├── .gitignore
├── LICENSE
├── README.md
├── app
├── __init__.py
├── add.py
├── db.py
├── db_wrappers.py
├── exporting.py
├── home.py
├── projects.py
├── schema.sql
├── static
│ ├── add.css
│ ├── manifest.css
│ ├── projects.css
│ ├── style.css
│ ├── style_viewer.css
│ ├── table.css
│ └── view.css
├── style.py
├── tasks.py
├── templates
│ ├── add.html
│ ├── base.html
│ ├── export.html
│ ├── exports.html
│ ├── home.html
│ ├── manifest.html
│ ├── project.html
│ ├── projects.html
│ ├── style.html
│ ├── tasks.html
│ └── view.html
├── utils.py
└── view.py
├── requirements.txt
├── sample-datasets
└── alpaca.json
└── screenshots
├── praetor-completion-screen.png
├── praetor-prompts-screen.png
└── praetor-style-screen.png
/.gitignore:
--------------------------------------------------------------------------------
1 | venv
2 | instance
3 | app/__pycache__
4 |
--------------------------------------------------------------------------------
/LICENSE:
--------------------------------------------------------------------------------
1 |
2 | Apache License
3 | Version 2.0, January 2004
4 | http://www.apache.org/licenses/
5 |
6 | TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
7 |
8 | 1. Definitions.
9 |
10 | "License" shall mean the terms and conditions for use, reproduction,
11 | and distribution as defined by Sections 1 through 9 of this document.
12 |
13 | "Licensor" shall mean the copyright owner or entity authorized by
14 | the copyright owner that is granting the License.
15 |
16 | "Legal Entity" shall mean the union of the acting entity and all
17 | other entities that control, are controlled by, or are under common
18 | control with that entity. For the purposes of this definition,
19 | "control" means (i) the power, direct or indirect, to cause the
20 | direction or management of such entity, whether by contract or
21 | otherwise, or (ii) ownership of fifty percent (50%) or more of the
22 | outstanding shares, or (iii) beneficial ownership of such entity.
23 |
24 | "You" (or "Your") shall mean an individual or Legal Entity
25 | exercising permissions granted by this License.
26 |
27 | "Source" form shall mean the preferred form for making modifications,
28 | including but not limited to software source code, documentation
29 | source, and configuration files.
30 |
31 | "Object" form shall mean any form resulting from mechanical
32 | transformation or translation of a Source form, including but
33 | not limited to compiled object code, generated documentation,
34 | and conversions to other media types.
35 |
36 | "Work" shall mean the work of authorship, whether in Source or
37 | Object form, made available under the License, as indicated by a
38 | copyright notice that is included in or attached to the work
39 | (an example is provided in the Appendix below).
40 |
41 | "Derivative Works" shall mean any work, whether in Source or Object
42 | form, that is based on (or derived from) the Work and for which the
43 | editorial revisions, annotations, elaborations, or other modifications
44 | represent, as a whole, an original work of authorship. For the purposes
45 | of this License, Derivative Works shall not include works that remain
46 | separable from, or merely link (or bind by name) to the interfaces of,
47 | the Work and Derivative Works thereof.
48 |
49 | "Contribution" shall mean any work of authorship, including
50 | the original version of the Work and any modifications or additions
51 | to that Work or Derivative Works thereof, that is intentionally
52 | submitted to Licensor for inclusion in the Work by the copyright owner
53 | or by an individual or Legal Entity authorized to submit on behalf of
54 | the copyright owner. For the purposes of this definition, "submitted"
55 | means any form of electronic, verbal, or written communication sent
56 | to the Licensor or its representatives, including but not limited to
57 | communication on electronic mailing lists, source code control systems,
58 | and issue tracking systems that are managed by, or on behalf of, the
59 | Licensor for the purpose of discussing and improving the Work, but
60 | excluding communication that is conspicuously marked or otherwise
61 | designated in writing by the copyright owner as "Not a Contribution."
62 |
63 | "Contributor" shall mean Licensor and any individual or Legal Entity
64 | on behalf of whom a Contribution has been received by Licensor and
65 | subsequently incorporated within the Work.
66 |
67 | 2. Grant of Copyright License. Subject to the terms and conditions of
68 | this License, each Contributor hereby grants to You a perpetual,
69 | worldwide, non-exclusive, no-charge, royalty-free, irrevocable
70 | copyright license to reproduce, prepare Derivative Works of,
71 | publicly display, publicly perform, sublicense, and distribute the
72 | Work and such Derivative Works in Source or Object form.
73 |
74 | 3. Grant of Patent License. Subject to the terms and conditions of
75 | this License, each Contributor hereby grants to You a perpetual,
76 | worldwide, non-exclusive, no-charge, royalty-free, irrevocable
77 | (except as stated in this section) patent license to make, have made,
78 | use, offer to sell, sell, import, and otherwise transfer the Work,
79 | where such license applies only to those patent claims licensable
80 | by such Contributor that are necessarily infringed by their
81 | Contribution(s) alone or by combination of their Contribution(s)
82 | with the Work to which such Contribution(s) was submitted. If You
83 | institute patent litigation against any entity (including a
84 | cross-claim or counterclaim in a lawsuit) alleging that the Work
85 | or a Contribution incorporated within the Work constitutes direct
86 | or contributory patent infringement, then any patent licenses
87 | granted to You under this License for that Work shall terminate
88 | as of the date such litigation is filed.
89 |
90 | 4. Redistribution. You may reproduce and distribute copies of the
91 | Work or Derivative Works thereof in any medium, with or without
92 | modifications, and in Source or Object form, provided that You
93 | meet the following conditions:
94 |
95 | (a) You must give any other recipients of the Work or
96 | Derivative Works a copy of this License; and
97 |
98 | (b) You must cause any modified files to carry prominent notices
99 | stating that You changed the files; and
100 |
101 | (c) You must retain, in the Source form of any Derivative Works
102 | that You distribute, all copyright, patent, trademark, and
103 | attribution notices from the Source form of the Work,
104 | excluding those notices that do not pertain to any part of
105 | the Derivative Works; and
106 |
107 | (d) If the Work includes a "NOTICE" text file as part of its
108 | distribution, then any Derivative Works that You distribute must
109 | include a readable copy of the attribution notices contained
110 | within such NOTICE file, excluding those notices that do not
111 | pertain to any part of the Derivative Works, in at least one
112 | of the following places: within a NOTICE text file distributed
113 | as part of the Derivative Works; within the Source form or
114 | documentation, if provided along with the Derivative Works; or,
115 | within a display generated by the Derivative Works, if and
116 | wherever such third-party notices normally appear. The contents
117 | of the NOTICE file are for informational purposes only and
118 | do not modify the License. You may add Your own attribution
119 | notices within Derivative Works that You distribute, alongside
120 | or as an addendum to the NOTICE text from the Work, provided
121 | that such additional attribution notices cannot be construed
122 | as modifying the License.
123 |
124 | You may add Your own copyright statement to Your modifications and
125 | may provide additional or different license terms and conditions
126 | for use, reproduction, or distribution of Your modifications, or
127 | for any such Derivative Works as a whole, provided Your use,
128 | reproduction, and distribution of the Work otherwise complies with
129 | the conditions stated in this License.
130 |
131 | 5. Submission of Contributions. Unless You explicitly state otherwise,
132 | any Contribution intentionally submitted for inclusion in the Work
133 | by You to the Licensor shall be under the terms and conditions of
134 | this License, without any additional terms or conditions.
135 | Notwithstanding the above, nothing herein shall supersede or modify
136 | the terms of any separate license agreement you may have executed
137 | with Licensor regarding such Contributions.
138 |
139 | 6. Trademarks. This License does not grant permission to use the trade
140 | names, trademarks, service marks, or product names of the Licensor,
141 | except as required for reasonable and customary use in describing the
142 | origin of the Work and reproducing the content of the NOTICE file.
143 |
144 | 7. Disclaimer of Warranty. Unless required by applicable law or
145 | agreed to in writing, Licensor provides the Work (and each
146 | Contributor provides its Contributions) on an "AS IS" BASIS,
147 | WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
148 | implied, including, without limitation, any warranties or conditions
149 | of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
150 | PARTICULAR PURPOSE. You are solely responsible for determining the
151 | appropriateness of using or redistributing the Work and assume any
152 | risks associated with Your exercise of permissions under this License.
153 |
154 | 8. Limitation of Liability. In no event and under no legal theory,
155 | whether in tort (including negligence), contract, or otherwise,
156 | unless required by applicable law (such as deliberate and grossly
157 | negligent acts) or agreed to in writing, shall any Contributor be
158 | liable to You for damages, including any direct, indirect, special,
159 | incidental, or consequential damages of any character arising as a
160 | result of this License or out of the use or inability to use the
161 | Work (including but not limited to damages for loss of goodwill,
162 | work stoppage, computer failure or malfunction, or any and all
163 | other commercial damages or losses), even if such Contributor
164 | has been advised of the possibility of such damages.
165 |
166 | 9. Accepting Warranty or Additional Liability. While redistributing
167 | the Work or Derivative Works thereof, You may choose to offer,
168 | and charge a fee for, acceptance of support, warranty, indemnity,
169 | or other liability obligations and/or rights consistent with this
170 | License. However, in accepting such obligations, You may act only
171 | on Your own behalf and on Your sole responsibility, not on behalf
172 | of any other Contributor, and only if You agree to indemnify,
173 | defend, and hold each Contributor harmless for any liability
174 | incurred by, or claims asserted against, such Contributor by reason
175 | of your accepting any such warranty or additional liability.
176 |
177 | END OF TERMS AND CONDITIONS
178 |
179 | APPENDIX: How to apply the Apache License to your work.
180 |
181 | To apply the Apache License to your work, attach the following
182 | boilerplate notice, with the fields enclosed by brackets "[]"
183 | replaced with your own identifying information. (Don't include
184 | the brackets!) The text should be enclosed in the appropriate
185 | comment syntax for the file format. We also recommend that a
186 | file or class name and description of purpose be included on the
187 | same "printed page" as the copyright notice for easier
188 | identification within third-party archives.
189 |
190 | Copyright [yyyy] [name of copyright owner]
191 |
192 | Licensed under the Apache License, Version 2.0 (the "License");
193 | you may not use this file except in compliance with the License.
194 | You may obtain a copy of the License at
195 |
196 | http://www.apache.org/licenses/LICENSE-2.0
197 |
198 | Unless required by applicable law or agreed to in writing, software
199 | distributed under the License is distributed on an "AS IS" BASIS,
200 | WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
201 | See the License for the specific language governing permissions and
202 | limitations under the License.
203 |
--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
1 | # Praetor
2 |
3 | Praetor is a dataset exploration tool. It is made for LLM finetuning data and prompts.
4 |
5 | ## Screenshots
6 |
7 | 
8 |
9 | ---
10 |
11 | 
12 |
13 | ## Why
14 |
15 | I got sick of scanning through large json files to try and get a feeling for a dataset. Then when I wanted to remove some data, I was using control F - this is obviously dumb. Praetor is a flask app that lets you mix, edit, view, and export datasets for LLMs. To keep it simple, the frontend is very straightforward. It uses an sqlite database and jinja templating under the hood. You can import and export datasets in a json format, so it is very easy to use Praetor along with other tools and scripts you already have in place.
16 |
17 | ## About
18 |
19 | Praetor is a super lightweight finetuning data and prompt management tool. The setup requirements are minimal, and so is the complexity.
20 |
21 | In general, the system works like this: you start with a Project. A project has associated with it some prompt styles, which define format strings for the prompts. You can then add prompts in a particular style and then one or more completions to those prompts. When you're done editing prompts or completions, you can export that data in a json format. You can also import data in the same format.
22 |
23 | Thus the hierarchy goes:
24 |
25 | ```
26 | Projects --> Styles --> Prompts --> Completions
27 | ```
28 |
29 | ## Usage
30 |
31 | First, it is recommended that you use a virtual environment if you do not already have one set up for your project. You'll need to install the requirements in any case (it's pretty much just Flask!):
32 |
33 | ```
34 | python3 -m venv venv
35 | source venv/bin/activate
36 | pip install -r requirements.txt
37 | ```
38 |
39 | Then you will need to initialize the database using
40 |
41 | ```
42 | flask init-db
43 | ```
44 |
45 | And then run the server:
46 |
47 | ```
48 | flask run
49 | ```
50 |
51 | ## License
52 |
53 | This software is distributed under the Apache 2.0 open source license.
54 |
55 | ## Contributing
56 |
57 | Everyone should feel free to open an issue if you have any comments, questions, or suggestions.
58 |
59 | Pull requests should be submitted to the development branch. Please do not make additions to the requirements unless they are necessary. Frontend frameworks are not necessary. The frontend should be as simple and functional as possible.
60 |
--------------------------------------------------------------------------------
/app/__init__.py:
--------------------------------------------------------------------------------
1 | import os
2 |
3 | from flask import Flask
4 |
5 |
6 | def create_app(test_config=None):
7 | # create and configure the app
8 | app = Flask(__name__, instance_relative_config=True)
9 | app.config.from_mapping(
10 | SECRET_KEY='dev',
11 | DATABASE=os.path.join(app.instance_path, 'app.sqlite'),
12 | EXPORTS_PATH=os.path.join(app.instance_path, 'exports')
13 | )
14 |
15 | if test_config is None:
16 | # load the instance config, if it exists, when not testing
17 | app.config.from_pyfile('config.py', silent=True)
18 | else:
19 | # load the test config if passed in
20 | app.config.from_mapping(test_config)
21 |
22 | # ensure the instance folder exists
23 | try:
24 | os.makedirs(app.instance_path)
25 | except OSError:
26 | pass
27 | try:
28 | os.makedirs(os.path.join(app.instance_path, 'exports'))
29 | except OSError:
30 | pass
31 |
32 | from . import view
33 | app.register_blueprint(view.bp)
34 |
35 | from . import home
36 | app.register_blueprint(home.bp)
37 |
38 | from . import add
39 | app.register_blueprint(add.bp)
40 |
41 | from . import tasks
42 | app.register_blueprint(tasks.bp)
43 |
44 | from . import exporting
45 | app.register_blueprint(exporting.bp)
46 |
47 | from . import projects
48 | app.register_blueprint(projects.bp)
49 |
50 | from . import style
51 | app.register_blueprint(style.bp)
52 |
53 | # connect database
54 | from . import db
55 | db.init_app(app)
56 |
57 | return app
58 |
--------------------------------------------------------------------------------
/app/add.py:
--------------------------------------------------------------------------------
1 | from flask import (
2 | Blueprint,
3 | render_template,
4 | request,
5 | redirect
6 | )
7 | from app.db import get_db
8 | from app.db_wrappers import add_prompt, add_bulk, get_project_by_id, get_style_by_id, get_styles_by_project_id, get_keys_by_style_id, check_running
9 | from app.utils import tag_string_to_list
10 | import json
11 |
12 | bp = Blueprint('add', __name__)
13 |
14 | @bp.route('/add', methods=('GET', 'POST'))
15 | def add():
16 |
17 | db = get_db()
18 |
19 | if request.method == "POST" and not check_running(db):
20 |
21 | project_id = request.form.get('project_id')
22 | style_id = request.form.get('style_id')
23 |
24 | # Was it a bulk upload?
25 | file = request.files.get('file')
26 | if file:
27 | file_contents = file.read()
28 | # Read as json
29 | # btw this is probably a security vulnerability
30 | # Probably should check for filesize as well
31 | data = json.loads(file_contents)
32 | tags = request.form.get("tags")
33 | tags = tag_string_to_list(tags)
34 | res = add_bulk(db, data, tags, project_id, style_id)
35 | return redirect("/tasks")
36 | else:
37 | style_keys = get_keys_by_style_id(db, style_id)
38 | style = get_style_by_id(db, style_id)
39 |
40 | # don't need completion key for style_keys
41 | if 'completion_key' in style:
42 | completion_key = style['completion_key']
43 | style_keys = [x for x in style_keys if x['name'] != completion_key]
44 |
45 | keys = {x['name']: request.form.get("key." + x['name']) for x in style_keys}
46 | tags = request.form.get("tags")
47 | if tags:
48 | tags = tag_string_to_list(tags)
49 | else:
50 | tags = []
51 | id = add_prompt(db, keys=keys, tags=tags, project_id=project_id, style_id=style_id)
52 | return redirect(f"/view?prompt_id={id}")
53 | else:
54 | project_id = request.args.get('project_id')
55 | style_id = request.args.get('style_id')
56 |
57 | error = None
58 | if request.method == 'POST' and check_running(db):
59 | project_id = request.form.get('project_id')
60 | style_id = request.form.get('style_id')
61 | error = "Cannot add new data while current task is incomplete."
62 |
63 | style = get_style_by_id(db, style_id)
64 | project = get_project_by_id(db, project_id)
65 | style_keys = get_keys_by_style_id(db, style_id)
66 | # don't need completion key for style_keys
67 | if 'completion_key' in style:
68 | completion_key = style['completion_key']
69 | style_keys = [x for x in style_keys if x['name'] != completion_key]
70 |
71 | return render_template('add.html', style=style, project=project, style_keys=style_keys, error=error)
--------------------------------------------------------------------------------
/app/db.py:
--------------------------------------------------------------------------------
1 | import sqlite3
2 | import json
3 | import click
4 | from flask import current_app, g
5 | import os
6 | import shutil
7 |
8 |
9 | def dict_factory(cursor, row):
10 | """
11 | Converts each row to a dictionary
12 | """
13 | d = {}
14 | for idx, col in enumerate(cursor.description):
15 | d[col[0]] = row[idx]
16 | return d
17 |
18 | def get_db():
19 | if 'db' not in g:
20 |
21 | need_to_init = not os.path.exists(current_app.config['DATABASE'])
22 |
23 | g.db = sqlite3.connect(
24 | current_app.config['DATABASE'],
25 | detect_types=sqlite3.PARSE_DECLTYPES
26 | )
27 | g.db.row_factory = dict_factory
28 |
29 | if need_to_init:
30 | with current_app.open_resource('schema.sql') as f:
31 | g.db.executescript(f.read().decode('utf8'))
32 |
33 | return g.db
34 |
35 |
36 | def get_tmp_db(instance_path, old_db_path):
37 |
38 | # TODO: make unique for each process?
39 | new_db_path = os.path.join(instance_path, "tmp.sqlite")
40 | shutil.copyfile(old_db_path, new_db_path)
41 |
42 | db = sqlite3.connect(
43 | new_db_path,
44 | detect_types=sqlite3.PARSE_DECLTYPES
45 | )
46 | db.row_factory = dict_factory
47 |
48 | return db, new_db_path
49 |
50 |
51 | def close_db(e=None):
52 | db = g.pop('db', None)
53 |
54 | if db is not None:
55 | db.close()
56 |
57 | def init_db():
58 | db = get_db()
59 |
60 | with current_app.open_resource('schema.sql') as f:
61 | db.executescript(f.read().decode('utf8'))
62 |
63 |
64 | @click.command('init-db')
65 | def init_db_command():
66 | """Clear the existing data and create new tables."""
67 | init_db()
68 | click.echo('Initialized the database.')
69 |
70 | def init_app(app):
71 | app.teardown_appcontext(close_db)
72 | app.cli.add_command(init_db_command)
73 |
74 |
75 | class SQLiteJSONEncoder(json.JSONEncoder):
76 | def default(self, obj):
77 | if isinstance(obj, sqlite3.Row):
78 | return dict(obj)
79 | return json.JSONEncoder.default(self, obj)
--------------------------------------------------------------------------------
/app/db_wrappers.py:
--------------------------------------------------------------------------------
1 | import multiprocessing
2 | import json
3 | import os
4 | from unicodedata import name
5 | from app.db import SQLiteJSONEncoder
6 | from flask import current_app, session
7 | from app.utils import get_named_arguments
8 | from app.db import get_tmp_db
9 | import shutil
10 | import psutil
11 |
12 |
13 | """
14 |
15 | All functions that depend on the peculiarities of the database
16 |
17 | """
18 |
19 | # EXAMPLES
20 |
21 | def add_example(db, prompt_id, completion, tags):
22 | c = db.cursor()
23 | c.execute("INSERT INTO examples (completion, prompt_id) VALUES (?, ?)", (completion, prompt_id))
24 | item_id = c.lastrowid
25 |
26 | # Add tags
27 | for tag in tags:
28 | c.execute("INSERT INTO tags (example_id, value) VALUES (?, ?)", (item_id, tag))
29 | db.commit()
30 | return item_id
31 |
32 |
33 | def delete_example(db, example_id):
34 | db.execute("DELETE FROM examples WHERE id = ?", (example_id,))
35 | db.commit()
36 |
37 |
38 | def update_example(db, example_id, completion, tags):
39 | c = db.cursor()
40 |
41 | # Remove all tags
42 | c.execute("DELETE FROM tags WHERE example_id = ?", (example_id,))
43 | # Update text
44 | c.execute("UPDATE examples SET completion = ? WHERE id = ?", (completion, example_id))
45 | for t in tags:
46 | c.execute("INSERT INTO tags (example_id, value) VALUES (?, ?)", (example_id, t))
47 | db.commit()
48 | return example_id
49 |
50 |
51 | # PROMPTS
52 |
53 | def delete_prompt(db, prompt_id):
54 |
55 | # Remove all tags
56 | # Remove all prompt_values
57 | # remove associated examples
58 |
59 | db.execute("DELETE FROM prompts WHERE id = ?", (prompt_id,))
60 | db.execute("DELETE FROM tags WHERE prompt_id = ?", (prompt_id,))
61 | db.execute("DELETE FROM prompt_values WHERE prompt_id = ?", (prompt_id,))
62 | db.execute("DELETE FROM examples WHERE prompt_id = ?", (prompt_id,))
63 | db.commit()
64 |
65 | def update_prompt(db, prompt_id, prompt_values, tags):
66 | c = db.cursor()
67 |
68 | # Update prompt values
69 | for key in prompt_values:
70 | c.execute("SELECT * FROM prompt_values WHERE prompt_id = ? AND key = ?", (prompt_id, key))
71 | res = c.fetchone()
72 | if res:
73 | sql = """
74 | UPDATE prompt_values SET value = ? WHERE prompt_id = ? AND key = ?
75 | """
76 | c.execute(sql, (prompt_values[key], prompt_id, key))
77 | else:
78 | sql = """
79 | INSERT INTO prompt_values (value, prompt_id, key) VALUES (?, ?, ?)
80 | """
81 | c.execute(sql, (prompt_values[key], prompt_id, key))
82 | # Update tags
83 | # Remove all prior tags
84 | sql = """
85 | DELETE FROM tags
86 | WHERE prompt_id = ?
87 | """
88 | c.execute(sql, (prompt_id,))
89 | for tag in tags:
90 | c.execute("INSERT INTO tags (value, prompt_id) VALUES (?, ?)", (tag, prompt_id))
91 |
92 | db.commit()
93 | return prompt_id
94 |
95 | # Adds prompt to database and returns that prompt's id
96 | def add_prompt(db, **kwargs):
97 |
98 | tags = kwargs['tags']
99 | keys = kwargs['keys']
100 | project_id = kwargs['project_id']
101 | style_id = kwargs['style_id']
102 |
103 | c = db.cursor()
104 |
105 | c.execute("INSERT INTO prompts (project_id, style) VALUES (?, ?)", (project_id, style_id))
106 | prompt_id = c.lastrowid
107 |
108 | for t in tags:
109 | c.execute("INSERT INTO tags (value, prompt_id) VALUES (?, ?)", (t, prompt_id))
110 |
111 | for k in keys:
112 | c.execute("INSERT INTO prompt_values (prompt_id, key, value) VALUES (?, ?, ?)", (prompt_id, k, keys[k]))
113 |
114 | db.commit()
115 | return prompt_id
116 |
117 | # Takes data and inserts prompts and examples
118 | # Data is a list of dictionaries (i.e. json data)
119 | # tags is a list
120 | def add_bulk(db, data, tags, project_id, style_id):
121 | c = db.cursor()
122 | sql = """
123 | INSERT INTO tasks (`type`, `status`)
124 | VALUES ("bulk_upload",
125 | "in_progress"
126 | );
127 | """
128 | c.execute(sql)
129 | task_id = c.lastrowid
130 | db.commit()
131 |
132 | p = multiprocessing.Process(target=add_bulk_background, args=(current_app.instance_path, current_app.config['DATABASE'], task_id, data, tags, project_id, style_id))
133 | p.start()
134 |
135 | sql = """
136 | UPDATE tasks
137 | SET pid = ?
138 | WHERE id = ?
139 | """
140 | db.execute(sql, (p.pid, task_id))
141 | db.commit()
142 |
143 | # Make session token to warn user about parallelism
144 | session['warn_parallelism'] = True
145 |
146 | status = {
147 | 'pid': p.pid,
148 | 'status': 'in_progress'
149 | }
150 | return status
151 |
152 | # NOTE: doesn't support example tags yet or multiple examples per prompt
153 | def add_bulk_background(instance_path, old_db_path, task_id, data, tags, project_id, style_id):
154 | db, new_db_path = get_tmp_db(instance_path, old_db_path)
155 |
156 | try:
157 | c = db.cursor()
158 |
159 | # Get the correct keys and note which is the completion key
160 | sql = """
161 | SELECT * FROM style_keys
162 | WHERE style_id = ?
163 | """
164 | res = c.execute(sql, (style_id,))
165 | style_keys = res.fetchall()
166 |
167 | # Get the style
168 | sql = """
169 | SELECT * FROM styles
170 | WHERE id = ?
171 | """
172 | res = c.execute(sql, (style_id))
173 | style_info = res.fetchone()
174 |
175 | completion_key = style_info['completion_key']
176 |
177 | prompt_values_keys = [x['name'] for x in style_keys if x['name'] != completion_key]
178 | db.commit()
179 |
180 | for item in data:
181 |
182 | # Add new prompt
183 | sql = """
184 | INSERT INTO prompts (style, project_id) VALUES (?, ?)
185 | """
186 | c.execute(sql, (style_id, project_id))
187 | prompt_id = c.lastrowid
188 |
189 | # Add examples and prompt values
190 | for key in item:
191 | if key == completion_key:
192 | sql = """
193 | INSERT INTO examples (prompt_id, completion) VALUES (?, ?)
194 | """
195 | c.execute(sql, (prompt_id, item[key]))
196 | elif key in prompt_values_keys: # need to avoid tags, other irrelevant values
197 | sql = """
198 | INSERT INTO prompt_values (prompt_id, key, value) VALUES (?, ?, ?)
199 | """
200 | c.execute(sql, (prompt_id, key, item[key]))
201 |
202 | # Add tags to prompt
203 | for tag in tags:
204 | sql = """
205 | INSERT INTO tags (prompt_id, value) VALUES (?, ?)
206 | """
207 | c.execute(sql, (prompt_id, tag))
208 |
209 | db.commit()
210 |
211 | sql = """
212 | UPDATE tasks
213 | SET status = 'completed'
214 | WHERE id = ?
215 | """
216 | db.execute(sql, (task_id,))
217 | db.commit()
218 | except:
219 | sql = """
220 | UPDATE tasks
221 | SET status = 'failed'
222 | WHERE id = ?;
223 | """
224 | db.execute(sql, (task_id,))
225 | db.commit()
226 |
227 | shutil.copyfile(new_db_path, old_db_path)
228 |
229 | """
230 |
231 | Will export a json file, which is a list of dictionaries with each key matching
232 | a named argument in the template format string.
233 | All named arguments are present in every dictionary.
234 | Does not include tags at the moment.
235 |
236 | If filename is None or "", the filename becomes "export.json"
237 |
238 | """
239 | def export(db, filename, tags=[], content="", example="", project_id=None, style_id=None):
240 |
241 | if not filename:
242 | filename = "export.json"
243 |
244 | c = db.cursor()
245 | sql = """
246 | INSERT INTO tasks (`type`, `status`)
247 | VALUES ("export",
248 | "in_progress"
249 | );
250 | """
251 | c.execute(sql)
252 | db.commit()
253 |
254 | task_id = c.lastrowid
255 |
256 | p = multiprocessing.Process(target=export_background, args=(current_app.instance_path, current_app.config['DATABASE'], current_app.config['EXPORTS_PATH'], task_id, filename, content, tags, example, style_id, project_id))
257 | p.start()
258 |
259 | sql = """
260 | UPDATE tasks SET `pid` = ?
261 | WHERE id = ?
262 | """
263 | c.execute(sql, (p.pid, task_id))
264 |
265 | # Make session token to warn user about parallelism
266 | session['warn_parallelism'] = True
267 |
268 | db.commit()
269 | status = {
270 | 'pid': p.pid,
271 | 'status': 'in_progress'
272 | }
273 | return status
274 |
275 | def export_background(instance_path, old_db_path, exports_path, task_id, filename, content, tags, example, style_id, project_id):
276 |
277 | db, new_db_path = get_tmp_db(instance_path, old_db_path)
278 |
279 | # The try catch is not compehensive
280 | # There should be an option for the user to check on the program itself (via its pid)
281 | try:
282 | if not example:
283 | example = "%"
284 | if not content:
285 | content = "%"
286 |
287 | args = [example]
288 |
289 | tag_query_str = ""
290 | if tags and len(tags) > 0:
291 | tag_query_str = f"JOIN tags ON prompts.id = tags.prompt_id AND ("
292 | for i, tag in enumerate(tags):
293 | tag_query_str += f"tags.value LIKE ?"
294 | args.append(tag)
295 | if i < len(tags) - 1:
296 | tag_query_str += " OR "
297 | tag_query_str += ")"
298 |
299 | args.append(content)
300 |
301 | proj_id_query = ""
302 | if project_id:
303 | proj_id_query = "AND prompts.project_id = ?"
304 | args.append(project_id)
305 |
306 | style_id_query = ""
307 | if style_id:
308 | style_id_query = "AND prompts.style = ?"
309 | args.append(style_id)
310 |
311 | # notice that in string, there is no option to put "LEFT" join on examples
312 | # this is beacuse we really do only want prompts with examples in this case
313 | sql = f"""
314 | SELECT DISTINCT prompts.*, styles.template as template, styles.completion_key as completion_key
315 | FROM prompts
316 | JOIN prompt_values ON prompts.id = prompt_values.prompt_id
317 | JOIN styles ON prompts.style = styles.id AND prompt_values.key = styles.preview_key
318 | JOIN examples ON prompts.id = examples.prompt_id AND examples.completion LIKE ?
319 | {tag_query_str}
320 | WHERE prompt_values.value LIKE ?
321 | {proj_id_query}
322 | {style_id_query}
323 | """
324 | c = db.cursor()
325 | res = c.execute(sql, tuple(args))
326 | prompts = res.fetchall()
327 |
328 | path = os.path.join(exports_path, filename)
329 | fhand = open(path, 'w')
330 | fhand.write('[\n')
331 |
332 | encoder = SQLiteJSONEncoder(indent=4)
333 |
334 | for k, prompt in enumerate(prompts):
335 | sql = """
336 | SELECT * FROM examples
337 | WHERE prompt_id = ?
338 | """
339 | examples = c.execute(sql, (prompt['id'],)).fetchall()
340 |
341 | sql = """
342 | SELECT * FROM prompt_values
343 | WHERE prompt_id = ?
344 | """
345 | prompt_values = c.execute(sql, (prompt['id'],)).fetchall()
346 | prompt_value_kwargs = {x['key']: x['value'] for x in prompt_values}
347 |
348 | template = prompt['template']
349 | named_args = get_named_arguments(template)
350 |
351 | kwargs = {}
352 | for arg in named_args:
353 | if arg in prompt_value_kwargs:
354 | kwargs[arg] = prompt_value_kwargs[arg]
355 | elif arg != prompt['completion_key']:
356 | kwargs[arg] = ""
357 |
358 | for i, ex in enumerate(examples):
359 | kwargs[prompt['completion_key']] = ex['completion']
360 |
361 | json_data = encoder.encode(kwargs)
362 |
363 | fhand.write(json_data)
364 |
365 | if i < len(examples) - 1:
366 | fhand.write(",\n")
367 |
368 | if k < len(prompts) - 1:
369 | fhand.write(",\n")
370 |
371 | fhand.write(']')
372 | fhand.close()
373 |
374 | sql = """
375 | INSERT INTO exports (`filename`)
376 | VALUES (?);
377 | """
378 | db.execute(sql, (filename,))
379 |
380 | sql = """
381 | UPDATE tasks
382 | SET status = 'completed'
383 | WHERE id = ?;
384 | """
385 | db.execute(sql, (task_id,))
386 | db.commit()
387 | except Exception as e:
388 | sql = """
389 | UPDATE tasks
390 | SET status = 'failed'
391 | WHERE pid = ?;
392 | """
393 | db.execute(sql, (os.getpid(),))
394 | db.commit()
395 | print(f"Error occurred: {e}")
396 |
397 | shutil.copyfile(new_db_path, old_db_path)
398 |
399 | def search_prompts(db, limit=None, offset=None, content_arg=None, example_arg=None, tags_arg=None, project_id=None, style_id=None):
400 |
401 | offset = 0 if not offset else offset
402 | limit = 100 if not limit else limit
403 | content_arg = "%" if not content_arg else "%" + content_arg + "%"
404 |
405 | x = "" if example_arg else "LEFT"
406 | example_arg = "%" if not example_arg else "%" + example_arg + "%"
407 |
408 | args = [example_arg]
409 |
410 | tag_query_str = ""
411 | if tags_arg and len(tags_arg) > 0:
412 | tag_query_str = f"JOIN tags ON prompts.id = tags.prompt_id AND ("
413 | for i, tag in enumerate(tags_arg):
414 | tag_query_str += f"tags.value LIKE ?"
415 | args.append(tag)
416 | if i < len(tags_arg) - 1:
417 | tag_query_str += " OR "
418 | tag_query_str += ")"
419 |
420 | args.extend([content_arg])
421 |
422 | proj_id_query = ""
423 | if project_id:
424 | proj_id_query = "AND prompts.project_id = ?"
425 | args.append(project_id)
426 |
427 | style_id_query = ""
428 | if style_id:
429 | style_id_query = "AND prompts.style = ?"
430 | args.append(style_id)
431 |
432 | args.extend([limit, offset])
433 |
434 | sql = f"""
435 | WITH main_search AS
436 | (
437 | SELECT DISTINCT prompts.*, prompt_values.*
438 | FROM prompts
439 | JOIN prompt_values ON prompts.id = prompt_values.prompt_id
440 | JOIN styles ON prompts.style = styles.id AND prompt_values.key = styles.preview_key
441 | {x} JOIN examples ON prompts.id = examples.prompt_id AND examples.completion LIKE ?
442 | {tag_query_str}
443 | WHERE prompt_values.value LIKE ?
444 | {proj_id_query}
445 | {style_id_query}
446 | )
447 | SELECT main_search.*, GROUP_CONCAT(t.value) AS tags, COUNT(*) OVER() AS total_results
448 | FROM main_search
449 | LEFT JOIN tags t ON main_search.prompt_id = t.prompt_id
450 | GROUP BY main_search.id
451 | LIMIT ?
452 | OFFSET ?
453 | """
454 |
455 | results = db.execute(sql, tuple(args))
456 | fetched = results.fetchall()
457 | total_results = 0 if len(fetched) == 0 else fetched[0]['total_results']
458 | return fetched, total_results
459 |
460 |
461 | def check_running(db):
462 | tasks = get_tasks(db)
463 | has_oustanding = False
464 | for task in tasks:
465 | if task['status'] == 'in_progress':
466 | task_id = task['id']
467 | pid = task['pid']
468 |
469 | def update_records(task_id):
470 | sql = """
471 | UPDATE tasks
472 | SET status = 'failed'
473 | WHERE id = ?;
474 | """
475 | db.execute(sql, (task_id,))
476 | db.commit()
477 | try:
478 | process = psutil.Process(pid)
479 | # Check if process with pid exists
480 | if process.is_running() and process.ppid() == os.getpid():
481 | # is still in progress, nothing to do
482 | has_oustanding = True
483 | else:
484 | update_records(task_id)
485 | except psutil.NoSuchProcess:
486 | update_records(task_id)
487 | if not has_oustanding:
488 | session['warn_parallelism'] = False
489 |
490 | return has_oustanding
491 |
492 |
493 | def get_tasks(db):
494 | sql = """
495 | SELECT * FROM tasks ORDER BY created_at DESC
496 | """
497 | tasks = db.execute(sql)
498 | return tasks.fetchall()
499 |
500 | def get_exports(db):
501 | sql = """
502 | SELECT * FROM exports ORDER BY created_at DESC
503 | """
504 | exports = db.execute(sql)
505 | return exports.fetchall()
506 |
507 | def get_export_by_id(db, id):
508 | sql = """
509 | SELECT * FROM exports WHERE id = ?
510 | """
511 | export = db.execute(sql, id)
512 | return export.fetchone()
513 |
514 | def get_prompt_by_id(db, id):
515 | sql = """
516 | SELECT * FROM prompts WHERE id = ?
517 | """
518 | prompt = db.execute(sql, (id,))
519 | return prompt.fetchone()
520 |
521 | def get_examples_by_prompt_id(db, prompt_id, with_tags=True):
522 | if with_tags:
523 | sql = """
524 | SELECT e.*, GROUP_CONCAT(t.value) AS tags
525 | FROM examples e
526 | LEFT JOIN tags t ON e.id = t.example_id
527 | WHERE e.prompt_id = ?
528 | GROUP BY e.id
529 | """
530 | else:
531 | sql = """
532 | SELECT * FROM examples WHERE prompt_id = ?
533 | """
534 | examples = db.execute(sql, (prompt_id,))
535 | return examples.fetchall()
536 |
537 | def get_projects(db):
538 | sql = """
539 | SELECT * FROM projects ORDER BY created_at DESC
540 | """
541 | examples = db.execute(sql)
542 | return examples.fetchall()
543 |
544 | def get_project_by_id(db, id):
545 | sql = """
546 | SELECT * FROM projects WHERE id = ?
547 | """
548 | examples = db.execute(sql, (id,))
549 | return examples.fetchone()
550 |
551 | def get_styles_by_project_id(db, id):
552 | sql = """
553 | SELECT * FROM styles WHERE project_id = ?
554 | """
555 | examples = db.execute(sql, (id,))
556 | return examples.fetchall()
557 |
558 | def get_styles(db):
559 | sql = """
560 | SELECT * FROM styles ORDER BY created_at DESC
561 | """
562 | styles = db.execute(sql)
563 | return styles.fetchall()
564 |
565 | def get_style_by_id(db, id):
566 | sql = """
567 | SELECT * FROM styles WHERE id = ?
568 | """
569 | style = db.execute(sql, (id,))
570 | return style.fetchone()
571 |
572 | def get_keys_by_style_id(db, id):
573 | sql = """
574 | SELECT * FROM style_keys WHERE style_id = ?
575 | """
576 | keys = db.execute(sql, (id,))
577 | return keys.fetchall()
578 |
579 | def get_tags_by_prompt_id(db, prompt_id):
580 | sql = """
581 | SELECT * FROM tags WHERE prompt_id = ?
582 | """
583 | tags = db.execute(sql, (prompt_id,))
584 | return tags.fetchall()
585 |
586 | def get_prompt_values_by_prompt_id(db, prompt_id):
587 | c = db.cursor()
588 | sql = """
589 | SELECT * FROM prompt_values WHERE prompt_id = ?
590 | """
591 | vals = c.execute(sql, (prompt_id,))
592 | return vals.fetchall()
593 |
594 | def add_project(db, name, description):
595 | sql = """
596 | INSERT INTO projects (name, desc)
597 | VALUES (?, ?)
598 | """
599 | c = db.cursor()
600 | c.execute(sql, (name, description))
601 | proj_id = c.lastrowid
602 | db.commit()
603 | return proj_id
604 |
605 | def add_style(db, idtext, format_string, completion_key, preview_key, project_id, style_keys):
606 | sql = """
607 | INSERT INTO styles (id_text, template, completion_key, preview_key, project_id)
608 | VALUES (?, ?, ?, ?, ?)
609 | """
610 | c = db.cursor()
611 | c.execute(sql, (idtext, format_string, completion_key, preview_key, project_id))
612 | style_id = c.lastrowid
613 | db.commit()
614 |
615 | # Now add style keys
616 | for key in style_keys:
617 | sql = """
618 | INSERT INTO style_keys (name, style_id)
619 | VALUES (?, ?)
620 | """
621 | c.execute(sql, (key, style_id))
622 | db.commit()
623 |
624 | return style_id
625 |
626 | def delete_project(db, project_id):
627 | c = db.cursor()
628 | # Remove all associated completions
629 | sql = """
630 | DELETE FROM examples
631 | WHERE prompt_id IN (
632 | SELECT id FROM prompts
633 | WHERE project_id = ?
634 | );
635 | """
636 | c.execute(sql, (project_id,))
637 |
638 | # Remove all associated prompts
639 | c.execute("DELETE FROM prompts WHERE project_id = ?", (project_id,))
640 |
641 | # Remove all associated styles
642 | c.execute("DELETE FROM styles WHERE project_id = ?", (project_id,))
643 |
644 | # Remove project
645 | c.execute("DELETE FROM projects WHERE id = ?", (project_id,))
646 |
647 | db.commit()
648 |
649 | def update_project(db, project_id, description, name):
650 | c = db.cursor()
651 |
652 | if description:
653 | c.execute("UPDATE projects SET desc = ? WHERE id = ?", (description, project_id))
654 | if name:
655 | c.execute("UPDATE projects SET name = ? WHERE id = ?", (name, project_id))
656 |
657 | db.commit()
658 |
659 | def update_style(db, style_id, id_text, template, completion_key, preview_key):
660 | c = db.cursor()
661 |
662 | new_keys = get_named_arguments(template)
663 | c.execute("SELECT * FROM style_keys WHERE style_id = ?", (style_id,))
664 | old_keys = c.fetchall()
665 |
666 | missing_keys = [x['name'] for x in old_keys if x['name'] not in new_keys]
667 | added_keys = [x for x in new_keys if x not in [y['name'] for y in old_keys]]
668 |
669 | for key in missing_keys:
670 | c.execute("DELETE FROM style_keys WHERE name LIKE ? AND style_id = ?", (key, style_id))
671 | # Also we delete all the prompt values associated with the removed key
672 | c.execute("""
673 | DELETE FROM prompt_values
674 | WHERE prompt_values.key LIKE ?
675 | AND prompt_values.prompt_id IN (
676 | SELECT prompts.id FROM prompts
677 | JOIN styles ON prompts.style = styles.id
678 | WHERE styles.id = ?
679 | )
680 | """, (key, style_id))
681 |
682 | for key in added_keys:
683 | sql = """
684 | INSERT INTO style_keys (name, style_id)
685 | VALUES (?, ?)
686 | """
687 | c.execute(sql, (key, style_id))
688 |
689 | if id_text:
690 | c.execute("UPDATE styles SET id_text = ? WHERE id = ?", (id_text, style_id))
691 | if template:
692 | c.execute("UPDATE styles SET template = ? WHERE id = ?", (template, style_id))
693 | if completion_key:
694 | c.execute("UPDATE styles SET completion_key = ? WHERE id = ?", (completion_key, style_id))
695 | if preview_key:
696 | c.execute("UPDATE styles SET preview_key = ? WHERE id = ?", (preview_key, style_id))
697 |
698 | db.commit()
699 |
700 |
701 | def delete_style(db, style_id):
702 | c = db.cursor()
703 | # Remove all associated completions
704 | sql = """
705 | DELETE FROM examples
706 | WHERE prompt_id IN (
707 | SELECT id FROM prompts
708 | WHERE style = ?
709 | );
710 | """
711 | c.execute(sql, (style_id,))
712 |
713 | # Remove all associated prompts
714 | c.execute("DELETE FROM prompts WHERE style = ?", (style_id,))
715 |
716 | # Remove the style
717 | c.execute("DELETE FROM styles WHERE id = ?", (style_id,))
718 |
719 | db.commit()
720 |
721 |
--------------------------------------------------------------------------------
/app/exporting.py:
--------------------------------------------------------------------------------
1 | import os
2 | from flask import (
3 | Blueprint,
4 | redirect,
5 | render_template,
6 | request,
7 | send_file,
8 | current_app,
9 | session
10 | )
11 | from app.db import get_db
12 | from app.db_wrappers import export, get_exports, get_export_by_id, get_styles, get_projects, check_running
13 | from app.utils import tag_string_to_list
14 |
15 | bp = Blueprint('exporting', __name__)
16 |
17 | @bp.route('/export', methods=('GET', 'POST'))
18 | def exp():
19 | db = get_db()
20 | if request.method == "POST":
21 |
22 | if check_running(db):
23 | return render_template('export.html', styles=get_styles(db), projects=get_projects(db), error="Cannot create new export while current task is incomplete.")
24 |
25 | filename = request.form.get('filename') if request.form.get('filename') else "export.json"
26 |
27 | tags = request.form.get('tags')
28 | tags = tag_string_to_list(tags)
29 |
30 | content = request.form.get('content')
31 | style_id = request.form.get('style_id')
32 | project_id = request.form.get('project_id')
33 | example = request.form.get('example')
34 |
35 | export(db, filename=filename, tags=tags, content=content, example=example, project_id=project_id, style_id=style_id)
36 | return redirect("/tasks")
37 |
38 | return render_template('export.html', styles=get_styles(db), projects=get_projects(db))
39 |
40 | @bp.route('/exports', methods=('GET',))
41 | def exps():
42 | download_id = request.args.get('download_id')
43 | db = get_db()
44 | if download_id:
45 | row = get_export_by_id(db, download_id)
46 | path = os.path.join(current_app.config.get('EXPORTS_PATH'), row['filename'])
47 | return send_file(path, as_attachment=True)
48 |
49 | exports = get_exports(db)
50 |
51 | return render_template('exports.html', exports=exports)
52 |
--------------------------------------------------------------------------------
/app/home.py:
--------------------------------------------------------------------------------
1 | from flask import (
2 | Blueprint,
3 | render_template,
4 | request
5 | )
6 | from app.db import get_db
7 | from app.db_wrappers import delete_project, get_projects, search_prompts, get_styles, delete_prompt
8 | from app.utils import tag_string_to_list
9 | import json
10 |
11 | bp = Blueprint('home', __name__)
12 |
13 | @bp.route('/', methods=('GET',))
14 | def home():
15 | return render_template('home.html')
16 |
17 |
18 | @bp.route('/manifest', methods=('GET', 'POST'))
19 | def manifest():
20 |
21 | db = get_db()
22 |
23 | # When a user selects prompts via check box and clicks a button
24 | # This will trigger a special json response, not the web page
25 | if request.method == 'POST':
26 | form_data = request.json
27 | selected_prompts = form_data['prompt_ids']
28 | action = form_data['action']
29 | if action == 'delete':
30 | # One day, this should probably be batched
31 | for prompt_id in selected_prompts:
32 | delete_prompt(db, prompt_id)
33 | response_text = json.dumps({'response': 'success', 'prompts_affected': selected_prompts, 'action': action})
34 | return response_text
35 |
36 | limit = 100
37 | offset = request.args.get('offset')
38 | if not offset:
39 | offset = 0
40 |
41 | content_arg = request.args.get("content")
42 | example_arg = request.args.get("example")
43 | tags_arg = request.args.get("tags")
44 | tags_arg = tag_string_to_list(tags_arg)
45 | project_id_arg = request.args.get("project_id")
46 | if project_id_arg == "":
47 | project_id_arg = None
48 | style_id_arg = request.args.get("style_id")
49 | if style_id_arg == "":
50 | style_id_arg = None
51 |
52 | projects = get_projects(db)
53 | styles = get_styles(db)
54 |
55 | prompts, total_results = search_prompts(db, limit, offset, content_arg, example_arg, tags_arg, project_id_arg, style_id_arg)
56 |
57 | return render_template('manifest.html', prompts=prompts, page_size=limit, total_results=total_results, projects=projects, styles=styles)
58 |
--------------------------------------------------------------------------------
/app/projects.py:
--------------------------------------------------------------------------------
1 | from flask import (
2 | Blueprint,
3 | render_template,
4 | request,
5 | redirect
6 | )
7 | from app.db import get_db
8 | from app.db_wrappers import delete_project, get_projects, get_project_by_id, get_styles_by_project_id, add_project, add_style, delete_project, update_project
9 | from app.utils import get_named_arguments
10 |
11 | bp = Blueprint('projects', __name__)
12 |
13 | @bp.route('/projects', methods=('GET', 'POST'))
14 | def projects():
15 |
16 | if request.method == "POST":
17 | name = request.form.get('name')
18 | description = request.form.get('description')
19 | add_project(get_db(), name, description)
20 |
21 | projects = get_projects(get_db())
22 | return render_template('projects.html', projects=projects)
23 |
24 | @bp.route('/project', methods=('GET', 'POST'))
25 | def project():
26 |
27 | db = get_db()
28 |
29 | errors = []
30 |
31 | if request.method == 'POST':
32 | project_id = request.form.get('project_id')
33 | form_type = request.form.get('form_type')
34 |
35 | if form_type == 'delete':
36 | if project_id:
37 | delete_project(db, project_id)
38 | # if there's no project id, we'll just redirect the user back to projects
39 | elif form_type == 'update':
40 | name = request.form.get('name')
41 | description = request.form.get('description')
42 | update_project(db, project_id, description, name)
43 | else: # form_type might be null
44 | # Add style
45 |
46 | idtext = request.form.get('id_text')
47 | format_string = request.form.get('format_string')
48 | completion_key = request.form.get('completion_key')
49 | preview_key = request.form.get('preview_key')
50 |
51 | style_keys = get_named_arguments(format_string)
52 |
53 | # In the future, should throw errors when completion/preview key not in style_keys
54 | if completion_key not in style_keys:
55 | errors.append("Completion key not in the format string")
56 | elif preview_key not in style_keys:
57 | errors.append("Preview key not in the format string")
58 | else:
59 | add_style(db, idtext, format_string, completion_key, preview_key, project_id, style_keys)
60 |
61 | project_id = request.args.get("id")
62 |
63 | if not project_id:
64 | return redirect('/projects')
65 |
66 | project = get_project_by_id(get_db(), project_id)
67 | styles = get_styles_by_project_id(get_db(), project_id)
68 | return render_template('project.html', project=project, styles=styles, errors=errors)
69 |
--------------------------------------------------------------------------------
/app/schema.sql:
--------------------------------------------------------------------------------
1 | DROP TABLE IF EXISTS examples;
2 | DROP TABLE IF EXISTS prompts;
3 | DROP TABLE IF EXISTS metrics;
4 | DROP TABLE IF EXISTS tasks;
5 | DROP TABLE IF EXISTS exports;
6 | DROP TABLE IF EXISTS styles;
7 | DROP TABLE IF EXISTS projects;
8 | DROP TABLE IF EXISTS tags;
9 | DROP TABLE IF EXISTS prompt_values;
10 | DROP TABLE IF EXISTS style_keys;
11 |
12 | CREATE TABLE examples (
13 | `id` INTEGER PRIMARY KEY AUTOINCREMENT,
14 | `prompt_id` INTEGER,
15 | `completion` TEXT,
16 | `created_at` TIMESTAMP DEFAULT CURRENT_TIMESTAMP
17 | );
18 |
19 | CREATE TABLE tags (
20 | `id` INTEGER PRIMARY KEY AUTOINCREMENT,
21 | `value` TEXT,
22 | `prompt_id` INTEGER, /* Optional prompt association */
23 | `example_id` INTEGER /* Optional example association */
24 | );
25 |
26 | CREATE TABLE prompts (
27 | `id` INTEGER PRIMARY KEY AUTOINCREMENT,
28 | `style` INTEGER,
29 | `project_id` INTEGER,
30 | `created_at` TIMESTAMP DEFAULT CURRENT_TIMESTAMP
31 | );
32 |
33 | CREATE TABLE prompt_values (
34 | `id` INTEGER PRIMARY KEY AUTOINCREMENT,
35 | `prompt_id` INTEGER,
36 | `key` TEXT,
37 | `value` TEXT
38 | );
39 |
40 | CREATE TABLE styles (
41 | `id` INTEGER PRIMARY KEY AUTOINCREMENT,
42 | `id_text` TEXT, /* like 'instruct' or 'chat' */
43 | `project_id` INTEGER,
44 | `template` TEXT,
45 | `completion_key` TEXT, /* Where the LLM is meant to add */
46 | `preview_key` TEXT, /* For table previews of the prompts */
47 | `created_at` TIMESTAMP DEFAULT CURRENT_TIMESTAMP
48 | );
49 |
50 | CREATE TABLE style_keys (
51 | `id` INTEGER PRIMARY KEY AUTOINCREMENT,
52 | `name` TEXT,
53 | `style_id` INTEGER
54 | );
55 |
56 | CREATE TABLE projects (
57 | `id` INTEGER PRIMARY KEY AUTOINCREMENT,
58 | `name` TEXT,
59 | `desc` TEXT,
60 | `created_at` TIMESTAMP DEFAULT CURRENT_TIMESTAMP
61 | );
62 |
63 | CREATE TABLE metrics (
64 | `id` INTEGER PRIMARY KEY AUTOINCREMENT,
65 | `name` TEXT,
66 | `score` TEXT,
67 | `example_id` INTEGER,
68 | `created_at` TIMESTAMP DEFAULT CURRENT_TIMESTAMP
69 | );
70 |
71 | CREATE TABLE tasks (
72 | `id` INTEGER PRIMARY KEY AUTOINCREMENT,
73 | `user_id` INTEGER,
74 | `type` TEXT,
75 | `pid` INTEGER,
76 | `status` TEXT,
77 | `created_at` TIMESTAMP DEFAULT CURRENT_TIMESTAMP
78 | );
79 |
80 | CREATE TABLE exports (
81 | `id` INTEGER PRIMARY KEY AUTOINCREMENT,
82 | `user_id` INTEGER,
83 | `filename` TEXT,
84 | `created_at` TIMESTAMP DEFAULT CURRENT_TIMESTAMP
85 | );
86 |
87 | INSERT INTO projects (`name`, `desc`)
88 | VALUES ("Sample Project",
89 | "My first project"
90 | );
91 |
92 | INSERT INTO styles (`id_text`, `project_id`, `template`, `completion_key`, `preview_key`)
93 | VALUES ("instruct",
94 | 1,
95 | "Instruction: {instruction}
96 | Input: {input}
97 | Output: {output}",
98 | "output",
99 | "instruction"
100 | );
101 |
102 | INSERT INTO style_keys (`name`, `style_id`)
103 | VALUES ("instruction",
104 | 1
105 | );
106 |
107 | INSERT INTO style_keys (`name`, `style_id`)
108 | VALUES ("input",
109 | 1
110 | );
111 |
112 | INSERT INTO style_keys (`name`, `style_id`)
113 | VALUES ("output",
114 | 1
115 | );
116 |
117 | INSERT INTO prompts (style, project_id)
118 | VALUES (1,
119 | 1
120 | );
121 |
122 | INSERT INTO prompt_values (`prompt_id`, `key`, `value`)
123 | VALUES (1,
124 | "instruction",
125 | "Write a line of Python that prints Hello World"
126 | );
127 |
128 | INSERT INTO examples (prompt_id, completion)
129 | VALUES (1,
130 | "print('Hello World')"
131 | );
132 |
133 | INSERT INTO tags (`example_id`, `value`)
134 | VALUES (1,
135 | "coding"
136 | );
137 |
--------------------------------------------------------------------------------
/app/static/add.css:
--------------------------------------------------------------------------------
1 |
2 | .prompt-textarea {
3 | width: 50%;
4 | height: 10em;
5 | }
6 |
--------------------------------------------------------------------------------
/app/static/manifest.css:
--------------------------------------------------------------------------------
1 | #search-form {
2 | max-width: calc(100% - 2em);
3 | margin-bottom: 1em;
4 | background-color: #f2f2f2;
5 | padding: 1em;
6 | }
7 |
8 | label {
9 | margin: 10px 0;
10 | }
11 |
12 | button{
13 | margin-top: 10px;
14 | }
15 |
16 | .pagination {
17 | display: flex;
18 | padding: .5rem;
19 | flex-wrap: wrap;
20 | }
21 |
22 | .pagination a {
23 | margin-right: 1rem;
24 | text-decoration: none;
25 | color: black;
26 | }
27 |
28 | .pagination .selected-page-number {
29 | font-weight: bold;
30 | text-decoration: underline;
31 | }
32 |
--------------------------------------------------------------------------------
/app/static/projects.css:
--------------------------------------------------------------------------------
1 |
2 | #add-project textarea{
3 | width: 50%;
4 | height: 10em;
5 | }
6 |
7 | #add-project input[type=text]{
8 | width: 50%;
9 | }
--------------------------------------------------------------------------------
/app/static/style.css:
--------------------------------------------------------------------------------
1 | html {
2 | font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, Oxygen, Ubuntu, Cantarell, 'Open Sans', 'Helvetica Neue', sans-serif;
3 | background: white;
4 | }
5 |
6 | html, body {
7 | height: 100%;
8 | }
9 |
10 | nav {
11 | padding-bottom: .5em;
12 | border-bottom: 1px black solid;
13 | }
14 |
15 | nav a{
16 | all: unset;
17 | }
18 |
19 | .nav-bar-text, .nav-bar-main-header {
20 | padding-right: .75em;
21 | cursor: pointer;
22 | }
23 |
24 | .nav-bar-main-header {
25 | font-size: 1.5em;
26 | }
27 |
28 | #footer {
29 | height: auto;
30 | padding: 20px;
31 | background-color: #f0f0f0;
32 | color: #808080;
33 | font-size: .75em;
34 | }
35 |
36 | #footer a{
37 | all: unset;
38 | }
39 |
40 | #wrapper {
41 | min-height: calc(100% + 40px);
42 | display: flex;
43 | flex-direction: column;
44 | }
45 |
46 | #page-content {
47 | flex-grow: 1;
48 | margin-bottom: 200px; /* minimum 200px distance from the footer */
49 | }
50 |
51 | .warning {
52 | background-color: #f0f0f0;
53 | text-align: center;
54 | padding: 5px;
55 | margin-bottom: 10px;
56 | }
--------------------------------------------------------------------------------
/app/static/style_viewer.css:
--------------------------------------------------------------------------------
1 | .textarea-template {
2 | width: 50%;
3 | height: 20em;
4 | }
--------------------------------------------------------------------------------
/app/static/table.css:
--------------------------------------------------------------------------------
1 | table {
2 | border-collapse: collapse;
3 | width: 100%;
4 | }
5 |
6 | td, th {
7 | border: 1px solid #ddd;
8 | padding: .5rem;
9 | text-align: left;
10 | }
11 |
12 | th {
13 | background-color: #f2f2f2;
14 | }
--------------------------------------------------------------------------------
/app/static/view.css:
--------------------------------------------------------------------------------
1 |
2 | .prompt, .completion-editor {
3 | padding: 12px;
4 | background-color: #f0f0f0;
5 | border: 1px solid #808080;
6 | }
7 |
8 | #add-completion {
9 | display: none;
10 | }
11 |
12 | #add-completion textarea{
13 | width: 50%;
14 | height: 10em;
15 | }
16 |
17 | .completion-editor {
18 | width: calc(100% - 24px);
19 | margin-top: .5em;
20 | }
21 |
22 | .edit-completion-button, .edit-prompt-button, .edit-style{
23 | margin-top: .5em;
24 | }
25 |
26 | .completions-header {
27 | font-size: 1.5em;
28 | }
29 |
30 | .delete-button {
31 | margin-top: .5em;
32 | }
--------------------------------------------------------------------------------
/app/style.py:
--------------------------------------------------------------------------------
1 | from flask import (
2 | Blueprint,
3 | render_template,
4 | request,
5 | redirect
6 | )
7 | from app.db import get_db
8 | from app.db_wrappers import delete_style, get_project_by_id, get_style_by_id, get_keys_by_style_id, update_style
9 | from app.utils import get_named_arguments
10 | import json
11 |
12 | bp = Blueprint('style', __name__)
13 |
14 | @bp.route('/style', methods=('GET', 'POST'))
15 | def style():
16 |
17 | errors = []
18 |
19 | db = get_db()
20 | if request.method == "POST":
21 | style_id = request.form.get('style_id')
22 | form_type = request.form.get('form_type')
23 | if form_type == "update":
24 | template = request.form.get('template')
25 | id_text = request.form.get('id_text')
26 | completion_key = request.form.get('completion_key')
27 | preview_key = request.form.get('preview_key')
28 |
29 | keys = get_named_arguments(template)
30 | if preview_key not in keys:
31 | errors.append("Preview key must be in template")
32 | if completion_key not in keys:
33 | errors.append("Completion key must be in template")
34 |
35 | if len(errors) == 0:
36 | update_style(db, style_id, id_text, template, completion_key, preview_key)
37 | elif form_type == 'delete':
38 | if style_id:
39 | delete_style(db, style_id)
40 |
41 |
42 | id = request.args.get('id')
43 |
44 | if not id:
45 | return redirect("/projects")
46 |
47 | style = get_style_by_id(db, id)
48 | proj_id = style['project_id']
49 | project = get_project_by_id(db, proj_id)
50 |
51 | keys = get_keys_by_style_id(db, id)
52 |
53 | return render_template('style.html', style=style, project=project, keys=keys, errors=errors)
54 |
--------------------------------------------------------------------------------
/app/tasks.py:
--------------------------------------------------------------------------------
1 | from flask import (
2 | Blueprint,
3 | render_template,
4 | request
5 | )
6 | from app.db import get_db
7 | from app.db_wrappers import get_tasks, check_running
8 |
9 | bp = Blueprint('tasks', __name__)
10 |
11 | @bp.route('/tasks', methods=('GET',))
12 | def tasks():
13 | db = get_db()
14 | check_running(db)
15 | tasks = get_tasks(db)
16 | return render_template('tasks.html', tasks=tasks)
17 |
--------------------------------------------------------------------------------
/app/templates/add.html:
--------------------------------------------------------------------------------
1 | {% extends 'base.html' %}
2 |
3 | {% block title %}Add{% endblock %}
4 |
5 | {% block meta %}
6 |
11 |
Add Data
12 | {% if error %}
13 |
14 | Error: {{ error }}
15 |
16 | {% endif %}
17 |
18 |
38 |
39 |
59 |
{% block title %}{% endblock %} - Praetor
4 |
10 | Warning: no changes may be made to the database while a
task is still outstanding.
11 |
14 | Praetor Prompts Projects Exports Tasks
20 |
21 |
22 | {% block content %}{% endblock %}
23 |
24 |
30 |
--------------------------------------------------------------------------------
/app/templates/export.html:
--------------------------------------------------------------------------------
1 | {% extends 'base.html' %}
2 |
3 | {% block title %}Export{% endblock %}
4 |
5 | {% block meta %}
6 | {% endblock %}
7 |
8 | {% block content %}
9 |
10 |
11 | Export
12 |
13 | {% if error %}
14 |
15 | Error: {{ error }}
16 |
17 | {% endif %}
18 |
19 |
56 |
57 |
61 |
11 |
12 | Exports
13 |
14 |
15 | Exports live in instance/exports, but downloading one will add a copy to your downloads folder.
16 |
17 |
18 |
19 |
20 | Download
21 | Time
22 | Filename
23 |
24 |
25 |
26 | {% set ns = namespace(count=0) %}
27 | {% for row in exports %}
28 | {% set ns.count = ns.count + 1 %}
29 |
30 | Download {{ row.created_at }}
32 | {{ row.filename }}
33 |
34 | {% endfor %}
35 | {% if ns.count == 0 %}
36 | No exports made yet - you can export completions from the prompt manifest screen (the prompt search area).
39 |
40 |
7 |
8 | Welcome!
9 |
10 |
11 | Welcome to Praetor, a dataset exploration tool.
12 |
13 |
14 | See your prompt manifest here .
15 |
16 |
17 |
18 | About
19 |
20 |
21 | Praetor is a super lightweight finetuning data and prompt management tool.
22 | The setup requirements are minimal, and so is the complexity.
23 |
24 |
25 | In general, the system works like this: you start with a Project.
26 | A project has associated with it some prompt styles, which define format strings for the prompts.
27 | You can then add prompts in a particular style and then one or more completions to those prompts.
28 | When you're done editing prompts or completions, you can export that data in a json format.
29 | You can also import data in the same format.
30 |
Projects --> Styles --> Prompts --> Completions
33 |
34 |
35 | Sample Data
36 |
37 |
38 | A sample project is added for you when the database is initialized. The existing style placed in the project will work with alpaca.json,
39 | which is in the sample-datasets folder. Go to Projects >> Sample Project >> add data (in style "instruct") >> Bulk Upload from sample-datasets/alpaca.json.
40 |
41 |
42 |
15 |
16 | Prompts
17 |
18 |
19 |
61 |
62 | Total Results:
63 | {{ "{:,d}".format(total_results) }}
64 |
65 | Total Selected:
66 | 🗑
76 |
77 |
78 |
79 |
101 |
102 | {% set total_pages = ((total_results - 1) // page_size) + 1 %}
103 | {% set current_offset = query_params['offset']|default(0)|int %}
104 | {% set current_page = current_offset // page_size + 1 %}
105 | {% set page_window = 5 %}
106 | {% set last_shown_page = [current_page+page_window, total_pages]|min %}
107 |
108 |
138 |
139 |
220 |
30 |
85 |
Styles
86 |
98 |
Add Style
99 |
100 |
101 | A style is defined by its format string , which is text including named arguments in {curly braces}.
102 | idtext is a short name to which you can refer to the style, like `instruct` or `chat`.
104 | preview key is the named argument of your format string that will appear in the prompt manifest (viewer).
106 | completion key is the space into which your language model's completion will go.
108 |
109 |
125 |
126 |
127 |
128 |
129 | {% for error in errors %}
130 |
131 | Error submitting style: {{error}}
132 |
133 | {% endfor %}
134 |
135 |
136 |
137 |
138 | Link
139 | Name
140 | Add Data in Style
141 | Time Created
142 |
143 |
144 |
145 | {% for row in styles %}
146 |
147 | View {{ row.id_text }}
149 | Add Data {{ row.created_at }}
151 |
152 | {% endfor %}
153 |
154 |
155 |
12 |
13 | Projects
14 |
15 |
16 | Projects contain styles and prompts made in the image of those styles.
17 |
18 |
30 |
Add
31 |
32 |
33 |
34 |
43 |
46 |
47 |
48 |
49 | Link
50 | Time
51 | Name
52 | Description
53 |
54 |
55 |
56 | {% for row in projects %}
57 |
58 | View {{ row.created_at }}
60 | {{ row.name }}
61 | {{ row.desc }}
62 |
63 | {% endfor %}
64 |
65 |
66 |
23 |
24 | Style
25 |
26 |
107 |
11 |
12 | Tasks
13 |
14 |
15 | Tasks are long-running processes, such as large exports or large imports.
16 |
17 |
18 |
19 |
20 | Time
21 | Type
22 | Status
23 |
24 |
25 |
26 | {% for row in tasks %}
27 |
28 | {{ row.created_at }}
29 | {{ row.type }}
30 | {{ row.status }}
31 |
32 | {% endfor %}
33 |
34 |
35 |
23 | {% if prompt %}
24 |
25 |
26 | Prompt
27 |
28 |
47 |
53 |
54 |
55 |
56 |
57 |
60 |
Add
61 |
62 |
72 |
73 |
74 |
75 |
78 | {% for row in completions %}
79 |
95 |
102 |
103 | {% endfor %}
104 |
105 |
106 | {% else %}
107 |
No Prompt Found
108 | {% endif %}
109 |
\w+)}')
18 | return pattern.findall(fmt_string)
19 |
20 |
--------------------------------------------------------------------------------
/app/view.py:
--------------------------------------------------------------------------------
1 | from flask import (
2 | Blueprint,
3 | redirect,
4 | render_template,
5 | request
6 | )
7 | from app.db import get_db
8 | from app.db_wrappers import add_example, delete_example, get_keys_by_style_id, update_example, get_examples_by_prompt_id, get_prompt_by_id, get_prompt_values_by_prompt_id, get_style_by_id, get_tags_by_prompt_id, update_prompt, delete_prompt
9 | from app.utils import tag_string_to_list
10 |
11 | bp = Blueprint('view', __name__)
12 |
13 | @bp.route('/view', methods=('GET', 'POST'))
14 | def view():
15 |
16 | prompt_id = request.args.get("prompt_id")
17 |
18 | db = get_db()
19 |
20 | if request.method == "POST":
21 |
22 | update_type = request.form.get("update_type")
23 |
24 | if update_type == "update_prompt":
25 | new_prompt_values = {}
26 | for key, value in request.form.items():
27 | if key.find("key.") == 0:
28 | new_prompt_values[key[4:]] = value
29 | tags = request.form.get("tags")
30 | tags = tag_string_to_list(tags)
31 | update_prompt(db, prompt_id, new_prompt_values, tags)
32 | elif update_type == "delete_prompt":
33 | delete_prompt(db, prompt_id)
34 | elif update_type == "add_completion":
35 | completion = request.form.get("completion")
36 | tags = request.form.get("tags")
37 | tags = tag_string_to_list(tags)
38 | add_example(db, prompt_id, completion, tags)
39 | elif update_type == "update_completion":
40 | example_id = request.form.get("id")
41 | completion = request.form.get("completion")
42 | tags = request.form.get("tags")
43 | tags = tag_string_to_list(tags)
44 | update_example(db, example_id, completion, tags)
45 | elif update_type == "delete_completion":
46 | example_id = request.form.get("id")
47 | delete_example(db, example_id)
48 |
49 | prompt_dict = get_prompt_by_id(db, prompt_id)
50 | tags = None
51 | prompt_values = None
52 | completions = None
53 | style = None
54 | tags_str = None
55 | if prompt_dict:
56 | tags = get_tags_by_prompt_id(db, prompt_id)
57 | tags_str = ",".join([tag['value'] for tag in tags])
58 |
59 | prompt_values = get_prompt_values_by_prompt_id(db, prompt_id)
60 | style = get_style_by_id(db, prompt_dict['style'])
61 | style_keys = get_keys_by_style_id(db, prompt_dict['style'])
62 | prompt_value_keys = [x['key'] for x in prompt_values]
63 | for key in style_keys:
64 | if key['name'] == style['completion_key']:
65 | continue
66 | if key['name'] not in prompt_value_keys:
67 | prompt_values.append({
68 | 'key': key['name'],
69 | 'value': ""
70 | })
71 |
72 |
73 | completions = get_examples_by_prompt_id(db, prompt_id, with_tags=True)
74 |
75 | return render_template('view.html', prompt=prompt_dict, style=style, prompt_values=prompt_values, prompt_tags=tags_str, completions=completions)
--------------------------------------------------------------------------------
/requirements.txt:
--------------------------------------------------------------------------------
1 | flask
2 | psutil
3 |
--------------------------------------------------------------------------------
/screenshots/praetor-completion-screen.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/goodreasonai/praetor-data/b7af7eaea1c2e5657d9b3f3c9321650c5886aac9/screenshots/praetor-completion-screen.png
--------------------------------------------------------------------------------
/screenshots/praetor-prompts-screen.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/goodreasonai/praetor-data/b7af7eaea1c2e5657d9b3f3c9321650c5886aac9/screenshots/praetor-prompts-screen.png
--------------------------------------------------------------------------------
/screenshots/praetor-style-screen.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/goodreasonai/praetor-data/b7af7eaea1c2e5657d9b3f3c9321650c5886aac9/screenshots/praetor-style-screen.png
--------------------------------------------------------------------------------