search

deepjavalibrary / djl

An Engine-Agnostic Deep Learning Framework in Java
https://djl.ai
Apache License 2.0
4.12k stars 654 forks source link

Tensorflow Docs not working? #846

Closed nikkisingh111333 closed 3 years ago

nikkisingh111333 commented 3 years ago

hey ,i want to train a object detector and want to do it in tensorflow..rather than MxNet...heres what i m doing following the docs provided on djl website : MY SETUP !curl -O https://raw.githubusercontent.com/aws-samples/d2l-java/master/tools/colab_build.sh && bash colab_build.sh !sudo rm /usr/local/cuda !ln -s /usr/local/cuda-10.1 /usr/local/cuda !ln -s /usr/local/cuda/lib64/libcudart.so /usr/lib64-nvidia !nvcc --version !python3 install.py --sys-prefix !pip install pyngrok --quiet

DEPENDENCY PART %maven org.slf4j:slf4j-api:1.7.26 %maven org.slf4j:slf4j-simple:1.7.26 %maven net.java.dev.jna:jna:5.3.0 %maven ai.djl:basicdataset:0.10.0 %maven ai.djl:api:0.10.0 %maven ai.djl:model-zoo:0.10.0 %maven ai.djl.tensorflow:tensorflow-native-auto:2.3.1 %maven ai.djl.tensorflow:tensorflow-engine:0.10.0 %maven ai.djl.tensorflow:tensorflow-model-zoo:0.10.0 %maven ai.djl.tensorflow:tensorflow-api:0.10.0 My imports: import ai.djl.Application; import ai.djl.Model; import ai.djl.ModelException; import ai.djl.engine.Engine; import ai.djl.inference.Predictor; import ai.djl.modality.cv.Image; import ai.djl.modality.cv.ImageFactory; import ai.djl.modality.cv.output.BoundingBox; import ai.djl.modality.cv.output.DetectedObjects; import ai.djl.modality.cv.output.Rectangle; import ai.djl.modality.cv.util.NDImageUtils; import ai.djl.ndarray.NDArray; import ai.djl.ndarray.NDList; import ai.djl.ndarray.NDManager; import ai.djl.ndarray.types.DataType; import ai.djl.repository.zoo.Criteria; import ai.djl.repository.zoo.ModelZoo; import ai.djl.repository.zoo.ZooModel; import ai.djl.training.util.ProgressBar; import ai.djl.translate.Batchifier; import ai.djl.translate.TranslateException; import ai.djl.translate.Translator; import ai.djl.translate.TranslatorContext; import ai.djl.util.JsonUtils; import com.google.gson.annotations.SerializedName; import java.io.IOException; import java.io.InputStream; import java.net.URL; import java.nio.charset.StandardCharsets; import java.nio.file.Files; import java.nio.file.Path; import java.nio.file.Paths; import java.util.ArrayList; import java.util.List; import java.util.Map; import java.util.Objects; import java.util.Scanner; import java.util.concurrent.ConcurrentHashMap; import org.slf4j.Logger; import org.slf4j.LoggerFactory;

testing: var engine=Engine.getInstance().getEngineName(); engine ERROR I M GETTING ai.djl.engine.EngineException: No deep learning engine found. Please refer to https://github.com/awslabs/djl/blob/master/docs/development/troubleshooting.md for more details. at ai.djl.engine.Engine.getInstance(Engine.java:120) at .do_it$Aux(#99:1) at .(#99:1) WHAT I M DOING WRONG when hitting this code on python enviornment import tensorflow as tf print(tf.__version__) out:2.4.1 ..tensorflow 2.4 is there ..how to make it work in djl ...guide me pleaase

stu1130 commented 3 years ago

@nikkisingh111333 we don't support TensorFlow training. Currently only MXNet engine is fully supported for training capability. PyTorch is still working in progress and has not yet well-tested.

nikkisingh111333 commented 3 years ago

okk so it means i cannot import tensorflow model which i hve trained on teachable machine into djl...please i want to ask few things: 1.how can i train ssd model oer any other object detector in djl..i mean fine tune it..all example i m seeing are just using pre trained model instead of training them on out own dataset 2.when will be TF for djl please i want to create some awesome projects which i m unable to do since i cannot find right AI environment in java ..i m Counting on DJL..

On Mon, 12 Apr 2021 at 19:59, Jake Lee @.***> wrote:

@nikkisingh111333 https://github.com/nikkisingh111333 we don't support TensorFlow training. Currently only MXNet engine is fully supported for training capability. PyTorch is still working in progress and has not yet well-tested.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/awslabs/djl/issues/846#issuecomment-817860594, or unsubscribe https://github.com/notifications/unsubscribe-auth/AL3BT7ESTSWBUSVTAEOKYQTTIL7URANCNFSM42ZJNP5A .

zachgk commented 3 years ago

You can find examples of training in https://github.com/awslabs/djl/tree/master/examples/src/main/java/ai/djl/examples/training