SNAP prediction failed and :out_of_bound exon of the .zff file

[ctsjyyk]

I trained my data using RNA-Seq, and since I don't know the strand direction, I set the strand option to no. This is my command.
funannotate train -i ../06.repeat/repeat2gff/$1.masked.fa -o $1
--left ../0.data/transcriptome/SRR26155732_1.clean.fastq ../0.data/transcriptome/SRR26155735_1.clean.fastq ../0.data/transcriptome/SRR26155>
--right ../0.data/transcriptome/SRR26155732_2.clean.fastq ../0.data/transcriptome/SRR26155735_2.clean.fastq ../0.data/transcriptome/SRR26155>
--trinity total_cdhit.rename.fa
--stranded no --jaccard_clip --species "Alternaria alternata"
--strain $1 --cpus 12
There were no errors in this step, so I started my prediction.
funannotate predict -i ../06.repeat/repeat2gff/$1.masked.fa
-o $1 -s "Alternaria alternata" --strain $1 --cpus 12 --force
The EVM results indicate that the remaining is 0, and then SNAP reported an error. An example of the error is as follows:
g_914-T1 1 1 2 + errors(2): exon-1:out_of_bounds exon-2:out_of_bounds
g_914-T1 1 1 2 + errors(2): exon-1:out_of_bounds exon-2:out_of_bounds
g_886-T1 1 1 2 + errors(2): exon-1:out_of_bounds exon-2:out_of_bounds
g_887-T1 1 1 2 + errors(2): exon-1:out_of_bounds exon-2:out_of_bounds
g_888-T1 1 1 3 + errors(3): exon-1:out_of_bounds exon-2:out_of_bounds exon-3:out_of_bounds
g_925-T1 1 1 2 + errors(2): exon-1:out_of_bounds exon-2:out_of_bounds
g_889-T1 1 1 3 + errors(3): exon-1:out_of_bounds exon-2:out_of_bounds exon-3:out_of_bounds
g_890-T1 1 1 4 + errors(4): exon-1:out_of_bounds exon-2:out_of_bounds exon-3:out_of_bounds exon-4:out_of_bounds
g_927-T1 1 1 2 + errors(2): exon-1:out_of_bounds exon-2:out_of_bounds
g_906-T1 1 1 3 + errors(3): exon-1:out_of_bounds exon-2:out_of_bounds exon-3:out_of_bounds
g_891-T1 1 1 3 + errors(3): exon-1:out_of_bounds exon-2:out_of_bounds exon-3:out_of_bounds
I saw similar issues on GitHub, but my training data comes from PASA, not BUSCO, and I don't understand where the problem lies. Additionally, I am missing SignalP.

[hyphaltip]

Were there stats from training that indicated number of predicted genes. It's unclear where this problem is coming from as I don't see with the current config I have and training and predict so need a way to replicate.

What are the g_XXX-T1 values coming from. Are those names in the trinity file?

What is the version of Augustus installed?