# Tuning Help **URL:** https://community.konduit.ai/c/tuning-help/11.md [Latest](https://community.konduit.ai/latest.md) · [Categories](https://community.konduit.ai/categories.md) --- ## [Problems with getting your Neural Network to work? Read this first!](https://community.konduit.ai/t/problems-with-getting-your-neural-network-to-work-read-this-first/35)
**Author:** [@treo](https://community.konduit.ai/u/treo)\ **Replies:** 0\ **Last updated:** [November 22, 2019, 12:40pm UTC](https://community.konduit.ai/t/problems-with-getting-your-neural-network-to-work-read-this-first/35 "2019-11-22T12:40:50Z")
A place for tuning related questions. Tuning questions are those like “what settings should I use for my net?” or “why am I getting poor performance from my network?”. Providing help for tuning neural networks can be qu… --- ## [Neural Network Struggling to Reproduce Tails in Bi-Modal Distribution](https://community.konduit.ai/t/neural-network-struggling-to-reproduce-tails-in-bi-modal-distribution/3227)
**Author:** [@genlarose](https://community.konduit.ai/u/genlarose)\ **Replies:** 1\ **Last updated:** [August 27, 2024, 11:11am UTC](https://community.konduit.ai/t/neural-network-struggling-to-reproduce-tails-in-bi-modal-distribution/3227 "2024-08-27T11:11:34Z")
Hello, I wrote a DL4J program in Kotlin which uses a single neural network for generating data mimicking the training data’s distribution. In this particular case, I am trying to generate data following a bi-modal normal… --- ## [Neural Network Struggling to Reproduce Tails in Bi-Modal Distribution](https://community.konduit.ai/t/neural-network-struggling-to-reproduce-tails-in-bi-modal-distribution/3219)
**Author:** [@genldeb](https://community.konduit.ai/u/genldeb)\ **Replies:** 0\ **Last updated:** [August 22, 2024, 1:16am UTC](https://community.konduit.ai/t/neural-network-struggling-to-reproduce-tails-in-bi-modal-distribution/3219 "2024-08-22T01:16:38Z")
I wrote a DL4J program in Kotlin which uses a single neural network for generating data mimicking the training data’s distribution. In this particular case, I am trying to generate data following a bi-modal normal distri… --- ## [Does NormalizerStandardize.transform(dataSet) support multi-threads?](https://community.konduit.ai/t/does-normalizerstandardize-transform-dataset-support-multi-threads/3075)
**Author:** [@baedaron](https://community.konduit.ai/u/baedaron)\ **Replies:** 3\ **Last updated:** [March 8, 2024, 2:11pm UTC](https://community.konduit.ai/t/does-normalizerstandardize-transform-dataset-support-multi-threads/3075 "2024-03-08T14:11:01Z")
hi, I have very big data and want to normalize and save them in advance to avoid repeated normalization during every epoch. so I want to use multi-threads to reduce normalization time. but one question is “Does Normal… --- ## [GPU Performance is worse than CPU](https://community.konduit.ai/t/gpu-performance-is-worse-than-cpu/3070)
**Author:** [@baedaron](https://community.konduit.ai/u/baedaron)\ **Replies:** 4\ **Last updated:** [February 15, 2024, 10:45pm UTC](https://community.konduit.ai/t/gpu-performance-is-worse-than-cpu/3070 "2024-02-15T22:45:28Z")
I am training 2 Auto-Encoders. One on CPU and the other on GPU. With same training data types. See below 2 logs(iteration time intervals). OS: Windows 11 DL4J Version: M2.1 Please help! CPU LOG: GPU LOG: --- ## [NVIDIA Tensor Cores Usage](https://community.konduit.ai/t/nvidia-tensor-cores-usage/3072)
**Author:** [@seger7024](https://community.konduit.ai/u/seger7024)\ **Replies:** 0\ **Last updated:** [February 13, 2024, 8:57am UTC](https://community.konduit.ai/t/nvidia-tensor-cores-usage/3072 "2024-02-13T08:57:54Z")
Hi! Thanks for the great project. I have used DL4J in the past and I am about to begin again a small project with it. I hope its development will continue. I have a rather newbie question but I have not come across an … --- ## [Imported model have lower accuracy](https://community.konduit.ai/t/imported-model-have-lower-accuracy/2711)
**Author:** [@minyueyong](https://community.konduit.ai/u/minyueyong)\ **Replies:** 3\ **Last updated:** [August 1, 2023, 11:33pm UTC](https://community.konduit.ai/t/imported-model-have-lower-accuracy/2711 "2023-08-01T23:33:18Z")
I want to import a tensorflow model.h5 file. The model will take in an image and predict whether its a real or fake face. But the newly imported DL4J model is not as accurate as the original tensorflow model. Is there … --- ## [ It seems CPU is utilized, though BACKEND\_PRIORITY\_GPU=10 and BACKEND\_PRIORITY\_CPU=1](https://community.konduit.ai/t/it-seems-cpu-is-utilized-though-backend-priority-gpu-10-and-backend-priority-cpu-1/2552)
**Author:** [@baedaron](https://community.konduit.ai/u/baedaron)\ **Replies:** 2\ **Last updated:** [May 8, 2023, 2:29am UTC](https://community.konduit.ai/t/it-seems-cpu-is-utilized-though-backend-priority-gpu-10-and-backend-priority-cpu-1/2552 "2023-05-08T02:29:26Z")
LOG . . . \[2023-05-06 21:45:59\] \[main\] INFO Nd4jBackend - Loaded \[JCublasBackend\] backend \[2023-05-06 21:46:01\] \[main\] INFO NativeOpsHolder - Number of threads used for linear algebra: 32 \[2023-05-06 21:46:01\] \[m… --- ## [The path of Protoc in DL4J](https://community.konduit.ai/t/the-path-of-protoc-in-dl4j/2469)
**Author:** [@yaoyao](https://community.konduit.ai/u/yaoyao)\ **Replies:** 3\ **Last updated:** [April 10, 2023, 12:13pm UTC](https://community.konduit.ai/t/the-path-of-protoc-in-dl4j/2469 "2023-04-10T12:13:09Z")
I am not familiar with protoc. Need I generate from all the proto files in \\deeplearning4j-master\\nd4j\\nd4j-backends\\nd4j-api-parent\\nd4j-api\\src\\main\\protobuf ? And the output path should be deeplearning4j-master\\nd4j… --- ## [Hyperparameters have no effect](https://community.konduit.ai/t/hyperparameters-have-no-effect/2328)
**Author:** [@Chondron](https://community.konduit.ai/u/Chondron)\ **Replies:** 14\ **Last updated:** [March 14, 2023, 12:09pm UTC](https://community.konduit.ai/t/hyperparameters-have-no-effect/2328 "2023-03-14T12:09:46Z")
Greetings. As part of my MSc research project I’m using DL4J in conjunction with wekaDeepLearning4J (version 1.7.2) to perform a land classification project. I’m using the Dl4jMlpClassifier amd Dl4jResNet50 model. I’m… --- ## [How can I turn a network into stub codes?](https://community.konduit.ai/t/how-can-i-turn-a-network-into-stub-codes/1905)
**Author:** [@huzpsb](https://community.konduit.ai/u/huzpsb)\ **Replies:** 52\ **Last updated:** [October 3, 2022, 8:33am UTC](https://community.konduit.ai/t/how-can-i-turn-a-network-into-stub-codes/1905 "2022-10-03T08:33:29Z")
Greetings! I have trained a multi-layer network to recognize some specific soundwaves.It worked perfectly well,and costed only ~1ms to work.However,when I tried to migrate it to Android,I found to my astonishment that t… --- ## [Layers is null for a MultiLayerNetwork object](https://community.konduit.ai/t/layers-is-null-for-a-multilayernetwork-object/1768)
**Author:** [@jcabot](https://community.konduit.ai/u/jcabot)\ **Replies:** 0\ **Last updated:** [February 3, 2022, 8:46am UTC](https://community.konduit.ai/t/layers-is-null-for-a-multilayernetwork-object/1768 "2022-02-03T08:46:47Z")
I’ve defined the following MultiLayerNetwork for an intent detection scenario (for a given user input I want to predict the intent that corresponds to that input). I’m trying to define a network with three layers: an emb… --- ## [Unet predictions poor performance](https://community.konduit.ai/t/unet-predictions-poor-performance/1744)
**Author:** [@Purva-Chaudhari](https://community.konduit.ai/u/Purva-Chaudhari)\ **Replies:** 0\ **Last updated:** [January 8, 2022, 9:37am UTC](https://community.konduit.ai/t/unet-predictions-poor-performance/1744 "2022-01-08T09:37:31Z")
Hello all, I was trying UNet architecture on DL4J. I get white predictions (with just one thin line of black pixels) Here is how I am loading the data, normalizing it and assigning groundtruth File testData = … --- ## [Differences in Shakespeare generator example](https://community.konduit.ai/t/differences-in-shakespeare-generator-example/1653)
**Author:** [@pwilkin](https://community.konduit.ai/u/pwilkin)\ **Replies:** 0\ **Last updated:** [October 20, 2021, 3:57pm UTC](https://community.konduit.ai/t/differences-in-shakespeare-generator-example/1653 "2021-10-20T15:57:54Z")
Hello, I’m a new visitor to the deep learning community, trying to find my way around after reading the O’Reilly book about DeepLearning4J. I tried to adapt the example from the book (GravesLSTMCharModellingExample), onl… --- ## [Sequential model tuning suggestions](https://community.konduit.ai/t/sequential-model-tuning-suggestions/1557)
**Author:** [@mCvjetko](https://community.konduit.ai/u/mCvjetko)\ **Replies:** 1\ **Last updated:** [August 14, 2021, 10:43am UTC](https://community.konduit.ai/t/sequential-model-tuning-suggestions/1557 "2021-08-14T10:43:30Z")
Hello, I’m making a model for music generation and looking for suggestions on how to improve it. The dataset consists of a feature and label file, respectively. A single feature is a 32 timestep long sequence, and each … --- ## [Missing Learning Rate Scheduler](https://community.konduit.ai/t/missing-learning-rate-scheduler/373)
**Author:** [@yptheangel](https://community.konduit.ai/u/yptheangel)\ **Replies:** 5\ **Last updated:** [July 20, 2021, 7:31am UTC](https://community.konduit.ai/t/missing-learning-rate-scheduler/373 "2021-07-20T07:31:30Z")
Hi, I am trying to implement a learning rate scheduler/policy in my Image Classifier. As I have checked the suggested LeNetExample, the latest version here does not implement the policy anymore and I saw the implementat… --- ## [Multi Label Sequence Classification](https://community.konduit.ai/t/multi-label-sequence-classification/1079)
**Author:** [@DarkWingMcQuack](https://community.konduit.ai/u/DarkWingMcQuack)\ **Replies:** 4\ **Last updated:** [June 24, 2021, 2:32pm UTC](https://community.konduit.ai/t/multi-label-sequence-classification/1079 "2021-06-24T14:32:09Z")
Hello Everybody, I am trying to do a multi label classification for sequences. I started with a simple LSTM and used this as configuration: new NeuralNetConfiguration.Builder() .seed(seed) .updater(new… --- ## [Snapshots feedback](https://community.konduit.ai/t/snapshots-feedback/1432)
**Author:** [@agibsonccc](https://community.konduit.ai/u/agibsonccc)\ **Replies:** 0\ **Last updated:** [June 8, 2021, 11:27pm UTC](https://community.konduit.ai/t/snapshots-feedback/1432 "2021-06-08T23:27:15Z")
Hello all, Some people have had issues with snapshots being out of date. We’ve added some automation that will run the actions every other day. The main issue with snapshots has mainly been the OSSRH expiry of publishe… --- ## [Convert 2d into 3d for using LSTM](https://community.konduit.ai/t/convert-2d-into-3d-for-using-lstm/1376)
**Author:** [@ramarro123](https://community.konduit.ai/u/ramarro123)\ **Replies:** 3\ **Last updated:** [May 12, 2021, 6:48am UTC](https://community.konduit.ai/t/convert-2d-into-3d-for-using-lstm/1376 "2021-05-12T06:48:21Z")
Hello, i would like to do some experiment with LSTM. Attached at the end a pseudocode of my experiment. I basically create a 2d array, with feature(s) that i generate in some way (generally speaking some forecast api … --- ## [Attributes normalization](https://community.konduit.ai/t/attributes-normalization/1365)
**Author:** [@ramarro123](https://community.konduit.ai/u/ramarro123)\ **Replies:** 1\ **Last updated:** [May 6, 2021, 7:01am UTC](https://community.konduit.ai/t/attributes-normalization/1365 "2021-05-06T07:01:12Z")
Hello, a general question about attributes normalization. i do something like DataNormalization normalizer = new NormalizerStandardize(); normalizer.fit(trainingData); normalizer.transform(trainingData); normalizer.tr… --- ## [Attribute selection](https://community.konduit.ai/t/attribute-selection/1366)
**Author:** [@ramarro123](https://community.konduit.ai/u/ramarro123)\ **Replies:** 2\ **Last updated:** [May 6, 2021, 6:50am UTC](https://community.konduit.ai/t/attribute-selection/1366 "2021-05-06T06:50:47Z")
Hello, there is a way to select (automatically with some sort of algo or function in deeplearning4j or other libs) the best attributes from a set of attributes? the example like this one 0,1,2,red 0,3,4,green 0,5,6,… --- ## [Using same data for Training and Testing](https://community.konduit.ai/t/using-same-data-for-training-and-testing/1160)
**Author:** [@artinmare](https://community.konduit.ai/u/artinmare)\ **Replies:** 1\ **Last updated:** [February 6, 2021, 9:52am UTC](https://community.konduit.ai/t/using-same-data-for-training-and-testing/1160 "2021-02-06T09:52:43Z")
Hi There, I wonder what would happen if I use the same Image I use for training process in the testing ? Is this some kind of bad practice ? Since the dataset is very specific. Just to visualize it better, if I train a … --- ## [How can I fix my model which seemingly isnt training well?](https://community.konduit.ai/t/how-can-i-fix-my-model-which-seemingly-isnt-training-well/1131)
**Author:** [@fletchni](https://community.konduit.ai/u/fletchni)\ **Replies:** 2\ **Last updated:** [January 27, 2021, 1:35am UTC](https://community.konduit.ai/t/how-can-i-fix-my-model-which-seemingly-isnt-training-well/1131 "2021-01-27T01:35:08Z")
First of all, I have gone through the suggested materials(both the visualization and troubleshooting pages on deeplearning4j’s site). To start off I will describe my model. I have built a RNN with an LSTM layer, which I… --- ## [Using LSTM to predict Random Numbers from Java's Random() class](https://community.konduit.ai/t/using-lstm-to-predict-random-numbers-from-javas-random-class/1098)
**Author:** [@sqljunkey](https://community.konduit.ai/u/sqljunkey)\ **Replies:** 0\ **Last updated:** [January 3, 2021, 9:49pm UTC](https://community.konduit.ai/t/using-lstm-to-predict-random-numbers-from-javas-random-class/1098 "2021-01-03T21:49:04Z")
Hi, I was interested to try using LSTM to predict Random numbers from Java’s Random class. I use random.getNextDouble() which gives you a random number between 0 and 1. I put these results in one sequence file for the fe… --- ## [Weird results from my LSTM prediction](https://community.konduit.ai/t/weird-results-from-my-lstm-prediction/1083)
**Author:** [@fletchni](https://community.konduit.ai/u/fletchni)\ **Replies:** 0\ **Last updated:** [December 27, 2020, 11:05pm UTC](https://community.konduit.ai/t/weird-results-from-my-lstm-prediction/1083 "2020-12-27T23:05:51Z")
Hey everyone, I am using deeplearning4j for stock predictions. I train my results and then predict both on the train data and the test data. When i print out the results of the output() method, it seems to work normally… --- ## [How to interpret training evaluation?](https://community.konduit.ai/t/how-to-interpret-training-evaluation/1074)
**Author:** [@fletchni](https://community.konduit.ai/u/fletchni)\ **Replies:** 0\ **Last updated:** [December 24, 2020, 2:28am UTC](https://community.konduit.ai/t/how-to-interpret-training-evaluation/1074 "2020-12-24T02:28:24Z")
Hello everyone, Im using an RNN with LSTM layer to make predictions on a stock, and I also implement the deeplearning4j UI while training. The charts look very bumpy, which I know the fact that I am predicting stocks w… --- ## [How do I use a MNIST model to verify a number image?](https://community.konduit.ai/t/how-do-i-use-a-mnist-model-to-verify-a-number-image/1021)
**Author:** [@TRSK-CONMI](https://community.konduit.ai/u/TRSK-CONMI)\ **Replies:** 6\ **Last updated:** [December 10, 2020, 6:08am UTC](https://community.konduit.ai/t/how-do-i-use-a-mnist-model-to-verify-a-number-image/1021 "2020-12-10T06:08:06Z")
Issue Description expected behavior:Mnist model is obtained from DL4J encountered behavior:How do I load the model? Version Information Deeplearning4j version: 1.0.0 beta7 Platform information (OS, etc): deepin20 CUDA… --- ## [Can I use DL4J's model in C?](https://community.konduit.ai/t/can-i-use-dl4js-model-in-c/1028)
**Author:** [@TRSK-CONMI](https://community.konduit.ai/u/TRSK-CONMI)\ **Replies:** 0\ **Last updated:** [December 3, 2020, 6:50am UTC](https://community.konduit.ai/t/can-i-use-dl4js-model-in-c/1028 "2020-12-03T06:50:30Z")
I don’t know if the scenario for this model is limited to JVMS, or can it be used in other environments? Thanks. --- ## [How do I test my digital images with my training model?](https://community.konduit.ai/t/how-do-i-test-my-digital-images-with-my-training-model/1025)
**Author:** [@TRSK-CONMI](https://community.konduit.ai/u/TRSK-CONMI)\ **Replies:** 2\ **Last updated:** [December 2, 2020, 9:59am UTC](https://community.konduit.ai/t/how-do-i-test-my-digital-images-with-my-training-model/1025 "2020-12-02T09:59:55Z")
Now I’m ready to load the MNIST model, but I don’t know how to get the model to output the number of this image. Thanks. --- ## [YOLO evaluation](https://community.konduit.ai/t/yolo-evaluation/987)
**Author:** [@lavajaw](https://community.konduit.ai/u/lavajaw)\ **Replies:** 0\ **Last updated:** [November 11, 2020, 6:23pm UTC](https://community.konduit.ai/t/yolo-evaluation/987 "2020-11-11T18:23:43Z")
Hello, I am trying to train YOLO model (have same setup like TinyYOLO-ZooModel). When first epoch is finished and evaluation started I am receiving error: New shape length doesn’t match original length: \[13520\] vs \[32… [Next page](https://community.konduit.ai/c/tuning-help/11.md?page=1)