Last used & min count settings

Also some performance improvements
This commit is contained in:
DominikDoom
2023-11-29 17:45:51 +01:00
parent 15478e73b5
commit a156214a48
4 changed files with 41 additions and 25 deletions

View File

@@ -173,7 +173,16 @@ function flatten(obj, roots = [], sep = ".") {
}
// Calculate biased tag score based on post count and frequent usage
function calculateUsageBias(count, uses) {
function calculateUsageBias(result, count, uses, lastUseDate) {
// Guard for minimum usage count & last usage date
const diffTime = Math.abs(Date.now() - (lastUseDate || Date.now()));
const diffDays = Math.ceil(diffTime / (1000 * 60 * 60 * 24));
if (uses < TAC_CFG.frequencyMinCount || diffDays > TAC_CFG.frequencyMaxAge) {
uses = 0;
} else if (uses != 0) {
result.usageBias = true;
}
switch (TAC_CFG.frequencyFunction) {
case "Logarithmic (weak)":
return Math.log(1 + count) + Math.log(1 + uses);
@@ -187,7 +196,14 @@ function calculateUsageBias(count, uses) {
}
// Beautify return type for easier parsing
function mapUseCountArray(useCounts) {
return useCounts.map(useCount => {return {"name": useCount[0], "type": useCount[1], "count": useCount[2]}});
return useCounts.map(useCount => {
return {
"name": useCount[0],
"type": useCount[1],
"count": useCount[2],
"lastUseDate": useCount[3]
}
});
}
// Call API endpoint to increase bias of tag in the database
function increaseUseCount(tagName, type, negative = false) {

View File

@@ -1,4 +1,4 @@
const styleColors = {
const styleColors = {
"--results-bg": ["#0b0f19", "#ffffff"],
"--results-border-color": ["#4b5563", "#e5e7eb"],
"--results-border-width": ["1px", "1.5px"],
@@ -224,6 +224,8 @@ async function syncOptions() {
modelSortOrder: opts["tac_modelSortOrder"],
frequencySort: opts["tac_frequencySort"],
frequencyFunction: opts["tac_frequencyFunction"],
frequencyMinCount: opts["tac_frequencyMinCount"],
frequencyMaxAge: opts["tac_frequencyMaxAge"],
// Insertion related settings
replaceUnderscores: opts["tac_replaceUnderscores"],
escapeParentheses: opts["tac_escapeParentheses"],
@@ -1169,7 +1171,6 @@ async function autocomplete(textArea, prompt, fixedTag = null) {
// Request use counts from the DB
const counts = await getUseCounts(names, types, isNegative);
const usedResults = counts.filter(c => c.count > 0).map(c => c.name);
// Sort all
results = results.sort((a, b) => {
@@ -1178,19 +1179,16 @@ async function autocomplete(textArea, prompt, fixedTag = null) {
const aUseStats = counts.find(c => c.name === aName && c.type === a.type);
const bUseStats = counts.find(c => c.name === bName && c.type === b.type);
const aUses = aUseStats?.count || 0;
const bUses = bUseStats?.count || 0;
const aLastUseDate = Date.parse(aUseStats?.lastUseDate);
const bLastUseDate = Date.parse(bUseStats?.lastUseDate);
const aWeight = calculateUsageBias(a.count, aUseStats ? aUseStats.count : 0);
const bWeight = calculateUsageBias(b.count, bUseStats ? bUseStats.count : 0);
const aWeight = calculateUsageBias(a, a.count, aUses, aLastUseDate);
const bWeight = calculateUsageBias(b, b.count, bUses, bLastUseDate);
return bWeight - aWeight;
});
// Mark results
results.forEach(r => {
const name = r.type === ResultType.chant ? r.aliases : r.text;
if (usedResults.includes(name))
r.usageBias = true;
});
}
// Slice if the user has set a max result count and we are not in a extra networks / wildcard list

View File

@@ -421,6 +421,8 @@ def on_ui_settings():
"tac_modelSortOrder": shared.OptionInfo("Name", "Model sort order", gr.Dropdown, lambda: {"choices": list(sort_criteria.keys())}).info("Order for extra network models and wildcards in dropdown"),
"tac_frequencySort": shared.OptionInfo(True, "Locally record tag usage and sort frequent tags higher").info("Will also work for extra networks, keeping the specified base order"),
"tac_frequencyFunction": shared.OptionInfo("Logarithmic (weak)", "Function to use for frequency sorting", gr.Dropdown, lambda: {"choices": list(frequency_sort_functions.keys())}).info("; ".join([f'<b>{key}</b>: {val}' for key, val in frequency_sort_functions.items()])),
"tac_frequencyMinCount": shared.OptionInfo(3, "Minimum number of uses for a tag to be considered frequent").info("Tags with less uses than this will not be sorted higher, even if the sorting function would normally result in a higher position."),
"tac_frequencyMaxAge": shared.OptionInfo(30, "Maximum days since last use for a tag to be considered frequent").info("Similar to the above, tags that haven't been used in this many days will not be sorted higher."),
# Insertion related settings
"tac_replaceUnderscores": shared.OptionInfo(True, "Replace underscores with spaces on insertion"),
"tac_escapeParentheses": shared.OptionInfo(True, "Escape parentheses on insertion"),
@@ -597,7 +599,7 @@ def api_tac(_: gr.Blocks, app: FastAPI):
if get:
ret = func()
if ret is list:
ret = [{"name": t[0], "type": t[1], "count": t[2]} for t in ret]
ret = [{"name": t[0], "type": t[1], "count": t[2], "lastUseDate": t[3]} for t in ret]
return JSONResponse({"result": ret})
else:
func()

View File

@@ -91,7 +91,7 @@ class TagFrequencyDb:
with transaction() as cursor:
cursor.execute(
f"""
SELECT name, type, {count_str}
SELECT name, type, {count_str}, last_used
FROM tag_frequency
ORDER BY {count_str} DESC
"""
@@ -105,15 +105,15 @@ class TagFrequencyDb:
with transaction() as cursor:
cursor.execute(
f"""
SELECT {count_str}
SELECT {count_str}, last_used
FROM tag_frequency
WHERE name = ? AND type = ?
""",
(tag,ttype),
(tag, ttype),
)
tag_count = cursor.fetchone()
return tag_count[0] if tag_count else 0
return tag_count[0], tag_count[1] if tag_count else 0
def get_tag_counts(self, tags: list[str], ttypes: list[str], negative=False):
count_str = "count_neg" if negative else "count_pos"
@@ -121,18 +121,18 @@ class TagFrequencyDb:
for tag, ttype in zip(tags, ttypes):
cursor.execute(
f"""
SELECT {count_str}
SELECT {count_str}, last_used
FROM tag_frequency
WHERE name = ? AND type = ?
""",
(tag,ttype),
(tag, ttype),
)
tag_count = cursor.fetchone()
yield (tag, ttype, tag_count[0]) if tag_count else (tag, ttype, 0)
yield (tag, ttype, tag_count[0], tag_count[1]) if tag_count else (tag, ttype, 0)
def increase_tag_count(self, tag, ttype, negative=False):
pos_count = self.get_tag_count(tag, ttype, False)
neg_count = self.get_tag_count(tag, ttype, True)
pos_count = self.get_tag_count(tag, ttype, False)[0]
neg_count = self.get_tag_count(tag, ttype, True)[0]
if negative:
neg_count += 1
@@ -156,7 +156,7 @@ class TagFrequencyDb:
set_str = "count_pos = 0"
elif negative:
set_str = "count_neg = 0"
with transaction() as cursor:
cursor.execute(
f"""
@@ -164,5 +164,5 @@ class TagFrequencyDb:
SET {set_str}
WHERE name = ? AND type = ?
""",
(tag,ttype),
(tag, ttype),
)