This commit is contained in:
20
docs/snippets/r/utils/embedding_with_chunking.md
Normal file
20
docs/snippets/r/utils/embedding_with_chunking.md
Normal file
@@ -0,0 +1,20 @@
|
||||
```r title="R"
|
||||
library(kreuzberg)
|
||||
|
||||
config <- list(
|
||||
chunking = list(
|
||||
max_characters = 1024L,
|
||||
overlap = 100L,
|
||||
embedding = list(
|
||||
model = list(type = "preset", name = "balanced"),
|
||||
normalize = TRUE,
|
||||
batch_size = 32L
|
||||
)
|
||||
)
|
||||
)
|
||||
|
||||
json <- extract_file_sync("document.pdf", "application/pdf", config)
|
||||
result <- jsonlite::fromJSON(json, simplifyVector = FALSE)
|
||||
|
||||
cat(sprintf("Chunks with embeddings: %d\n", length(result$chunks)))
|
||||
```
|
||||
Reference in New Issue
Block a user