I could not access the results of previous experiments in Experiment Manager app in MATLAB R2020b for a deep learning project. How can I fix it?

7 vues (au cours des 30 derniers jours)
I am using Experiment Manager app in MATLAB R2020b for a deep learning project. Ealier I could access the results of all the previous experiments. But, right now when I open the project I could see only the experiment files (Please see the attached figure). Not the results genearted out of it. While I checked the root folder, the 'Results' folder still contains all the data. But, it some how fails to load in the Experiment Manager app interface. It seems like the 'connection' between the 'Experiment' and the 'Results' is lost some how. Without visualising the results all together, it is impossible to compare. If I run any new experiments, the results will be visible. But not the previous results. This happened after crashing of one of my experiments which was running over night and the subsequent shutdown of the PC (NVIDIA GeForce RTX 2070, Intel Core i7-9750H, 16GB DDR4, 1TB PCIe Nvme SSD, Windows 10).
The details of the deep learning project is given below:
> Type of experiment: Deep learning classification problem using LSTM
> Hyper parameter tuning strategy: Bayesian Optimization' or 'Exhaustive Sweep'
> No. of experiments: 23
> Execution time: 1 - 50 hrs
> No. of results generated: 2 - 7
> Execution environment: GPU
> Tool boxes: Deep Learning Tool Box, MATLAB, Statistics and Machine Learning Tool box
Please help me load the results of the experiments in the Experiment Manager app interface.
  2 commentaires
AS
AS le 14 Déc 2020
Modifié(e) : AS le 14 Déc 2020
Hi,
Which NVIDIA graphics driver are you using? 441.66 or above?
What does gpuDevice return?
APARNA HARICHANDRAN
APARNA HARICHANDRAN le 15 Déc 2020
Which NVIDIA graphics driver are you using? 441.66 or above?
Game ready driver version 460.79
What does gpuDevice return?
CUDADevice with properties:
Name: 'GeForce RTX 2060'
Index: 1
ComputeCapability: '7.5'
SupportsDouble: 1
DriverVersion: 11.2000
ToolkitVersion: 10.2000
MaxThreadsPerBlock: 1024
MaxShmemPerBlock: 49152
MaxThreadBlockSize: [1024 1024 64]
MaxGridSize: [2.1475e+09 65535 65535]
SIMDWidth: 32
TotalMemory: 6.4425e+09
AvailableMemory: 5.2446e+09
MultiprocessorCount: 30
ClockRateKHz: 1200000
ComputeMode: 'Default'
GPUOverlapsTransfers: 1
KernelExecutionTimeout: 1
CanMapHostMemory: 1
DeviceSupported: 1
DeviceSelected: 1

Connectez-vous pour commenter.

Réponses (0)

Catégories

En savoir plus sur Deep Learning Toolbox dans Help Center et File Exchange

Produits


Version

R2020b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by