add the model and python app

app.py

@@ -0,0 +1,149 @@
+import gradio as gr
+from setfit import SetFitModel
+from sentence_transformers import SentenceTransformer, util
+import re
+import numpy as np
+
+print("Loading Models...")
+try:
+    food_model = SetFitModel.from_pretrained("filter-search-model")
+    vector_model = SentenceTransformer('all-MiniLM-L6-v2')
+    print("Models Loaded Successfully!")
+except Exception as e:
+    print(f"Error loading models: {e}")
+    print("Did you upload the 'filter-search-model' folder correctly?")
+
+tag_definitions = {
+    # --- RELIGION ---
+    "religion": {
+        'none': 'secular non-religious atheism no religion',
+        'taoist': 'taoist taoism yin yang dao',
+        'chinese_folk': 'chinese folk religion ancestor worship shenism',
+        'buddhist': 'buddhist temple buddhism monk zen lotus',
+        'christian': 'christian church jesus catholic protestant cross bible',
+        'hinduism': 'hindu temple hinduism shiva vishnu',
+        'vietnamese_folk': 'vietnamese folk religion mother goddess dao mau thanh mau',
+        'khmer': 'khmer religion theravada buddhism cambodian style',
+        'muslim': 'muslim islam mosque halal allah'
+    },
+    # --- ARCHITECTURE STYLE ---
+    "arch_style": {
+        'french': 'french colonial architecture indochina villa balcony yellow walls shutters',
+        'modernism': 'modern architecture modernism brutalist concrete glass sleek minimal',
+        'east asian': 'east asian oriental wooden structure curved roof',
+        'contemporary': 'contemporary design new recent 21st century current',
+        'baroque': 'baroque architecture ornate dramatic detailed grandiose european',
+        'rococo': 'rococo style decorative pastel intricate playful',
+        'vietnamese': 'traditional vietnamese architecture wooden red tile roof',
+        'chinese': 'chinese architecture dragon courtyard feng shui',
+        'romanesque': 'romanesque arches thick walls sturdy rounded',
+        'gothic': 'gothic architecture pointed arches stained glass spires cathedral',
+        'renaissance': 'renaissance symmetry domes columns classical proportion',
+        'neo-romanesque': 'neo-romanesque revival 19th century style',
+        'indochine': 'indochine style fusion french colonial and vietnamese tropical',
+        'neo-baroque': 'neo-baroque revival grand opera house style',
+        'hindu': 'hindu architecture cham style champa towers sandstone',
+        'art deco': 'art deco retro geometric stylized 1920s 1930s',
+        'khmer': 'khmer architecture angkor wat style stone temple prasat',
+        'islamic': 'islamic architecture domes minarets geometric patterns',
+        'flamboyant': 'flamboyant gothic flame-like intricate stone tracery',
+        'beaux arts': 'beaux arts style monumental classical grand french school',
+        'art nouveau': 'art nouveau organic lines flowery decorative curved'
+    },
+    # --- BUILDING TYPE ---
+    "building_type": {
+        'marketplace': 'marketplace market bazaar shopping buy food groceries ben thanh',
+        'community_centre': 'community center public gathering cultural house hall',
+        'courthouse': 'courthouse law legal judge justice government',
+        'hospital': 'hospital clinic doctor medical emergency healthcare',
+        'mortuary': 'mortuary funeral home dead burial services',
+        'place_of_worship': 'place of worship temple church pagoda shrine mosque pray spiritual',
+        'institution': 'institution school university college education academy library',
+        'tomb': 'tomb grave mausoleum cemetery burial site',
+        'heritage': 'heritage site historical building landmark monument preservation',
+        'zoo': 'zoo animals botanical garden nature park',
+        'museum': 'museum exhibition history art gallery display',
+        'commercial': 'commercial building shop store mall office business trade',
+        'government': 'government building administrative town hall committee',
+        'residential': 'residential house home villa apartment living',
+        'library': 'library books reading study archive',
+        'fountain': 'fountain water feature square plaza',
+        'theatre': 'theatre opera house performance stage cinema arts',
+        'pagoda': 'pagoda tiered tower buddhist temple asian',
+        'park': 'park green space garden nature relax trees',
+        'reservoir': 'reservoir water lake dam utility',
+        'restaurant': 'restaurant place to eat food dining hungry lunch dinner eatery',
+        'cafe': 'cafe coffee shop drink sip tea chill work',
+    }
+}
+
+# --- 3. PRE-COMPUTE VECTORS ---
+print("Pre-computing vectors...")
+encoded_db = {}
+for category, tags_map in tag_definitions.items():
+    encoded_db[category] = {
+        'embeddings': vector_model.encode(list(tags_map.values()), convert_to_tensor=True),
+        'names': list(tags_map.keys())
+    }
+print("Ready.")
+
+# --- 4. THE EXTRACTION LOGIC ---
+def extract_tags(user_text):
+    detected_data = {}
+    
+    # A. Age Extraction
+    match = re.search(r'\b(1[7-9]|20)\d{2}s?\b', user_text)
+    detected_data['age'] = match.group(0).rstrip('s') if match else None
+
+    # B. Vector Search (Arch/Religion/Building)
+    user_embedding = vector_model.encode(user_text, convert_to_tensor=True)
+    base_threshold = 0.28
+    
+    for category in ["religion", "arch_style", "building_type"]:
+        db_data = encoded_db[category]
+        scores = util.cos_sim(user_embedding, db_data['embeddings'])[0]
+        
+        candidates = []
+        for idx, score in enumerate(scores):
+            if score > base_threshold:
+                candidates.append((db_data['names'][idx], score.item()))
+        
+        candidates.sort(key=lambda x: x[1], reverse=True)
+        
+        if candidates:
+            best_score = candidates[0][1]
+            relative_cutoff = best_score * 0.85 
+            final_tags = [t[0] for t in candidates if t[1] >= relative_cutoff]
+            detected_data[category] = final_tags
+        else:
+            detected_data[category] = None
+
+    # C. Smart Food Logic (SetFit)
+    food_triggers = ['restaurant', 'cafe', 'street_food', 'marketplace']
+    buildings = detected_data.get('building_type') or []
+    is_food_venue = any(b in food_triggers for b in buildings)
+    
+    # Run Prediction
+    prediction_array = food_model.predict([user_text])
+    prediction_clean = prediction_array.tolist()
+    
+    # Trust logic: Only return food tags if venue is found OR if the prediction is extremely confident
+    # For now, we stick to your venue rule:
+    if is_food_venue:
+        detected_data['food_type'] = prediction_clean
+    else:
+        detected_data['food_type'] = None
+
+    return detected_data
+
+# --- 5. LAUNCH API ---
+# We define the Input/Output types clearly for the API
+iface = gr.Interface(
+    fn=extract_tags,
+    inputs=gr.Textbox(label="User Search", placeholder="e.g. I want a french villa"),
+    outputs=gr.JSON(label="Extracted Tags"),
+    title="Vietnam Travel AI API",
+    description="Send a POST request to /api/predict with {'data': ['your query']} to get JSON tags."
+)
+
+iface.launch()

filter-search-model/1_Pooling/config.json

@@ -0,0 +1,10 @@
+{
+    "word_embedding_dimension": 384,
+    "pooling_mode_cls_token": false,
+    "pooling_mode_mean_tokens": true,
+    "pooling_mode_max_tokens": false,
+    "pooling_mode_mean_sqrt_len_tokens": false,
+    "pooling_mode_weightedmean_tokens": false,
+    "pooling_mode_lasttoken": false,
+    "include_prompt": true
+}

filter-search-model/README.md

@@ -0,0 +1,393 @@
+---
+tags:
+- setfit
+- sentence-transformers
+- text-classification
+- generated_from_setfit_trainer
+widget:
+- text: Show me a university location
+- text: Best vietnam restaurant
+- text: crepe pancake french dessert sweet
+- text: Where can I find pot food?
+- text: fountain water feature square plaza
+metrics:
+- accuracy
+pipeline_tag: text-classification
+library_name: setfit
+inference: true
+base_model: sentence-transformers/all-MiniLM-L6-v2
+---
+
+# SetFit with sentence-transformers/all-MiniLM-L6-v2
+
+This is a [SetFit](https://github.com/huggingface/setfit) model that can be used for Text Classification. This SetFit model uses [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2) as the Sentence Transformer embedding model. A [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance is used for classification.
+
+The model has been trained using an efficient few-shot learning technique that involves:
+
+1. Fine-tuning a [Sentence Transformer](https://www.sbert.net) with contrastive learning.
+2. Training a classification head with features from the fine-tuned Sentence Transformer.
+
+## Model Details
+
+### Model Description
+- **Model Type:** SetFit
+- **Sentence Transformer body:** [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2)
+- **Classification head:** a [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance
+- **Maximum Sequence Length:** 256 tokens
+- **Number of Classes:** 90 classes
+<!-- - **Training Dataset:** [Unknown](https://huggingface.co/datasets/unknown) -->
+<!-- - **Language:** Unknown -->
+<!-- - **License:** Unknown -->
+
+### Model Sources
+
+- **Repository:** [SetFit on GitHub](https://github.com/huggingface/setfit)
+- **Paper:** [Efficient Few-Shot Learning Without Prompts](https://arxiv.org/abs/2209.11055)
+- **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
+
+### Model Labels
+| Label                 | Examples                                                                                                                                            |
+|:----------------------|:----------------------------------------------------------------------------------------------------------------------------------------------------|
+| mexican               | <ul><li>'Best salsa restaurant'</li><li>'Best taco restaurant'</li><li>'Looking for a taco meal'</li></ul>                                          |
+| korean                | <ul><li>'Looking for a k-food meal'</li><li>'Serving k-food dishes'</li><li>'I want to eat k-food'</li></ul>                                        |
+| zoo                   | <ul><li>'Find a nature in the city'</li><li>'Find a zoo in the city'</li><li>'I need to go to a zoo'</li></ul>                                      |
+| lebanese              | <ul><li>'Looking for a hummus meal'</li><li>'I need some middle'</li><li>'I want to eat lebanese'</li></ul>                                         |
+| chinese               | <ul><li>'Best noodles restaurant'</li><li>'Best wok restaurant'</li><li>'Delicious wok cuisine'</li></ul>                                           |
+| tomb                  | <ul><li>'Find a tomb in the city'</li><li>'Find a grave in the city'</li><li>'Where is the nearest tomb?'</li></ul>                                 |
+| french                | <ul><li>'I am craving fine'</li><li>'Examples of colonial structures'</li><li>'I want to eat fine'</li></ul>                                        |
+| thai                  | <ul><li>'Best thai restaurant'</li><li>'Serving pad dishes'</li><li>'I need some curry'</li></ul>                                                   |
+| steamed fish          | <ul><li>'I am craving steamed fish'</li><li>'I need some steamed fish'</li><li>'Looking for a steamed fish meal'</li></ul>                          |
+| modernism             | <ul><li>'Show me sleek architecture'</li><li>'Find me a sleek villa'</li><li>'Show me concrete architecture'</li></ul>                              |
+| diner                 | <ul><li>'Delicious comfort cuisine'</li><li>'I need some cheap'</li><li>'Best eatery restaurant'</li></ul>                                          |
+| marketplace           | <ul><li>'Show me a marketplace location'</li><li>'I need to go to a groceries'</li><li>'Find a market in the city'</li></ul>                        |
+| reservoir             | <ul><li>'Show me a dam location'</li><li>'reservoir water lake dam utility'</li><li>'Find a reservoir in the city'</li></ul>                        |
+| coffee                | <ul><li>'I am craving coffee'</li><li>'Delicious phe cuisine'</li><li>'Looking for a cappuccino meal'</li></ul>                                     |
+| hue style             | <ul><li>'I am craving vietnam'</li><li>'Looking for a noodle meal'</li><li>'Where can I find central food?'</li></ul>                               |
+| khmer                 | <ul><li>'Examples of stone structures'</li><li>'Locations for cambodian followers'</li><li>'Find me a temple villa'</li></ul>                       |
+| commercial            | <ul><li>'Where is the nearest business?'</li><li>'Show me a mall location'</li><li>'Where is the nearest commercial?'</li></ul>                     |
+| japanese              | <ul><li>'Delicious sashimi cuisine'</li><li>'Best udon restaurant'</li><li>'Looking for a tempura meal'</li></ul>                                   |
+| vietnamese            | <ul><li>'Best cuisine restaurant'</li><li>'Where can I find local food?'</li><li>'Looking for a tam meal'</li></ul>                                 |
+| contemporary          | <ul><li>'Looking for recent vibes'</li><li>'Find me a design villa'</li><li>'Looking for 21st vibes'</li></ul>                                      |
+| pizza                 | <ul><li>'I need some pizza'</li><li>'I am craving italian'</li><li>'Where can I find pie food?'</li></ul>                                           |
+| theatre               | <ul><li>'I need to go to a theatre'</li><li>'Show me a theatre location'</li><li>'theatre opera house performance stage cinema arts'</li></ul>      |
+| library               | <ul><li>'I need to go to a study'</li><li>'Where is the nearest archive?'</li><li>'Find a books in the city'</li></ul>                              |
+| neo-romanesque        | <ul><li>'Find me a 19th villa'</li><li>'Show me 19th architecture'</li><li>'Looking for 19th vibes'</li></ul>                                       |
+| sushi                 | <ul><li>'Looking for a fish meal'</li><li>'I am craving maki'</li><li>'I am craving nigiri'</li></ul>                                               |
+| nướng (grill)         | <ul><li>'I want to eat nướng (grill)'</li><li>'I am craving nướng (grill)'</li><li>'Serving nướng (grill) dishes'</li></ul>                         |
+| indochine             | <ul><li>'Find me a and villa'</li><li>'I want to see a fusion style building'</li><li>'Houses with and design'</li></ul>                            |
+| mediterranean         | <ul><li>'I want to eat greek'</li><li>'Delicious healthy cuisine'</li><li>'Serving greek dishes'</li></ul>                                          |
+| sandwich              | <ul><li>'Serving sub dishes'</li><li>'I want to eat panini'</li><li>'Looking for a sub meal'</li></ul>                                              |
+| wings                 | <ul><li>'I want to eat wings'</li><li>'Serving buffalo dishes'</li><li>'Where can I find buffalo food?'</li></ul>                                   |
+| international         | <ul><li>'Looking for a foreign meal'</li><li>'Serving western dishes'</li><li>'I want to eat imported'</li></ul>                                    |
+| mortuary              | <ul><li>'Where is the nearest burial?'</li><li>'Show me a mortuary location'</li><li>'Show me a services location'</li></ul>                        |
+| ice cream             | <ul><li>'I need some sweet'</li><li>'Delicious cold cuisine'</li><li>'Serving gelato dishes'</li></ul>                                              |
+| baroque               | <ul><li>'Houses with baroque design'</li><li>'Find me a ornate villa'</li><li>'I want to see a european style building'</li></ul>                   |
+| romanesque            | <ul><li>'Show me rounded architecture'</li><li>'Looking for thick vibes'</li><li>'Houses with thick design'</li></ul>                               |
+| hospital              | <ul><li>'I need to go to a doctor'</li><li>'Show me a emergency location'</li><li>'Where is the nearest clinic?'</li></ul>                          |
+| art nouveau           | <ul><li>'I want to see a lines style building'</li><li>'Show me lines architecture'</li><li>'Looking for organic vibes'</li></ul>                   |
+| curry                 | <ul><li>'Delicious coconut cuisine'</li><li>'I want to eat coconut'</li><li>'I want to eat stew'</li></ul>                                          |
+| neo-baroque           | <ul><li>'Looking for house vibes'</li><li>'Houses with opera design'</li><li>'Show me opera architecture'</li></ul>                                 |
+| argentinian           | <ul><li>'Where can I find argentinian food?'</li><li>'Serving american dishes'</li><li>'argentinian steak latin american meat asado'</li></ul>      |
+| grill                 | <ul><li>'I want to eat fire'</li><li>'Best smoke restaurant'</li><li>'Best meat restaurant'</li></ul>                                               |
+| christian             | <ul><li>'Locations for church followers'</li><li>'I want to visit a catholic place of worship'</li><li>'Locations for cross followers'</li></ul>    |
+| burger                | <ul><li>'Delicious burger cuisine'</li><li>'Best fast restaurant'</li><li>'I need some fast'</li></ul>                                              |
+| rococo                | <ul><li>'Looking for pastel vibes'</li><li>'Find me a decorative villa'</li><li>'Show me playful architecture'</li></ul>                            |
+| muslim                | <ul><li>'Find a mosque temple'</li><li>'Locations for muslim followers'</li><li>'Find a halal temple'</li></ul>                                     |
+| hinduism              | <ul><li>'Find a hindu temple'</li><li>'I want to visit a hindu place of worship'</li><li>'I want to visit a shiva place of worship'</li></ul>       |
+| seafood               | <ul><li>'Looking for a snail meal'</li><li>'Delicious crab cuisine'</li><li>'I need some shell'</li></ul>                                           |
+| asian                 | <ul><li>'Looking for a eastern meal'</li><li>'I need some oriental'</li><li>'I want to eat asian'</li></ul>                                         |
+| cafe                  | <ul><li>'Find a sip in the city'</li><li>'Where is the nearest drink?'</li><li>'I need to go to a drink'</li></ul>                                  |
+| park                  | <ul><li>'I need to go to a trees'</li><li>'Show me a space location'</li><li>'Find a relax in the city'</li></ul>                                   |
+| local food            | <ul><li>'Where can I find specialty food?'</li><li>'Looking for a street meal'</li><li>'Serving specialty dishes'</li></ul>                         |
+| thai hotpot           | <ul><li>'I need some sour'</li><li>'Best sour restaurant'</li><li>'Best thai hotpot restaurant'</li></ul>                                           |
+| flamboyant            | <ul><li>'Examples of flame-like structures'</li><li>'Houses with flame-like design'</li><li>'Find me a flame-like villa'</li></ul>                  |
+| gothic                | <ul><li>'Show me pointed architecture'</li><li>'Show me gothic architecture'</li><li>'Examples of gothic structures'</li></ul>                      |
+| buddhist              | <ul><li>'I want to visit a monk place of worship'</li><li>'Locations for buddhism followers'</li><li>'Locations for buddhist followers'</li></ul>   |
+| art deco              | <ul><li>'I want to see a stylized style building'</li><li>'Houses with retro design'</li><li>'Find me a retro villa'</li></ul>                      |
+| pagoda                | <ul><li>'Where is the nearest asian?'</li><li>'Show me a tiered location'</li><li>'Where is the nearest tower?'</li></ul>                           |
+| islamic               | <ul><li>'I want to see a patterns style building'</li><li>'Examples of minarets structures'</li><li>'Examples of patterns structures'</li></ul>     |
+| heritage              | <ul><li>'Where is the nearest heritage?'</li><li>'Find a monument in the city'</li><li>'I need to go to a heritage'</li></ul>                       |
+| government            | <ul><li>'Find a committee in the city'</li><li>'government building administrative town hall committee'</li><li>'Show me a town location'</li></ul> |
+| steakhouse            | <ul><li>'Delicious steakhouse cuisine'</li><li>'Serving steakhouse dishes'</li><li>'I am craving steakhouse'</li></ul>                              |
+| hotpot                | <ul><li>'I want to eat pot'</li><li>'hotpot lau soup pot shabu shabu sharing'</li><li>'Where can I find hotpot food?'</li></ul>                     |
+| banh cuon             | <ul><li>'Delicious breakfast cuisine'</li><li>'Looking for a rice meal'</li><li>'I am craving light'</li></ul>                                      |
+| noodles               | <ul><li>'Where can I find vermicelli food?'</li><li>'I am craving soup'</li><li>'I need some vermicelli'</li></ul>                                  |
+| fine dining           | <ul><li>'I am craving fine dining'</li><li>'I am craving upscale'</li><li>'Looking for a luxury meal'</li></ul>                                     |
+| stir-fried vegetables | <ul><li>'Serving vegetarian dishes'</li><li>'I need some vegetarian'</li><li>'Best glory restaurant'</li></ul>                                      |
+| courthouse            | <ul><li>'I need to go to a law'</li><li>'Find a legal in the city'</li><li>'I need to go to a justice'</li></ul>                                    |
+| none                  | <ul><li>'I want to visit a atheism place of worship'</li><li>'Find a atheism temple'</li><li>'Locations for atheism followers'</li></ul>            |
+| grilled fish          | <ul><li>'Where can I find grilled fish food?'</li><li>'grilled fish ca nuong seafood'</li><li>'Best grilled fish restaurant'</li></ul>              |
+| italian               | <ul><li>'I want to eat pasta'</li><li>'Delicious rome cuisine'</li><li>'Where can I find spaghetti food?'</li></ul>                                 |
+| beef bowl             | <ul><li>'I need some gyudon'</li><li>'Where can I find gyudon food?'</li><li>'Where can I find beef bowl food?'</li></ul>                           |
+| east asian            | <ul><li>'Find me a roof villa'</li><li>'Show me roof architecture'</li><li>'Find me a structure villa'</li></ul>                                    |
+| museum                | <ul><li>'I need to go to a gallery'</li><li>'Find a exhibition in the city'</li><li>'Find a art in the city'</li></ul>                              |
+| place_of_worship      | <ul><li>'Show me a pray location'</li><li>'I need to go to a pagoda'</li><li>'Find a church in the city'</li></ul>                                  |
+| barbecue              | <ul><li>'Where can I find smokehouse food?'</li><li>'Serving brisket dishes'</li><li>'barbecue bbq ribs brisket smokehouse'</li></ul>               |
+| residential           | <ul><li>'I need to go to a villa'</li><li>'I need to go to a apartment'</li><li>'Show me a living location'</li></ul>                               |
+| renaissance           | <ul><li>'I want to see a domes style building'</li><li>'Houses with classical design'</li><li>'Find me a symmetry villa'</li></ul>                  |
+| crepe                 | <ul><li>'Best crepe restaurant'</li><li>'Serving pancake dishes'</li><li>'Serving crepe dishes'</li></ul>                                           |
+| institution           | <ul><li>'Show me a institution location'</li><li>'Show me a education location'</li><li>'Where is the nearest institution?'</li></ul>               |
+| indian                | <ul><li>'Where can I find naan food?'</li><li>'Serving masala dishes'</li><li>'I am craving naan'</li></ul>                                         |
+| community_centre      | <ul><li>'Show me a public location'</li><li>'Where is the nearest house?'</li><li>'Find a center in the city'</li></ul>                             |
+| fountain              | <ul><li>'Show me a water location'</li><li>'I need to go to a water'</li><li>'Find a water in the city'</li></ul>                                   |
+| hindu                 | <ul><li>'Find me a hindu villa'</li><li>'Houses with hindu design'</li><li>'I want to see a champa style building'</li></ul>                        |
+| chinese_folk          | <ul><li>'Locations for worship followers'</li><li>'Find a ancestor temple'</li><li>'Locations for shenism followers'</li></ul>                      |
+| restaurant            | <ul><li>'I need to go to a dining'</li><li>'Find a lunch in the city'</li><li>'Where is the nearest dinner?'</li></ul>                              |
+| beaux arts            | <ul><li>'Houses with school design'</li><li>'Show me beaux arts architecture'</li><li>'Houses with beaux arts design'</li></ul>                     |
+| vietnamese_folk       | <ul><li>'I want to visit a mother place of worship'</li><li>'Locations for thanh followers'</li><li>'Locations for mau followers'</li></ul>         |
+| american              | <ul><li>'Serving usa dishes'</li><li>'Delicious usa cuisine'</li><li>'american food usa burger fries steak'</li></ul>                               |
+| taoist                | <ul><li>'I want to visit a yin place of worship'</li><li>'I want to visit a taoism place of worship'</li><li>'Find a taoist temple'</li></ul>       |
+| pasta                 | <ul><li>'Delicious carbonara cuisine'</li><li>'I want to eat carbonara'</li><li>'Where can I find carbonara food?'</li></ul>                        |
+
+## Uses
+
+### Direct Use for Inference
+
+First install the SetFit library:
+
+```bash
+pip install setfit
+```
+
+Then you can load this model and run inference.
+
+```python
+from setfit import SetFitModel
+
+# Download from the 🤗 Hub
+model = SetFitModel.from_pretrained("setfit_model_id")
+# Run inference
+preds = model("Best vietnam restaurant")
+```
+
+<!--
+### Downstream Use
+
+*List how someone could finetune this model on their own dataset.*
+-->
+
+<!--
+### Out-of-Scope Use
+
+*List how the model may foreseeably be misused and address what users ought not to do with the model.*
+-->
+
+<!--
+## Bias, Risks and Limitations
+
+*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
+-->
+
+<!--
+### Recommendations
+
+*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
+-->
+
+## Training Details
+
+### Training Set Metrics
+| Training set | Min | Median | Max |
+|:-------------|:----|:-------|:----|
+| Word count   | 3   | 4.8024 | 10  |
+
+| Label                 | Training Sample Count |
+|:----------------------|:----------------------|
+| american              | 17                    |
+| argentinian           | 36                    |
+| art deco              | 32                    |
+| art nouveau           | 20                    |
+| asian                 | 24                    |
+| banh cuon             | 46                    |
+| barbecue              | 21                    |
+| baroque               | 34                    |
+| beaux arts            | 14                    |
+| beef bowl             | 15                    |
+| buddhist              | 23                    |
+| burger                | 23                    |
+| cafe                  | 27                    |
+| chinese               | 73                    |
+| chinese_folk          | 17                    |
+| christian             | 25                    |
+| coffee                | 36                    |
+| commercial            | 25                    |
+| community_centre      | 33                    |
+| contemporary          | 42                    |
+| courthouse            | 25                    |
+| crepe                 | 15                    |
+| curry                 | 24                    |
+| diner                 | 46                    |
+| east asian            | 31                    |
+| fine dining           | 37                    |
+| flamboyant            | 15                    |
+| fountain              | 18                    |
+| french                | 83                    |
+| gothic                | 28                    |
+| government            | 13                    |
+| grill                 | 48                    |
+| grilled fish          | 16                    |
+| heritage              | 20                    |
+| hindu                 | 31                    |
+| hinduism              | 17                    |
+| hospital              | 22                    |
+| hotpot                | 28                    |
+| hue style             | 43                    |
+| ice cream             | 36                    |
+| indian                | 30                    |
+| indochine             | 24                    |
+| institution           | 26                    |
+| international         | 34                    |
+| islamic               | 18                    |
+| italian               | 20                    |
+| japanese              | 45                    |
+| khmer                 | 44                    |
+| korean                | 37                    |
+| lebanese              | 31                    |
+| library               | 14                    |
+| local food            | 37                    |
+| marketplace           | 28                    |
+| mediterranean         | 45                    |
+| mexican               | 29                    |
+| modernism             | 40                    |
+| mortuary              | 24                    |
+| museum                | 23                    |
+| muslim                | 20                    |
+| neo-baroque           | 23                    |
+| neo-romanesque        | 18                    |
+| none                  | 20                    |
+| noodles               | 14                    |
+| nướng (grill)         | 16                    |
+| pagoda                | 15                    |
+| park                  | 16                    |
+| pasta                 | 8                     |
+| pizza                 | 35                    |
+| place_of_worship      | 28                    |
+| renaissance           | 30                    |
+| reservoir             | 16                    |
+| residential           | 16                    |
+| restaurant            | 27                    |
+| rococo                | 29                    |
+| romanesque            | 27                    |
+| sandwich              | 23                    |
+| seafood               | 39                    |
+| steakhouse            | 14                    |
+| steamed fish          | 17                    |
+| stir-fried vegetables | 43                    |
+| sushi                 | 36                    |
+| taoist                | 16                    |
+| thai                  | 45                    |
+| thai hotpot           | 23                    |
+| theatre               | 23                    |
+| tomb                  | 16                    |
+| vietnamese            | 77                    |
+| vietnamese_folk       | 18                    |
+| wings                 | 33                    |
+| zoo                   | 21                    |
+
+### Training Hyperparameters
+- batch_size: (16, 16)
+- num_epochs: (1, 16)
+- max_steps: 2000
+- sampling_strategy: oversampling
+- body_learning_rate: (2e-05, 1e-05)
+- head_learning_rate: 0.01
+- loss: CosineSimilarityLoss
+- distance_metric: cosine_distance
+- margin: 0.25
+- end_to_end: False
+- use_amp: False
+- warmup_proportion: 0.1
+- l2_weight: 0.01
+- seed: 42
+- run_name: my_fresh_run_v2
+- evaluation_strategy: steps
+- eval_max_steps: -1
+- load_best_model_at_end: True
+
+### Training Results
+| Epoch  | Step | Training Loss | Validation Loss |
+|:------:|:----:|:-------------:|:---------------:|
+| 0.0005 | 1    | 0.2922        | -               |
+| 0.025  | 50   | 0.2078        | -               |
+| 0.05   | 100  | 0.1794        | -               |
+| 0.075  | 150  | 0.1687        | -               |
+| 0.1    | 200  | 0.1498        | 0.1183          |
+| 0.125  | 250  | 0.1504        | -               |
+| 0.15   | 300  | 0.1304        | -               |
+| 0.175  | 350  | 0.1323        | -               |
+| 0.2    | 400  | 0.1249        | 0.1001          |
+| 0.225  | 450  | 0.1234        | -               |
+| 0.25   | 500  | 0.1223        | -               |
+| 0.275  | 550  | 0.1133        | -               |
+| 0.3    | 600  | 0.1106        | 0.0899          |
+| 0.325  | 650  | 0.1026        | -               |
+| 0.35   | 700  | 0.1069        | -               |
+| 0.375  | 750  | 0.0975        | -               |
+| 0.4    | 800  | 0.0994        | 0.0772          |
+| 0.425  | 850  | 0.1006        | -               |
+| 0.45   | 900  | 0.0979        | -               |
+| 0.475  | 950  | 0.0817        | -               |
+| 0.5    | 1000 | 0.0966        | 0.0683          |
+| 0.525  | 1050 | 0.0856        | -               |
+| 0.55   | 1100 | 0.082         | -               |
+| 0.575  | 1150 | 0.0767        | -               |
+| 0.6    | 1200 | 0.0788        | 0.0620          |
+| 0.625  | 1250 | 0.0705        | -               |
+| 0.65   | 1300 | 0.0713        | -               |
+| 0.675  | 1350 | 0.0711        | -               |
+| 0.7    | 1400 | 0.0713        | 0.0575          |
+| 0.725  | 1450 | 0.074         | -               |
+| 0.75   | 1500 | 0.0618        | -               |
+| 0.775  | 1550 | 0.0701        | -               |
+| 0.8    | 1600 | 0.0674        | 0.0539          |
+| 0.825  | 1650 | 0.0685        | -               |
+| 0.85   | 1700 | 0.0678        | -               |
+| 0.875  | 1750 | 0.0671        | -               |
+| 0.9    | 1800 | 0.0657        | 0.0517          |
+| 0.925  | 1850 | 0.0593        | -               |
+| 0.95   | 1900 | 0.0641        | -               |
+| 0.975  | 1950 | 0.0659        | -               |
+| 1.0    | 2000 | 0.0629        | 0.0508          |
+
+### Framework Versions
+- Python: 3.12.12
+- SetFit: 1.1.3
+- Sentence Transformers: 5.1.2
+- Transformers: 4.57.2
+- PyTorch: 2.9.0+cu126
+- Datasets: 4.0.0
+- Tokenizers: 0.22.1
+
+## Citation
+
+### BibTeX
+```bibtex
+@article{https://doi.org/10.48550/arxiv.2209.11055,
+    doi = {10.48550/ARXIV.2209.11055},
+    url = {https://arxiv.org/abs/2209.11055},
+    author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},
+    keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
+    title = {Efficient Few-Shot Learning Without Prompts},
+    publisher = {arXiv},
+    year = {2022},
+    copyright = {Creative Commons Attribution 4.0 International}
+}
+```
+
+<!--
+## Glossary
+
+*Clearly define terms in order to be accessible across audiences.*
+-->
+
+<!--
+## Model Card Authors
+
+*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
+-->
+
+<!--
+## Model Card Contact
+
+*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
+-->

filter-search-model/config.json

@@ -0,0 +1,25 @@
+{
+  "architectures": [
+    "BertModel"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "dtype": "float32",
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 384,
+  "initializer_range": 0.02,
+  "intermediate_size": 1536,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 6,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "transformers_version": "4.57.2",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

filter-search-model/config_sentence_transformers.json

@@ -0,0 +1,14 @@
+{
+  "__version__": {
+    "sentence_transformers": "5.1.2",
+    "transformers": "4.57.2",
+    "pytorch": "2.9.0+cu126"
+  },
+  "model_type": "SentenceTransformer",
+  "prompts": {
+    "query": "",
+    "document": ""
+  },
+  "default_prompt_name": null,
+  "similarity_fn_name": "cosine"
+}

filter-search-model/config_setfit.json

@@ -0,0 +1,95 @@
+{
+  "labels": [
+    "american",
+    "argentinian",
+    "art deco",
+    "art nouveau",
+    "asian",
+    "banh cuon",
+    "barbecue",
+    "baroque",
+    "beaux arts",
+    "beef bowl",
+    "buddhist",
+    "burger",
+    "cafe",
+    "chinese",
+    "chinese_folk",
+    "christian",
+    "coffee",
+    "commercial",
+    "community_centre",
+    "contemporary",
+    "courthouse",
+    "crepe",
+    "curry",
+    "diner",
+    "east asian",
+    "fine dining",
+    "flamboyant",
+    "fountain",
+    "french",
+    "gothic",
+    "government",
+    "grill",
+    "grilled fish",
+    "heritage",
+    "hindu",
+    "hinduism",
+    "hospital",
+    "hotpot",
+    "hue style",
+    "ice cream",
+    "indian",
+    "indochine",
+    "institution",
+    "international",
+    "islamic",
+    "italian",
+    "japanese",
+    "khmer",
+    "korean",
+    "lebanese",
+    "library",
+    "local food",
+    "marketplace",
+    "mediterranean",
+    "mexican",
+    "modernism",
+    "mortuary",
+    "museum",
+    "muslim",
+    "neo-baroque",
+    "neo-romanesque",
+    "none",
+    "noodles",
+    "n\u01b0\u1edbng (grill)",
+    "pagoda",
+    "park",
+    "pasta",
+    "pizza",
+    "place_of_worship",
+    "renaissance",
+    "reservoir",
+    "residential",
+    "restaurant",
+    "rococo",
+    "romanesque",
+    "sandwich",
+    "seafood",
+    "steakhouse",
+    "steamed fish",
+    "stir-fried vegetables",
+    "sushi",
+    "taoist",
+    "thai",
+    "thai hotpot",
+    "theatre",
+    "tomb",
+    "vietnamese",
+    "vietnamese_folk",
+    "wings",
+    "zoo"
+  ],
+  "normalize_embeddings": false
+}

filter-search-model/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45f9966ed1b4481b29786c463dc03ddea3ab6529ebbb4229a0c11ba64ea8ad16
+size 90864192

filter-search-model/model_head.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:220dc87103dba7842ddbc931c391a678dfd3ba39670ca503843bef1886204247
+size 285607

filter-search-model/modules.json

@@ -0,0 +1,20 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  },
+  {
+    "idx": 2,
+    "name": "2",
+    "path": "2_Normalize",
+    "type": "sentence_transformers.models.Normalize"
+  }
+]

filter-search-model/sentence_bert_config.json

@@ -0,0 +1,4 @@
+{
+    "max_seq_length": 256,
+    "do_lower_case": false
+}

filter-search-model/special_tokens_map.json

@@ -0,0 +1,37 @@
+{
+  "cls_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "[MASK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

filter-search-model/tokenizer_config.json

@@ -0,0 +1,65 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "extra_special_tokens": {},
+  "mask_token": "[MASK]",
+  "max_length": 128,
+  "model_max_length": 256,
+  "never_split": null,
+  "pad_to_multiple_of": null,
+  "pad_token": "[PAD]",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
+  "sep_token": "[SEP]",
+  "stride": 0,
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
+  "unk_token": "[UNK]"
+}

requirements.txt

@@ -0,0 +1,5 @@
+setfit
+sentence-transformers
+scikit-learn
+pandas
+gradio