I installed the mirDeep2. The version is miRDeep2.0.0.8 and last modification time is 21/04/2016.

bowtie --version
#bowtie version 0.12.9
#64-bit
#Built on igm1
#Sun Dec 16 14:36:32 EST 2012
#Compiler: gcc version 4.1.2 20080704 (Red Hat 4.1.2-50)
#Options: -O3 -m64  -Wl,--hash-style=both  
#Sizeof {int, long, long long, void*, size_t, off_t}: {4, 8, 8, 8, 8, 8}

# fetch the raw data from GEO (GSE78031) and decompress it using fastq-dump
# the raw data format is fastq, SRR3175618 (SRX1590020): SRR3175618.fastq
# fetch the reference genome for rat from UCSC (rn4) :genome.fa
# fetch the hairpin and mature sequence information from miRBase: mature.fa/hairpin.fa
bowtie-build genome.fa rn4
mapper.pl SRR3175618.fastq -e -h -j -i -l 18 -m -p rn4 -s bowtie_rn4.fa -t bowtie_rn4.arf
quantifier.pl -p hairpin.fa -m mature.fa -t Rat -r bowtie_rn4.fa

I checked the result,miRNAs_expressed_all_samples_23_11_2016_t_17_27_18.csv, and found that each mirRNA has zero count. partial output from the csv file:

# head -n 5 *.csv
#miRNA  read_count      precursor       total
rno-let-7a-5p   0.00    rno-let-7a-1    0.00
rno-let-7a-1-3p 0.00    rno-let-7a-1    0.00
rno-let-7a-5p   0.00    rno-let-7a-2    0.00
rno-let-7a-2-3p 0.00    rno-let-7a-2    0.00

I also used miRDeep2.pl to process the data, and the quantification of each mirRNA is also zero. I checked the report.log, and there was no obvious error message. I changed the dataset and still kept the same output, namely, zero read_count. Did I miss something here? Any suggestions?