing tensorflow gpu via conda and installing numpy etc my code for fizyr keras-retinanet fails to run. Let me know if you can help. Looking for someone that has some experience with fizyr keras-retinanet
I am in a voluntary research group at my University, I have been asked to help adapt some models made in R using Keras-tensorflow. I have limited experience with either. After 2 weeks I am stuck on a number of errors that I have been unable to resolve. Some errors arising from my inexperience working with tensors or atomic vectors. Others due to deprecation of some functions like fit_generator() from the work I inherited. This is neither directly course nor thesis related, though is tangentially related to my thesis work. I will give proper acknowledgement of your contribution with any submitted models. Sample error attached.
we have to observe our live webcam/video stream where products will go through, on the products (t-shirts) are designs placed - this design has to be compared with the original image from DB. if there is something wrong with the image comparing, then set some value in DB. for example where products will go through, on the products (t-shirts) are designs placed - this design has to be compared with the original image from DB. if there is something wrong with the image comparing, then set some value in DB. for example
Hi I need to increase the accuracy for a model based on keras. Also, I need to find the model prediction and inference time. Its not very big work. Let's talk. Cannot pay more than the project bid amount.
Forecasting of weather variable using time-series geospatial data with the help of LSTM. One should be able to do the following: 1. Crop netCDF file for a region using latitude and longitude (100 years monthly multiple variable data) 2. extract variable values for a time period from the cropped region 3. process/parse/normalize geospatial data for multiple variables 4. use de...times 7. Find variable importance for various time periods and regions in the test dataset. 8. Interpretation of data in the deep learning model. 9. Plotting the prediction of forecasted data and comparison with original data. 65% of the work is done. Just need help with verification and the last couple of above-mentioned steps. Candidate should have a good understanding of xarray, numpy, gdal, keras and...
I want help to convert keras based code to pytorch framework. I basically look for someone with deep understanding in keras as well as pytorch, since the basic work in here is the conversion of keras code to pytorch.
replicate the results in python 3 in the following paper: Generating virtual scenarios of multivariate financial data for quantitative trading applications more details here: and here: there are ... there are various WGAN repositories on github that can be used a starting base, for example etc the aim is to be able to produce a large number of synthetic curves based on a pair (or more ) of original curves maintaining a degree of correlation between synthetic pairs. prefer pytorch but tensorflow keras ok
Hello everyone! Currently I'm looking for a Data Scientist specialized in the field of Financial Data Analsysis, Regression and Deep Learning including RNN, LSTM, etc. to apply in the cryptocurrency markets. ATTENTION: At the moment I only want to hire for Phase 1! But OF COURSE I am looking for a developer who has the necessary technical and interpersonal basis for intensive, friendly and long-term cooperation for all phases and above, who enjoys development and communication and who is always helpful. Please read the attached documents and fill in the questionary by adding some "X" and some numbers with paint or similar. For applications for this project, please respond with at least a message and a completed questionnairy. As as goal and for full payment I expect an ...
Code uses tensorflow keras to classify medical ultrasound images. It uses transfer learning to develop model to use in predictions Knowledge of python is required as well as flask Some enhancements in following areas 1. Review and enhance training and prediction code written using python and tensorflow keras numpy 2. The transfer learning and model needs reviews and upgrade 3. The model I had created ie "last_epoch" may or may not be the best one. I had run only 20 epochs and not sure if I made the appropriate changes to the pre-trained model ie vgg16 4. I think a deep learning exprrt needs to review and make corrections to the model architecture, weights, hyperparameters, optimizations etc 5. The current model ie last_epoch has associated weight files (stor...
Make a simple OCR model with keras and Python. Use the created model to recognize or extract text in real time using webcams
please do message me if you have worked on uart or any object detection project, I've 90% of it complete, just need to solve some errors. message me if you are interested
i have a python script that is wrote using the Keras Library. I would like it to function utilizing the GPU. I have it sitting on a 4X Nvedia GPU server but it does not seem to utilize the GPU. looking for someone to make this work properly. all tests (with tensorflow and keras) show that the GPUs are actually detected fine need an expert in running python scripts / Keras utilizing GPU
I have collected too many photos and categorized them I want someone who is able to choose the best deep learning approach CNN, RNN..., and tools i.e Keras .. etc and justify why the chosen approach is better please answer 1 + 1 = ? in the first line of ur proposal
I have collected too many photos and categorized them I want someone who is able to choose the best deep learning approach CNN, RNN..., and tools i.e Keras .. etc and justify why the chosen approach is better please answer 1 + 1 = ? in the first line of ur proposal
I have functioning Keras / Tensorflow in RStudio code that trains up a NN. But, I now need to retrain with different output layers using different optimization criteria for the outputs. Between TF bugs and a shortage of usable examples, I need some help. So, if you can help with the above, I could have you help with some additional tasks too. Again, needs to be in R (I don't know Python) and needs to be in Keras / TF (tried to get RTorch to work, but install failed again and again and again). It's a very niche project. I'm hoping someone out there can help. Thanks.
Need to build an LSTM algorithm to predict sequence categorization. Algorithm ideally built on Keras or Tensorflow. API and SQL experience also needed.
Must have: Semantic semantic network (SPARQL, linkde Data, RDF, OWL, swrl, RuleMl, etc.); Experience with ontologies ; Experience in operating and / or developing a knowledge storage system; Expe...(particularly graph databases); Experience with microservice architectures; Experience with Machine Learning (techniques and algorithms, such as k-NN, Naive Bayes, SVM, Decision tree, clustering, artificial neural networks, etc.); Experience Deep Learning; Experience in Natural Language Processing and in particular Information Extraction; Experience with standard data technologies Science (PyTorch, Pandas, Keras, SciPy, SciKit, Tensorflow,NumPy, MatLab, etc.); Experience in creating storage architecture and data processing and knowledge (data architectures, data models, data warehouses/da...
I am working on time series anomaly detection project using unsupervised approach which ...approach which is forecasting based. I have python based code written for anomaly detection approach which classifies anomalies by comparing actual value with predicted. Deep learning models are implemented using torch. I need to convert it using keras library instead of torch and improve the code in terms of any idea that could be possible. The code is research paper based which is fixed model used for unsupervised anomaly detection. Deliverables To completely understand the code Modify the code (convert the torch based CNN,LSTM models model into keras or tensorflow) Update our model in real time. Improve the code to continuously keep our model up to date to adapt to the latest behavi...
Hi. I need some help to build a Jupiter notebook (Python) for teaching how to: - define and train a model to binary classify dogs vs cats using AWS sagemaker - deploy the model to sagemaker - test it with random images from cats and dogs There ...(Python) for teaching how to: - define and train a model to binary classify dogs vs cats using AWS sagemaker - deploy the model to sagemaker - test it with random images from cats and dogs There are a couple of similar tutorials online but they need some improvements and I'm in a hurry: All code must be run from the notebook, training images uploaded to s3 Anybody willing to do it?
The task should be done in Google Colab with Python: I want to use Pose Estimation in Google Colab in order to upload an image and get coordinates of key points from the image. So I upload an RGB image of any size then transform it into numpy array. Pose Estimation gives me a dictionary with each key point annotated by 2 numbers - index of the pixel along axis 0 (Y direction ) and ...com/drive/1RCt7IFrJGHWh4hi3azLmi_3xIXW4SRjI Here is link which I found with PE in Google Colab. May be it will help. This code gets Youtube video as input and returns this video with drawn keypoints and lines. But I don't need the keypoins to be drawn on the image. I will do it myself with your code. I just need to get pixel coordinates. These coordinates will be used for machine learning with Kera...
I'm looking for a tutor who could teach me on a practical project, I'll give you the project link from github, and you tutor me by rebuilding the same project in keras framework
I have Human Activity Recognition using Accelerometer and CNN task using Keras. So if you are Tensorflow & Keras expert, please bid now.
Hi I am looking for expert in developing deep learning models using Keras API, Tensor Flow and Python. the details of project is attached. please go through the attached document carefully and please contact if you find this suitable for you ,as implementation of the same is expected earliest. please participate on the biding if you are sure enough to delivered the mentioned objectives in attached document .
I need someone to build a flight delay prediction model using artifitial neural network for academic purpose. The model should use pandas, tensorflow and keras. The model should take input from a dataset, train the neural network then test it. I uploaded a file of another ANN that predicts if a bank customer could leave the bank or not with the dataset as a sample and and it should be similar to it with some differences. Explaining the code after finishing is a must.
Need someone to develop a feedforward neuronal network with Python and Keras and based using on the MNIST data set. The goal is to test different hyperparameters of the model and to determine the performance of a final model with the best possible combination of hyperparameters.
Hi Aleksa F., I was wondering if you are familiar with deep learning libraries such as Keras, Tensorflow.
I want someone who is expert in CNN, LSTM, GRU. PYTHON , TENSORFLOW, KERAS You should have GPUs, because i will provide u with dataset, if u don't have GPUs , please don't comment My project is to creat a license plate recognition, I have some already written codes in github but needs some arrangements and developing . money will be paid when its done !! more details in private
Hi developers! My company needs a python module that can recognize two classes of "objects" in low-resolution images. The "objects" to be recognized are specific characteristics (anomalies) of the photographed surface. The stack is python + keras. The developer will receive a few thousand images already labeled for the two classes of anomalies and will have to create a neural network (of his choice) and two functions for training and use the NN. If necessary, he will have to process the images before training to improve recognition performance. At the end of the work, two python functions (or two notebooks containing them) must be delivered. The pseudo code of this functions is as follows: # TRAINING FUNCTION def train(images_folder_path): # Load image from...
I am looking for someone that can program a deep q learning bot with keras. the bot will receive live data stream and take actions based on the data.
Data Description: You are provided with a dataset of images of plant seedlings at various stages of grown. Each image has a filename that is its unique id. The dataset comprises 12 plant species. The goal of the project is to create a classifier capable of determining a plant's species from a photo.
I am looking for someone familiar with VAE, RNN, and GAN. Please if you do not have an experience with these do not replay. Also, new account or accounts with empty profile, do not replay There is an existing deep learning model that we trying to improve the resul...looking for someone familiar with VAE, RNN, and GAN. Please if you do not have an experience with these do not replay. Also, new account or accounts with empty profile, do not replay There is an existing deep learning model that we trying to improve the results little bet. The model is coded already. Link: Paper: I have an idea that we can add so the model result gets better
I need a professional freelancer who can complete my task. You must have knowledge of tensorflow and keras.
Implement 'Test Time Augmentation' and 'Weighted Box Fusion' techniques in existing YOLOv4+Keras/TensorFlow 2.0 project. these techniques may boost the current mAP score of the model. Reference code is available in pytorch. ping me if you can add these two.
I need an expert that can do hyperparameter optimization and find an optimal CNN architecture. The freelancer would have to implemet 1D CNN and 2D CNN. And preference will be given to those that have gpu at their disposal. Details will be provided in chat.
Implement a simple white box attack in Keras to attack a simple neural network
Open Source tools are an excellent choice for getting started with Machine learning. This article covers some of the top ML frameworks and tools.
Many front-end development tools in the market can help faster web development, but we will enlist only the top 6 from them. continue reading