vector-inspector 0.2.1__tar.gz → 0.2.3__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (35) hide show
  1. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/PKG-INFO +37 -16
  2. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/README.md +241 -225
  3. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/pyproject.toml +8 -1
  4. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/ui/main_window.py +1 -1
  5. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/tests/test_connections.py +60 -60
  6. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/tests/test_filter_service.py +101 -101
  7. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/tests/test_settings_service.py +101 -101
  8. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/tests/vector_inspector.py +35 -35
  9. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/__init__.py +0 -0
  10. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/__main__.py +0 -0
  11. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/core/__init__.py +0 -0
  12. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/core/connections/__init__.py +0 -0
  13. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/core/connections/base_connection.py +0 -0
  14. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/core/connections/chroma_connection.py +0 -0
  15. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/core/connections/qdrant_connection.py +0 -0
  16. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/core/connections/template_connection.py +0 -0
  17. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/main.py +0 -0
  18. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/services/__init__.py +0 -0
  19. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/services/backup_restore_service.py +0 -0
  20. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/services/filter_service.py +0 -0
  21. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/services/import_export_service.py +0 -0
  22. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/services/settings_service.py +0 -0
  23. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/services/visualization_service.py +0 -0
  24. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/ui/__init__.py +0 -0
  25. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/ui/components/__init__.py +0 -0
  26. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/ui/components/backup_restore_dialog.py +0 -0
  27. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/ui/components/filter_builder.py +0 -0
  28. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/ui/components/item_dialog.py +0 -0
  29. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/ui/components/loading_dialog.py +0 -0
  30. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/ui/views/__init__.py +0 -0
  31. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/ui/views/collection_browser.py +0 -0
  32. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/ui/views/connection_view.py +0 -0
  33. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/ui/views/metadata_view.py +0 -0
  34. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/ui/views/search_view.py +0 -0
  35. {vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/ui/views/visualization_view.py +0 -0
{vector_inspector-0.2.1 → vector_inspector-0.2.3}/PKG-INFO RENAMED
@@ -1,9 +1,13 @@
1
1
  Metadata-Version: 2.1
2
2
  Name: vector-inspector
3
- Version: 0.2.1
3
+ Version: 0.2.3
4
4
  Summary: A comprehensive desktop application for visualizing, querying, and managing vector database data
5
5
  Author-Email: Anthony Dawson <anthonypdawson+github@gmail.com>
6
6
  License: MIT
7
+ Project-URL: Homepage, https://vector-inspector.divinedevops.com
8
+ Project-URL: Source, https://github.com/anthony-dawson/vector-inspector
9
+ Project-URL: Issues, https://github.com/anthony-dawson/vector-inspector/issues
10
+ Project-URL: Documentation, https://github.com/anthony-dawson/vector-inspector#readme
7
11
  Requires-Python: ==3.12.*
8
12
  Requires-Dist: chromadb>=0.4.22
9
13
  Requires-Dist: qdrant-client>=1.7.0
@@ -17,21 +21,26 @@ Requires-Dist: plotly>=5.18.0
17
21
  Requires-Dist: sentence-transformers>=2.2.0
18
22
  Requires-Dist: fastembed>=0.7.4
19
23
  Requires-Dist: pyarrow>=14.0.0
24
+ Requires-Dist: pinecone>=8.0.0
20
25
  Description-Content-Type: text/markdown
21
26
 
27
+
22
28
  # Vector Inspector
29
+ ![PyPI](https://img.shields.io/pypi/v/vector-inspector)
30
+ [![PyPI Downloads](https://static.pepy.tech/personalized-badge/vector-inspector?period=total&units=INTERNATIONAL_SYSTEM&left_color=BLACK&right_color=GREEN&left_text=downloads)](https://pepy.tech/projects/vector-inspector)
23
31
 
24
32
 
25
33
  A comprehensive desktop application for visualizing, querying, and managing vector database data. Similar to SQL database viewers, Vector Inspector provides an intuitive GUI for exploring vector embeddings, metadata, and performing similarity searches across multiple vector database providers.
26
34
 
27
35
  ## Overview
28
36
 
37
+ Vector Inspector bridges the gap between vector databases and user-friendly data exploration tools. While vector databases are powerful for semantic search and AI applications, they often lack the intuitive inspection and management tools that traditional SQL databases have. This project aims to provide that missing layer.
38
+
29
39
  ## Table of Contents
30
40
 
31
41
  - [Overview](#overview)
32
42
  - [Key Features](#key-features)
33
43
  - [Architecture](#architecture)
34
- - [Application Structure](#application-structure)
35
44
  - [Use Cases](#use-cases)
36
45
  - [Feature Access](#feature-access)
37
46
  - [Roadmap](#roadmap)
@@ -42,8 +51,6 @@ A comprehensive desktop application for visualizing, querying, and managing vect
42
51
  - [License](#license)
43
52
  - [Acknowledgments](#acknowledgments)
44
53
 
45
- Vector Inspector bridges the gap between vector databases and user-friendly data exploration tools. While vector databases are powerful for semantic search and AI applications, they often lack the intuitive inspection and management tools that traditional SQL databases have. This project aims to provide that missing layer.
46
-
47
54
  ## Key Features
48
55
 
49
56
  ### 1. **Multi-Provider Support**
@@ -134,9 +141,13 @@ For detailed architecture information, see [docs/architecture.md](docs/architect
134
141
 
135
142
  ## Feature Access
136
143
 
137
- Vector Inspector is available in both free (open source) and Pro versions. The free version includes all core features for ChromaDB and basic Qdrant support, while Pro adds advanced analytics and additional providers.
144
+ Vector Inspector follows a user-friendly monetization model:
138
145
 
139
- See [FEATURES.md](FEATURES.md) for a complete feature comparison.
146
+ - **All vector database providers are free** — Try the full app with any database
147
+ - **Core workflows remain free** — Connect, browse, search, visualize, and manage your data
148
+ - **Pro adds power tools** — Advanced analytics, enterprise formats, workflow automation, and collaboration
149
+
150
+ **Nothing currently in Free will ever move to Pro.** See [FEATURES.md](FEATURES.md) for a detailed comparison.
140
151
 
141
152
  ## Roadmap
142
153
 
@@ -144,6 +155,7 @@ See [FEATURES.md](FEATURES.md) for a complete feature comparison.
144
155
 
145
156
  See [ROADMAP.md](ROADMAP.md) for the complete development roadmap and planned features.
146
157
 
158
+
147
159
  ## Installation
148
160
 
149
161
  ### From PyPI (Recommended)
@@ -153,12 +165,28 @@ pip install vector-inspector
153
165
  vector-inspector
154
166
  ```
155
167
 
168
+ ### From a Downloaded Wheel or Tarball (e.g., GitHub Release)
169
+
170
+ Download the `.whl` or `.tar.gz` file from the [GitHub Releases](https://github.com/anthonypdawson/vector-inspector/releases) page, then install with:
171
+
172
+ ```bash
173
+ pip install <your-filename.whl>
174
+ # or
175
+ pip install <your-filename.tar.gz>
176
+ ```
177
+
178
+ After installation, run the application with:
179
+
180
+ ```bash
181
+ vector-inspector
182
+ ```
183
+
156
184
  ### From Source
157
185
 
158
186
  ```bash
159
187
  # Clone the repository
160
- git clone https://github.com/anthonypdawson/vector-viewer.git
161
- cd vector-viewer
188
+ git clone https://github.com/anthonypdawson/vector-inspector.git
189
+ cd vector-inspector
162
190
 
163
191
  # Install dependencies using PDM
164
192
  pdm install
@@ -231,15 +259,8 @@ This project draws inspiration from:
231
259
 
232
260
  ---
233
261
 
234
- **Status**: ✅ Phase 2 Complete - Advanced Features Implemented!
235
262
 
236
- **What's New in Phase 2:**
237
- - 🔍 Advanced metadata filtering with customizable filter rules (AND/OR logic)
238
- - ✏️ Double-click to edit items directly in the data browser
239
- - 📥 Import data from CSV, JSON, and Parquet files
240
- - 📤 Export filtered data to CSV, JSON, and Parquet formats
241
- - 💾 Comprehensive backup and restore system for collections
242
- - 🔄 Metadata filters integrated with search for powerful queries
263
+ See [CHANGELOG.md](CHANGELOG.md) for the latest status and what's new in each release.
243
264
 
244
265
  See [GETTING_STARTED.md](GETTING_STARTED.md) for usage instructions and [IMPLEMENTATION_SUMMARY.md](IMPLEMENTATION_SUMMARY.md) for technical details.
245
266
 
{vector_inspector-0.2.1 → vector_inspector-0.2.3}/README.md RENAMED
@@ -1,225 +1,241 @@
1
- # Vector Inspector
2
-
3
-
4
- A comprehensive desktop application for visualizing, querying, and managing vector database data. Similar to SQL database viewers, Vector Inspector provides an intuitive GUI for exploring vector embeddings, metadata, and performing similarity searches across multiple vector database providers.
5
-
6
- ## Overview
7
-
8
- ## Table of Contents
9
-
10
- - [Overview](#overview)
11
- - [Key Features](#key-features)
12
- - [Architecture](#architecture)
13
- - [Application Structure](#application-structure)
14
- - [Use Cases](#use-cases)
15
- - [Feature Access](#feature-access)
16
- - [Roadmap](#roadmap)
17
- - [Installation](#installation)
18
- - [Configuration](#configuration)
19
- - [Development Setup](#development-setup)
20
- - [Contributing](#contributing)
21
- - [License](#license)
22
- - [Acknowledgments](#acknowledgments)
23
-
24
- Vector Inspector bridges the gap between vector databases and user-friendly data exploration tools. While vector databases are powerful for semantic search and AI applications, they often lack the intuitive inspection and management tools that traditional SQL databases have. This project aims to provide that missing layer.
25
-
26
- ## Key Features
27
-
28
- ### 1. **Multi-Provider Support**
29
- - Connect to vector databases:
30
- - ChromaDB (persistent local storage)
31
- - Qdrant (remote server or embedded local)
32
- - Unified interface regardless of backend provider
33
- - Automatically saves last connection configuration
34
-
35
- ### 2. **Data Visualization**
36
- - **Metadata Explorer**: Browse and filter vector entries by metadata fields
37
- - **Vector Dimensionality Reduction**: Visualize high-dimensional vectors in 2D/3D using:
38
- - t-SNE
39
- - UMAP
40
- - PCA
41
- - **Cluster Visualization**: Color-code vectors by metadata categories or clustering results
42
- - **Interactive Plots**: Zoom, pan, and select vectors for detailed inspection
43
- - **Data Distribution Charts**: Histograms and statistics for metadata fields
44
-
45
- ### 3. **Search & Query Interface**
46
- - **Similarity Search**:
47
- - Text-to-vector search (with embedding model integration)
48
- - Vector-to-vector search
49
- - Find similar items to selected entries
50
- - Adjustable top-k results and similarity thresholds
51
- - **Metadata Filtering**:
52
- - SQL-like query builder for metadata
53
- - Combine vector similarity with metadata filters
54
- - Advanced filtering: ranges, IN clauses, pattern matching
55
- - **Hybrid Search**: Combine semantic search with keyword search
56
- - **Query History**: Save and reuse frequent queries
57
-
58
- ### 4. **Data Management**
59
- - **Browse Collections/Indexes**: View all available collections with statistics
60
- - **CRUD Operations**:
61
- - View individual vectors and their metadata
62
- - Add new vectors (with auto-embedding options)
63
- - Update metadata fields
64
- - Delete vectors (single or batch)
65
- - **Bulk Import/Export**:
66
- - Import from CSV, JSON, Parquet
67
- - Export query results to various formats
68
- - Backup and restore collections
69
- - **Schema Inspector**: View collection configuration, vector dimensions, metadata schema
70
-
71
- ### 5. **SQL-Like Experience**
72
- - **Query Console**: Write queries in a familiar SQL-like syntax (where supported)
73
- - **Results Grid**:
74
- - Sortable, filterable table view
75
- - Pagination for large result sets
76
- - Column customization
77
- - **Data Inspector**: Click any row to see full details including raw vector
78
- - **Query Execution Plans**: Understand how queries are executed
79
- - **Auto-completion**: Intelligent suggestions for collection names, fields, and operations
80
-
81
- ### 6. **Advanced Features**
82
- - **Embedding Model Integration**:
83
- - Use OpenAI, Cohere, HuggingFace models for text-to-vector conversion
84
- - Local model support (sentence-transformers)
85
- - Custom model integration
86
- - **Vector Analysis**:
87
- - Compute similarity matrices
88
- - Identify outliers and anomalies
89
- - Cluster analysis with k-means, DBSCAN
90
- - **Embedding Inspector**:
91
- - For similar collections or items, automatically identify which vector dimensions (activations) most contribute to the similarity
92
- - Map key activations to interpretable concepts (e.g., 'humor', 'sadness', 'anger') using metadata or labels
93
- - Generate human-readable explanations for why items are similar
94
- - **Performance Monitoring**:
95
- - Query latency tracking
96
- - Index performance metrics
97
- - Connection health monitoring
98
-
99
- ## Architecture
100
-
101
- Vector Inspector is built with PySide6 (Qt for Python) for the GUI, providing a native desktop experience. The backend uses Python with support for multiple vector database providers through a unified interface.
102
-
103
- For detailed architecture information, see [docs/architecture.md](docs/architecture.md).
104
-
105
- ## Use Cases
106
-
107
- 1. **AI/ML Development**: Inspect embeddings generated during model development
108
- 2. **RAG System Debugging**: Verify what documents are being retrieved
109
- 3. **Data Quality Assurance**: Identify poorly embedded or outlier vectors
110
- 4. **Production Monitoring**: Check vector database health and data consistency
111
- 5. **Data Migration**: Transfer data between vector database providers
112
- 6. **Education**: Learn and experiment with vector databases interactively
113
-
114
- ## Feature Access
115
-
116
- Vector Inspector is available in both free (open source) and Pro versions. The free version includes all core features for ChromaDB and basic Qdrant support, while Pro adds advanced analytics and additional providers.
117
-
118
- See [FEATURES.md](FEATURES.md) for a complete feature comparison.
119
-
120
- ## Roadmap
121
-
122
- **Current Status**: ✅ Phase 2 Complete
123
-
124
- See [ROADMAP.md](ROADMAP.md) for the complete development roadmap and planned features.
125
-
126
- ## Installation
127
-
128
- ### From PyPI (Recommended)
129
-
130
- ```bash
131
- pip install vector-inspector
132
- vector-inspector
133
- ```
134
-
135
- ### From Source
136
-
137
- ```bash
138
- # Clone the repository
139
- git clone https://github.com/anthonypdawson/vector-viewer.git
140
- cd vector-viewer
141
-
142
- # Install dependencies using PDM
143
- pdm install
144
-
145
- # Launch application
146
- ./run.sh # Linux/macOS
147
- ./run.bat # Windows
148
- ```
149
-
150
- ## Configuration
151
-
152
- Paths are resolved relative to the project root (where `pyproject.toml` is). For example, entering `./data/chroma_db` will use the absolute path resolved from the project root.
153
-
154
- The application automatically saves your last connection configuration to `~/.vector-viewer/settings.json`. The next time you launch the application, it will attempt to reconnect using the last saved settings.
155
-
156
- Example settings structure:
157
- ```json
158
- {
159
- "last_connection": {
160
- "provider": "chromadb",
161
- "connection_type": "persistent",
162
- "path": "./data/chroma_db"
163
- }
164
- }
165
- ```
166
-
167
- ## Development Setup
168
-
169
- ```bash
170
- # Install PDM if you haven't already
171
- pip install pdm
172
-
173
- # Install dependencies with development tools (PDM will create venv automatically)
174
- pdm install -d
175
-
176
- # Run tests
177
- pdm run pytest
178
-
179
- # Run application in development mode
180
- ./run.sh # Linux/macOS
181
- ./run.bat # Windows
182
-
183
- # Or use Python module directly from src directory:
184
- cd src
185
- pdm run python -m vector_viewer
186
- ```
187
-
188
- ## Contributing
189
-
190
- Contributions are welcome! Areas where help is needed:
191
- - Additional vector database provider integrations
192
- - UI/UX improvements
193
- - Performance optimizations
194
- - Documentation
195
- - Test coverage
196
-
197
- Please see [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines.
198
-
199
- ## License
200
-
201
- MIT License - See [LICENSE](LICENSE) file for details.
202
-
203
- ## Acknowledgments
204
-
205
- This project draws inspiration from:
206
- - DBeaver (SQL database viewer)
207
- - MongoDB Compass (NoSQL database GUI)
208
- - Pinecone Console
209
- - Various vector database management tools
210
-
211
- ---
212
-
213
- **Status**: ✅ Phase 2 Complete - Advanced Features Implemented!
214
-
215
- **What's New in Phase 2:**
216
- - 🔍 Advanced metadata filtering with customizable filter rules (AND/OR logic)
217
- - ✏️ Double-click to edit items directly in the data browser
218
- - 📥 Import data from CSV, JSON, and Parquet files
219
- - 📤 Export filtered data to CSV, JSON, and Parquet formats
220
- - 💾 Comprehensive backup and restore system for collections
221
- - 🔄 Metadata filters integrated with search for powerful queries
222
-
223
- See [GETTING_STARTED.md](GETTING_STARTED.md) for usage instructions and [IMPLEMENTATION_SUMMARY.md](IMPLEMENTATION_SUMMARY.md) for technical details.
224
-
225
- **Contact**: Anthony Dawson
1
+
2
+ # Vector Inspector
3
+ ![PyPI](https://img.shields.io/pypi/v/vector-inspector)
4
+ [![PyPI Downloads](https://static.pepy.tech/personalized-badge/vector-inspector?period=total&units=INTERNATIONAL_SYSTEM&left_color=BLACK&right_color=GREEN&left_text=downloads)](https://pepy.tech/projects/vector-inspector)
5
+
6
+
7
+ A comprehensive desktop application for visualizing, querying, and managing vector database data. Similar to SQL database viewers, Vector Inspector provides an intuitive GUI for exploring vector embeddings, metadata, and performing similarity searches across multiple vector database providers.
8
+
9
+ ## Overview
10
+
11
+ Vector Inspector bridges the gap between vector databases and user-friendly data exploration tools. While vector databases are powerful for semantic search and AI applications, they often lack the intuitive inspection and management tools that traditional SQL databases have. This project aims to provide that missing layer.
12
+
13
+ ## Table of Contents
14
+
15
+ - [Overview](#overview)
16
+ - [Key Features](#key-features)
17
+ - [Architecture](#architecture)
18
+ - [Use Cases](#use-cases)
19
+ - [Feature Access](#feature-access)
20
+ - [Roadmap](#roadmap)
21
+ - [Installation](#installation)
22
+ - [Configuration](#configuration)
23
+ - [Development Setup](#development-setup)
24
+ - [Contributing](#contributing)
25
+ - [License](#license)
26
+ - [Acknowledgments](#acknowledgments)
27
+
28
+ ## Key Features
29
+
30
+ ### 1. **Multi-Provider Support**
31
+ - Connect to vector databases:
32
+ - ChromaDB (persistent local storage)
33
+ - Qdrant (remote server or embedded local)
34
+ - Unified interface regardless of backend provider
35
+ - Automatically saves last connection configuration
36
+
37
+ ### 2. **Data Visualization**
38
+ - **Metadata Explorer**: Browse and filter vector entries by metadata fields
39
+ - **Vector Dimensionality Reduction**: Visualize high-dimensional vectors in 2D/3D using:
40
+ - t-SNE
41
+ - UMAP
42
+ - PCA
43
+ - **Cluster Visualization**: Color-code vectors by metadata categories or clustering results
44
+ - **Interactive Plots**: Zoom, pan, and select vectors for detailed inspection
45
+ - **Data Distribution Charts**: Histograms and statistics for metadata fields
46
+
47
+ ### 3. **Search & Query Interface**
48
+ - **Similarity Search**:
49
+ - Text-to-vector search (with embedding model integration)
50
+ - Vector-to-vector search
51
+ - Find similar items to selected entries
52
+ - Adjustable top-k results and similarity thresholds
53
+ - **Metadata Filtering**:
54
+ - SQL-like query builder for metadata
55
+ - Combine vector similarity with metadata filters
56
+ - Advanced filtering: ranges, IN clauses, pattern matching
57
+ - **Hybrid Search**: Combine semantic search with keyword search
58
+ - **Query History**: Save and reuse frequent queries
59
+
60
+ ### 4. **Data Management**
61
+ - **Browse Collections/Indexes**: View all available collections with statistics
62
+ - **CRUD Operations**:
63
+ - View individual vectors and their metadata
64
+ - Add new vectors (with auto-embedding options)
65
+ - Update metadata fields
66
+ - Delete vectors (single or batch)
67
+ - **Bulk Import/Export**:
68
+ - Import from CSV, JSON, Parquet
69
+ - Export query results to various formats
70
+ - Backup and restore collections
71
+ - **Schema Inspector**: View collection configuration, vector dimensions, metadata schema
72
+
73
+ ### 5. **SQL-Like Experience**
74
+ - **Query Console**: Write queries in a familiar SQL-like syntax (where supported)
75
+ - **Results Grid**:
76
+ - Sortable, filterable table view
77
+ - Pagination for large result sets
78
+ - Column customization
79
+ - **Data Inspector**: Click any row to see full details including raw vector
80
+ - **Query Execution Plans**: Understand how queries are executed
81
+ - **Auto-completion**: Intelligent suggestions for collection names, fields, and operations
82
+
83
+ ### 6. **Advanced Features**
84
+ - **Embedding Model Integration**:
85
+ - Use OpenAI, Cohere, HuggingFace models for text-to-vector conversion
86
+ - Local model support (sentence-transformers)
87
+ - Custom model integration
88
+ - **Vector Analysis**:
89
+ - Compute similarity matrices
90
+ - Identify outliers and anomalies
91
+ - Cluster analysis with k-means, DBSCAN
92
+ - **Embedding Inspector**:
93
+ - For similar collections or items, automatically identify which vector dimensions (activations) most contribute to the similarity
94
+ - Map key activations to interpretable concepts (e.g., 'humor', 'sadness', 'anger') using metadata or labels
95
+ - Generate human-readable explanations for why items are similar
96
+ - **Performance Monitoring**:
97
+ - Query latency tracking
98
+ - Index performance metrics
99
+ - Connection health monitoring
100
+
101
+ ## Architecture
102
+
103
+ Vector Inspector is built with PySide6 (Qt for Python) for the GUI, providing a native desktop experience. The backend uses Python with support for multiple vector database providers through a unified interface.
104
+
105
+ For detailed architecture information, see [docs/architecture.md](docs/architecture.md).
106
+
107
+ ## Use Cases
108
+
109
+ 1. **AI/ML Development**: Inspect embeddings generated during model development
110
+ 2. **RAG System Debugging**: Verify what documents are being retrieved
111
+ 3. **Data Quality Assurance**: Identify poorly embedded or outlier vectors
112
+ 4. **Production Monitoring**: Check vector database health and data consistency
113
+ 5. **Data Migration**: Transfer data between vector database providers
114
+ 6. **Education**: Learn and experiment with vector databases interactively
115
+
116
+ ## Feature Access
117
+
118
+ Vector Inspector follows a user-friendly monetization model:
119
+
120
+ - **All vector database providers are free** — Try the full app with any database
121
+ - **Core workflows remain free** — Connect, browse, search, visualize, and manage your data
122
+ - **Pro adds power tools** — Advanced analytics, enterprise formats, workflow automation, and collaboration
123
+
124
+ **Nothing currently in Free will ever move to Pro.** See [FEATURES.md](FEATURES.md) for a detailed comparison.
125
+
126
+ ## Roadmap
127
+
128
+ **Current Status**: ✅ Phase 2 Complete
129
+
130
+ See [ROADMAP.md](ROADMAP.md) for the complete development roadmap and planned features.
131
+
132
+
133
+ ## Installation
134
+
135
+ ### From PyPI (Recommended)
136
+
137
+ ```bash
138
+ pip install vector-inspector
139
+ vector-inspector
140
+ ```
141
+
142
+ ### From a Downloaded Wheel or Tarball (e.g., GitHub Release)
143
+
144
+ Download the `.whl` or `.tar.gz` file from the [GitHub Releases](https://github.com/anthonypdawson/vector-inspector/releases) page, then install with:
145
+
146
+ ```bash
147
+ pip install <your-filename.whl>
148
+ # or
149
+ pip install <your-filename.tar.gz>
150
+ ```
151
+
152
+ After installation, run the application with:
153
+
154
+ ```bash
155
+ vector-inspector
156
+ ```
157
+
158
+ ### From Source
159
+
160
+ ```bash
161
+ # Clone the repository
162
+ git clone https://github.com/anthonypdawson/vector-inspector.git
163
+ cd vector-inspector
164
+
165
+ # Install dependencies using PDM
166
+ pdm install
167
+
168
+ # Launch application
169
+ ./run.sh # Linux/macOS
170
+ ./run.bat # Windows
171
+ ```
172
+
173
+ ## Configuration
174
+
175
+ Paths are resolved relative to the project root (where `pyproject.toml` is). For example, entering `./data/chroma_db` will use the absolute path resolved from the project root.
176
+
177
+ The application automatically saves your last connection configuration to `~/.vector-viewer/settings.json`. The next time you launch the application, it will attempt to reconnect using the last saved settings.
178
+
179
+ Example settings structure:
180
+ ```json
181
+ {
182
+ "last_connection": {
183
+ "provider": "chromadb",
184
+ "connection_type": "persistent",
185
+ "path": "./data/chroma_db"
186
+ }
187
+ }
188
+ ```
189
+
190
+ ## Development Setup
191
+
192
+ ```bash
193
+ # Install PDM if you haven't already
194
+ pip install pdm
195
+
196
+ # Install dependencies with development tools (PDM will create venv automatically)
197
+ pdm install -d
198
+
199
+ # Run tests
200
+ pdm run pytest
201
+
202
+ # Run application in development mode
203
+ ./run.sh # Linux/macOS
204
+ ./run.bat # Windows
205
+
206
+ # Or use Python module directly from src directory:
207
+ cd src
208
+ pdm run python -m vector_viewer
209
+ ```
210
+
211
+ ## Contributing
212
+
213
+ Contributions are welcome! Areas where help is needed:
214
+ - Additional vector database provider integrations
215
+ - UI/UX improvements
216
+ - Performance optimizations
217
+ - Documentation
218
+ - Test coverage
219
+
220
+ Please see [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines.
221
+
222
+ ## License
223
+
224
+ MIT License - See [LICENSE](LICENSE) file for details.
225
+
226
+ ## Acknowledgments
227
+
228
+ This project draws inspiration from:
229
+ - DBeaver (SQL database viewer)
230
+ - MongoDB Compass (NoSQL database GUI)
231
+ - Pinecone Console
232
+ - Various vector database management tools
233
+
234
+ ---
235
+
236
+
237
+ See [CHANGELOG.md](CHANGELOG.md) for the latest status and what's new in each release.
238
+
239
+ See [GETTING_STARTED.md](GETTING_STARTED.md) for usage instructions and [IMPLEMENTATION_SUMMARY.md](IMPLEMENTATION_SUMMARY.md) for technical details.
240
+
241
+ **Contact**: Anthony Dawson
{vector_inspector-0.2.1 → vector_inspector-0.2.3}/pyproject.toml RENAMED
@@ -1,6 +1,6 @@
1
1
  [project]
2
2
  name = "vector-inspector"
3
- version = "0.2.1"
3
+ version = "0.2.3"
4
4
  description = "A comprehensive desktop application for visualizing, querying, and managing vector database data"
5
5
  authors = [
6
6
  { name = "Anthony Dawson", email = "anthonypdawson+github@gmail.com" },
@@ -18,6 +18,7 @@ dependencies = [
18
18
  "sentence-transformers>=2.2.0",
19
19
  "fastembed>=0.7.4",
20
20
  "pyarrow>=14.0.0",
21
+ "pinecone>=8.0.0",
21
22
  ]
22
23
  requires-python = "==3.12.*"
23
24
  readme = "README.md"
@@ -25,6 +26,12 @@ readme = "README.md"
25
26
  [project.license]
26
27
  text = "MIT"
27
28
 
29
+ [project.urls]
30
+ Homepage = "https://vector-inspector.divinedevops.com"
31
+ Source = "https://github.com/anthony-dawson/vector-inspector"
32
+ Issues = "https://github.com/anthony-dawson/vector-inspector/issues"
33
+ Documentation = "https://github.com/anthony-dawson/vector-inspector#readme"
34
+
28
35
  [project.scripts]
29
36
  vector-inspector = "vector_inspector.main:main"
30
37
 
{vector_inspector-0.2.1 → vector_inspector-0.2.3}/src/vector_inspector/ui/main_window.py RENAMED
@@ -282,7 +282,7 @@ class MainWindow(QMainWindow):
282
282
  "<h2>Vector Inspector 0.1.0</h2>"
283
283
  "<p>A comprehensive desktop application for visualizing, "
284
284
  "querying, and managing vector database data.</p>"
285
- '<p><a href="https://github.com/anthonypdawson/vector-viewer" style="color:#2980b9;">GitHub Project Page</a></p>'
285
+ '<p><a href="https://github.com/anthonypdawson/vector-inspector" style="color:#2980b9;">GitHub Project Page</a></p>'
286
286
  "<hr />"
287
287
  "<p>Built with PySide6 and ChromaDB</p>"
288
288
  )