Skip to content
Snippets Groups Projects
Select Git revision
  • 9-mers
  • batch_outside_tf
  • ben_dev
  • ben_gru_dev
  • coa_analysis
  • master default protected
6 results
You can move around the graph by using the arrow keys.
Created with Raphaël 2.2.014Nov19Jun16141310647May4329Apr2615141211108754130Mar28144225Feb1514230Jan282726222120181612729Dec282217141110987653230Nov292625242322211918171615121110985432129Oct28272524222120191815131187654130Sep2924232215141398762131Aug1219Jul1330Jun282525MayAdded analysis for guppy accuracy over time9-mers9-mersSmall implementation changes.updatebatch_outside_tfbatch_outside_tfadd readmebug fixesupdateadapt to kmer counting to include 8 and 9 mersupdateupdateupdatestart with tflite implementationBalanced dataset for all species is now used to calculate performance. Also increased deepnano accuracy.additional args for inferencemastermasterbatch outside tfBaseless cutoff at 55% of kmers and some minor refactoringCalculate accuracies for all foldsBugfixed inferenceModel now outputs fraction of found kmersChanged default kmer cutoffUpdated help message of --parameter-fileMade ugly way to only look at balanced datasetsFixed memory issue in run_inference_naive.pyremove coa detection directoryadd script for conversion h5 to trtcoa_analysiscoa_analysisinference memory and batch size are argumentsImplemented functional API for compile_model.pyBugfix and modified implementationFind optimal cutoff and print PR curves for baseless inference on sampleChanged signature of scan_one_fileRefactoring for import statementCode to determine optimal cutoff based on k-foldAdded implementation to create in silico samples of readsRenamed compare_accuracy.py to compare_accuracy_per_read.pyrefactoring for jetson compatabilityChanged order of tools in bar chartUpdated inference procedure of deepnano and guppy to increase their performanceChanged some parametersNow handles missing neural networks correctlyChanged baseless_output to other directory8+9-mer models now implemented