diff --git a/load_vocab.py b/load_vocab.py
index 82ca400a03decf6405adbf5d12042d8a03471c9f..5a4f1f921636124a367d6c9ff8cbc1822b010e0f 100644
--- a/load_vocab.py
+++ b/load_vocab.py
@@ -4,6 +4,8 @@ import simfile
import os
from io import StringIO
from simfile.notes import NoteData
+from simfile.notes.timed import time_notes
+from simfile.timing import TimingData
def filter_whitespaces(notes):
@@ -24,9 +26,9 @@ def index_tokens(token_sequence):
def built_chart_vocabulary(song_dirs):
vocabulary_set = set()
- cols = 4
+ cols_init = 4
# once add zero string
- note_string = ["0"] * cols
+ note_string = ["0"] * cols_init
notedata_new = StringIO()
notedata_new.write("".join(note_string))
# print(notedata_new.getvalue())
@@ -46,57 +48,35 @@ def built_chart_vocabulary(song_dirs):
charts = sm_file.charts
# print("Built vocabulary")
for chart in charts:
- # print(chart.notes)
- # notes ist string
- #print(type(chart.notes))
- # from notes
- notedata = NoteData(chart)
- notes = NoteData.from_notes(notedata, cols)
- new_notes = []
-
-
- for i, note in enumerate(notes):
- note_string = ["0"] * cols
- for j, note_2 in enumerate(notes):
- if note.beat == note_2.beat:
- note_string[note.column] = str(note)
- note_string[note_2.column] = str(note_2)
- if note_string == ["0"] * cols:
- note_string[note.column] = str(note)
- notedata_new = StringIO()
- notedata_new.write("".join(note_string))
- #print(notedata_new.getvalue())
- new_notes.append(notedata_new.getvalue())
-
- #for note in notes:
- # new_notes.append(str(note))
- #print(notes)
- #print(str(notes))
- #print(new_notes)
- vocabulary_set.update(set(new_notes))
- #notes = NoteData.from_notes(notedata, cols)
- #notes = str(notes)
- #notes = chart.notes.replace('\n,\n', '\n')
- #notes = notes.replace(',', '')
- # print(chart.notes)
- #notes = filter_whitespaces(notes)
- #notes = notes.split("\n")
- # print(set(notes))
- #vocabulary_set.update(set(notes))
-
- # update vocabulary set if there are whitespaces and length not 4
- # updated_vocabulary_set = set()
- # for elem in vocabuary_set:
- # if len(elem) != 4:
- # elem = re.sub(r"\s+", "", elem, flags=re.UNICODE)
- #
- # if len(elem) == 4:
- # updated_vocabulary_set.update([elem])
+ # notes is string
+ note_data = NoteData(chart)
+ timing_data = TimingData(sm_file)
+ cols = note_data.columns
+
+ # dictionary
+ tmp = dict()
+ for timed_note in time_notes(note_data, timing_data):
+ if timed_note.time in tmp.keys():
+ tmp[timed_note.time] = tmp[timed_note.time][:timed_note.note.column] + str(timed_note.note) + \
+ tmp[timed_note.time][timed_note.note.column + 1:]
+ else:
+ note_string = ["0"] * cols
+ note_string[timed_note.note.column] = str(timed_note.note)
+ note_string_new = StringIO()
+ note_string_new.write("".join(note_string))
+ tmp[timed_note.time] = note_string_new.getvalue()
+
+ vocabulary_set.update(tmp.values())
+
+ # alternative dictionary
+ #tmp = dict()
+ #for i, note in enumerate(notes):
+ # tmp[note.beat].append(note)
+
+
print("----------------------------------")
print("Built Vocabulary:")
- #print(updated_vocabulary_set)
print(vocabulary_set)
- #indexed_vocabulary = index_tokens(updated_vocabulary_set)
indexed_vocabulary = index_tokens(vocabulary_set)
print("Indexing:")
print(indexed_vocabulary)
diff --git a/vocabulary_run.sbatch b/vocabulary_run.sbatch
index 37555ecbbf76a36fe4af66c6e1d353d5eecf341a..a7e7d8312334a2ead5ff9940446340714a336cbb 100644
--- a/vocabulary_run.sbatch
+++ b/vocabulary_run.sbatch
@@ -1,14 +1,13 @@
#!/bin/bash
-#SBATCH -t 1-00:00:00 # time limit set to 1 week
+#SBATCH -t 0-00:10:00 # time limit set to 10 minutes, 1 day: 1-00:00:00
#SBATCH --mem=1G # reserve 1GB of memory
-#SBATCH -J Load_Vocabulary # the job name
+#SBATCH -J Load_Vocabulary # the job name
#SBATCH --mail-type=END,FAIL,TIME_LIMIT # send notification emails
#SBATCH -n 5 # use 5 tasks
-#SBATCH --cpus-per-task=1 # use 1 thread per taks
+#SBATCH --cpus-per-task=1 # use 1 thread per task
#SBATCH -N 1 # request slots on 1 node
-#SBATCH --partition=informatik-mind # run on one of our DGX servers
-#SBATCH --output=/scratch/grzonkow/vocabulary_3.txt # capture output
-#SBATCH --error=/scratch/grzonkow/err_3.txt # and error streams
+#SBATCH --output=/scratch/grzonkow/vocabulary.txt # capture output
+#SBATCH --error=/scratch/grzonkow/err.txt # and error streams
module load anaconda3/latest
. $ANACONDA_HOME/etc/profile.d/conda.sh