upload version 1.1

README.md

@@ -59,14 +59,14 @@ Annotations guidelines were updated continuously during dataset refinement and m
 1. ingredient lists in all languages must be annotated.
 2. ingredients list should start with the first ingredient, without `ingredient` prefix ("Ingredients:", "Zutaten", "Ingrédients: ") or `language` prefix ("EN:", "FR - ",...)
 3. ingredient list containing single ingredients without any `ingredient` or `language` prefix should not be annotated. Otherwise, it's very difficult to know whether the mention is the ingredient list or just a random mention of an ingredient on the packaging.
-4. We have a very restrictive approach on where the ingredient list ends: we don't include any extra information (allergen, origin, trace, organic mentions) at the end of the ingredient list. The only exception is when this information is in bracket after the ingredient. This rule is in place to make it easier for the detector to know what is an ingredient list and what is not. Additional information can be added afterward as a post-processing step.
+4. We don't include most extra information (allergen, origin, trace) at the end of the ingredient list. We now include organic mentions (as of v1.1 of the dataset). Another exception is when the extra information is in bracket after the ingredient. This rule is in place to make it easier for the detector to know what is an ingredient list and what is not.
 
 ## Dataset schema
 
 The dataset is made of 2 JSONL files:
 
-- `ingredient_detection_dataset-v1_train.jsonl.gz`: train split, 5065 samples
-- `ingredient_detection_dataset-v1_test.jsonl.gz`: test split, 556 samples
+- `ingredient_detection_dataset_train.jsonl.gz`: train split, 5065 samples
+- `ingredient_detection_dataset_test.jsonl.gz`: test split, 556 samples
 
 Each sample has the following fields:
 
@@ -78,4 +78,17 @@ Each sample has the following fields:
 - `meta`: a dict containing additional meta-data about the sample:
     - `barcode`: the product barcode of the image that was used
     - `image_id`: unique digit identifier of the image for the product
-    - `url`: image URL from which the text was extracted
+    - `url`: image URL from which the text was extracted
+
+## Versions
+
+### v1.1
+
+New version of the dataset with the following modifications:
+
+- included “*organic”/”*issu de l’agriculture biologique” suffixes as part of the ingredient list. This suffix was added manually as a postprocessing step and it did not make much sense not to include it. This was only included if the mention was just after the ingredient list (if allergen info was in between it was not included).
+- Some other annotation errors were fixed as well.
+
+### v1.0
+
+First version of the dataset

ingredient_detection_dataset-v1_test.jsonl.gz → ingredient_detection_dataset_test.jsonl.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:83aeb4dd2551f178281f01d6c067a1a70585b4153fdf74aa43bdff43e87b27e6
-size 388781
+oid sha256:1fc0f5b627fac7e9825c96d0a11d6dca39865ffbfcab7ebc2529116aac7ade90
+size 365163

ingredient_detection_dataset-v1_train.jsonl.gz → ingredient_detection_dataset_train.jsonl.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e5c8af31ac05cc820f5083eaefae0ee7f7350677972b9c01480aa221c0f5e928
-size 3508881
+oid sha256:a7cdf1be46548e049328cb57a3e280e7c2d5c195c48d34832274ae1151565995
+size 3154206