Content-Length: 683405 | pFad | http://github.com/postgresml/postgresml/commit/56d6b54d8c0efc32d36fd5906472e4f5ec6f65dd

75 Updated docs (#951) · postgresml/postgresml@56d6b54 · GitHub
Skip to content

Commit 56d6b54

Browse files
SilasMarvinSilasMarvin
authored
Updated docs (#951)
1 parent fb27d39 commit 56d6b54

File tree

4 files changed

+88
-51
lines changed

4 files changed

+88
-51
lines changed

‎pgml-sdks/rust/pgml/javascript/README.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -375,7 +375,7 @@ const model = pgml.newModel(
375375
376376
**Use an OpenAI model**
377377
```javascript
378-
const model = pgml.newModel(name="text-embedding-ada-002", source="openai")
378+
const model = pgml.newModel("text-embedding-ada-002", "openai")
379379
```
380380
381381
### Splitters

‎pgml-sdks/rust/pgml/javascript/tests/typescript-tests/test.ts

Lines changed: 25 additions & 14 deletions
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
1-
import pgml from '../../index.js'
1+
import pgml from "../../index.js";
22

33
//github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com/
44
//github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com/
@@ -9,29 +9,21 @@ import pgml from '../../index.js'
99
//github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com/
1010
//github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com/
1111

12-
const DATABASE_URL = process.env.DATABASE_URL;
13-
if (!DATABASE_URL) {
14-
console.log("No DATABASE_URL environment variable found. Please set one")
15-
process.exit(1)
16-
}
1712
const LOG_LEVEL = process.env.LOG_LEVEL ? process.env.LOG_LEVEL : "ERROR";
18-
1913
pgml.js_init_logger(LOG_LEVEL);
2014

2115
const generate_dummy_documents = (count: number) => {
2216
let docs = [];
2317
for (let i = 0; i < count; i++) {
2418
docs.push({
25-
"id": i,
26-
"text": `This is a test document: ${i}`,
27-
"metadata": {
28-
"uuid": i * 10,
29-
"name": `Test Document ${i}`
30-
}
19+
id: i,
20+
text: `This is a test document: ${i}`,
21+
uuid: i * 10,
22+
name: `Test Document ${i}`,
3123
});
3224
}
3325
return docs;
34-
}
26+
};
3527

3628
//github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com/
3729
// Test the API exposed is correct //github.com///github.com///github.com///github.com///github.com//
@@ -118,6 +110,25 @@ it("can vector search with query builder with remote embeddings", async() => {
118110
await collection.archive();
119111
});
120112

113+
it("can vector search with query builder and metadata filtering", async () => {
114+
let model = pgml.newModel();
115+
let splitter = pgml.newSplitter();
116+
let pipeline = pgml.newPipeline("test_j_p_cvswqbamf_0", model, splitter);
117+
let collection = pgml.newCollection("test_j_c_cvswqbamf_4");
118+
await collection.upsert_documents(generate_dummy_documents(3));
119+
await collection.add_pipeline(pipeline);
120+
let results = await collection
121+
.query()
122+
.vector_recall("Here is some query", pipeline)
123+
.filter({
124+
metadata: {
125+
$or: [{ uuid: { $eq: 0 } }, { uuid: { $eq: 20 } }],
126+
},
127+
})
128+
.limit(10).fetch_all();
129+
expect(results).toHaveLength(2);
130+
await collection.archive();
131+
});
121132

122133
//github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com///github.com/
123134
// Test user output facing functions //github.com///github.com///github.com///github.com///

‎pgml-sdks/rust/pgml/python/README.md

Lines changed: 4 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -315,13 +315,13 @@ results = (
315315
{
316316
"$and": [
317317
{
318-
"$eq": {
319-
"uuid": 1
318+
"uuid": {
319+
"$eq": 1
320320
}
321321
},
322322
{
323-
"$lt": {
324-
"index": 100
323+
"index": {
324+
"$lt": 100
325325
}
326326
}
327327
]

‎pgml-sdks/rust/pgml/python/tests/test.py

Lines changed: 58 additions & 32 deletions
Original file line numberDiff line numberDiff line change
@@ -30,7 +30,8 @@ def generate_dummy_documents(count: int) -> List[Dict[str, Any]]:
3030
"id": i,
3131
"text": "This is a test document: {}".format(i),
3232
"some_random_thing": "This will be metadata on it",
33-
"metadata": {"uuid": i * 10, "name": "Test Document {}".format(i)},
33+
"uuid": i * 10,
34+
"name": "Test Document {}".format(i),
3435
}
3536
)
3637
return dummy_documents
@@ -135,6 +136,31 @@ async def test_can_vector_search_with_query_builder_with_remote_embeddings():
135136
await collection.archive()
136137

137138

139+
@pytest.mark.asyncio
140+
async def test_can_vector_search_with_query_builder_and_metadata_filtering():
141+
model = pgml.Model()
142+
splitter = pgml.Splitter()
143+
pipeline = pgml.Pipeline("test_p_p_tcvswqbamf_1", model, splitter)
144+
collection = pgml.Collection(name="test_p_c_tcvswqbamf_2")
145+
await collection.upsert_documents(generate_dummy_documents(3))
146+
await collection.add_pipeline(pipeline)
147+
results = (
148+
await collection.query()
149+
.vector_recall("Here is some query", pipeline)
150+
.filter({
151+
"metadata": {
152+
"uuid": {
153+
"$eq": 0
154+
}
155+
}
156+
})
157+
.limit(10)
158+
.fetch_all()
159+
)
160+
assert len(results) == 1
161+
await collection.archive()
162+
163+
138164
###################################################
139165
## Test user output facing functions ##############
140166
###################################################
@@ -158,17 +184,17 @@ async def test_pipeline_to_dict():
158184
###################################################
159185

160186

161-
def vector_search(collection_name, pipeline_name):
162-
collection = pgml.Collection(collection_name)
163-
pipeline = pgml.Pipeline(pipeline_name)
164-
result = asyncio.run(
165-
collection.query()
166-
.vector_recall("Here is some query", pipeline)
167-
.limit(10)
168-
.fetch_all()
169-
)
170-
print(result)
171-
return [0, 1, 2]
187+
# def vector_search(collection_name, pipeline_name):
188+
# collection = pgml.Collection(collection_name)
189+
# pipeline = pgml.Pipeline(pipeline_name)
190+
# result = asyncio.run(
191+
# collection.query()
192+
# .vector_recall("Here is some query", pipeline)
193+
# .limit(10)
194+
# .fetch_all()
195+
# )
196+
# print(result)
197+
# return [0, 1, 2]
172198

173199

174200
# @pytest.mark.asyncio
@@ -200,23 +226,23 @@ def vector_search(collection_name, pipeline_name):
200226
###################################################
201227

202228

203-
async def silas_test_add_pipeline():
204-
model = pgml.Model()
205-
splitter = pgml.Splitter()
206-
pipeline = pgml.Pipeline("silas_test_p_1", model, splitter)
207-
collection = pgml.Collection(name="silas_test_c_10")
208-
await collection.add_pipeline(pipeline)
209-
210-
async def silas_test_upsert_documents():
211-
collection = pgml.Collection(name="silas_test_c_9")
212-
await collection.upsert_documents(generate_dummy_documents(10))
213-
214-
async def silas_test_vector_search():
215-
pipeline = pgml.Pipeline("silas_test_p_1")
216-
collection = pgml.Collection(name="silas_test_c_9")
217-
results = await collection.vector_search("Here is some query", pipeline)
218-
print(results)
219-
220-
# asyncio.run(silas_test_add_pipeline())
221-
# asyncio.run(silas_test_upsert_documents())
222-
# asyncio.run(silas_test_vector_search())
229+
# async def test_add_pipeline():
230+
# model = pgml.Model()
231+
# splitter = pgml.Splitter()
232+
# pipeline = pgml.Pipeline("silas_test_p_1", model, splitter)
233+
# collection = pgml.Collection(name="silas_test_c_10")
234+
# await collection.add_pipeline(pipeline)
235+
#
236+
# async def test_upsert_documents():
237+
# collection = pgml.Collection(name="silas_test_c_9")
238+
# await collection.upsert_documents(generate_dummy_documents(10))
239+
#
240+
# async def test_vector_search():
241+
# pipeline = pgml.Pipeline("silas_test_p_1")
242+
# collection = pgml.Collection(name="silas_test_c_9")
243+
# results = await collection.vector_search("Here is some query", pipeline)
244+
# print(results)
245+
246+
# asyncio.run(test_add_pipeline())
247+
# asyncio.run(test_upsert_documents())
248+
# asyncio.run(test_vector_search())

0 commit comments

Comments
 (0)








ApplySandwichStrip

pFad - (p)hone/(F)rame/(a)nonymizer/(d)eclutterfier!      Saves Data!


--- a PPN by Garber Painting Akron. With Image Size Reduction included!

Fetched URL: http://github.com/postgresml/postgresml/commit/56d6b54d8c0efc32d36fd5906472e4f5ec6f65dd

Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy