diff --git a/content/data/datachain.yml b/content/data/datachain.yml index 1432c2a9ec..2c69177928 100644 --- a/content/data/datachain.yml +++ b/content/data/datachain.yml @@ -9,10 +9,10 @@ prompt = "How many people in the image?" - people = Dataset("s3://my-storage/") \ - .filter(C.name.glob("*.jpg")) \ - .map(DescribeImage(prompt, max_token = 300)) \ - .mutate(num_people = int(C.descr)) + people = DataChain.from_storage("s3://my-storage/", type="image") \ + .filter(C.name.glob("*.jpg")) \ + .map(DescribeImage(prompt, max_token = 300)) \ + .mutate(num_people = int(C.descr)) people.filter(C.num_people > 3) @@ -21,6 +21,9 @@ Save the results of a query in a dataset that you can use to train your ML models. terminal: | - $ datachain query my-query.py my-dataset - .......... - Dataset 'my-dataset' created + from datachain.lib.dc import C, DataChain + + images = DataChain.from_storage("s3://my-storage/", type="image") \ + .filter(C.name.glob("*.jpg")) + + images.save("fashion-product-images") diff --git a/src/components/Home/Hero/BetterTogether/index.tsx b/src/components/Home/Hero/BetterTogether/index.tsx index 8fa25d069c..a935c8a5a8 100644 --- a/src/components/Home/Hero/BetterTogether/index.tsx +++ b/src/components/Home/Hero/BetterTogether/index.tsx @@ -5,7 +5,7 @@ const BetterTogether = () => { return (

- Datachain and DVC: Better Together + DataChain and DVC: Better Together

Build the datasets you need without modifying your data sources. Create diff --git a/src/components/Home/Hero/HeroSection.tsx b/src/components/Home/Hero/HeroSection.tsx index 7553016cb6..0727475c84 100644 --- a/src/components/Home/Hero/HeroSection.tsx +++ b/src/components/Home/Hero/HeroSection.tsx @@ -190,7 +190,7 @@ const HeroSection = () => { }) }} > - Learn about Datachain + Learn about DataChain