Duplicate from lorenzoscottb/phrase-entailment

Co-authored-by: Lorenzo Bertolini <lorenzoscottb@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,34 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

OOV_Train_2.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6fb9dff8ebc3bfff5fe07d2ed347b1ed0d9d05d84643f51a91bcf16b5e54be32
+size 4128194

README.md

@@ -0,0 +1,14 @@
+---
+title: Phrase Entailment
+emoji: ❌✅
+colorFrom: blue
+colorTo: blue
+sdk: gradio
+sdk_version: 3.16.2
+app_file: app.py
+pinned: false
+license: cc-by-nc-2.0
+duplicated_from: lorenzoscottb/phrase-entailment
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,68 @@
+import gradio as gr
+import pickle
+import pandas as pd
+
+
+data = pickle.load(open("OOV_Train_2.pkl", "rb"))
+data = pd.DataFrame(
+    data, 
+    columns=["Input_Seq", "Label", "Adj_Class", "Adj", "Nn", "Hypr", "Adj_NN"]
+)
+
+adjs = set(data["Adj"])
+Nns  = set(list(data["Nn"]) + list(data["Hypr"]))
+all_set = set(list(adjs) + list(Nns))
+
+def test_input(words):
+    word_dict = ""
+    for w in words.split(","):
+        if w in all_set:
+            word_dict += "{} : in-distribution\n".format(w)
+        else:
+            word_dict += "{} : out-of-distribution\n".format(w)
+    return word_dict
+
+    
+title = "Phrase-Entailment Detection with BERT"
+
+description = """
+Did you know that logically speaking **A small cat is not a small animal**, and that **A fake smile is not a smile**? Learn more by testing our BERT model tuned to perform phrase-level adjective-noun entailment. The proposed model was tuned with a section of the PLANE (**P**hrase-**L**evel **A**djective-**N**oun **E**ntailment) dataset, introduced in COLING 2022 [Bertolini et al.,](https://aclanthology.org/2022.coling-1.359/). Please note that the scope of the model is not to run lexical-entailment or hypernym detection (e.g., *"A dog is an animal*"), but to perform a very specific subset of phrase-level compositional entailment over adjective-noun phrases. The type of question you can ask the model are limited, and should have one of three forms:
+
+- An *Adjective-Noun* is a *Noun* (e.g. A red car is a car)
+
+- An *Adjective-Noun* is a *Hypernym(Noun)* (e.g. A red car is a vehicle)
+
+- An *Adjective-Noun* is a *Adjective-Hypernym(Noun)* (e.g. A red car is a red vehicle)
+
+As in the examples above, the **adjective should be the same for both phrases**, and the **Hypernym(Noun) should be a true hypernym of the selected noun**.
+
+The current model achieves an accuracy of 90% on out-of-distribution evaluation.
+Use the next page to check if your test-items (i.e. adjective, noun and hypernyms) were part of the training data!"""
+
+examples = [["A red car is a vehicle"], ["A fake smile is a smile"], ["A small cat is a small animal"]]
+
+interface_model = gr.Interface.load(
+            "huggingface/lorenzoscottb/bert-base-cased-PLANE-ood-2",
+            description=description,
+            examples=examples,
+            title=title,
+)
+
+
+description_w = """
+You can use this page to test if a set of words was included in the training data used to tune the model. As in the samples below, use as input a series of words separated solely by a comma (e.g. *red,car,vehicle*).
+"""
+
+examples_w = [["red,car,vehicle"], ["fake,smile"], ["small,cat,animal"]]
+
+interface_words = gr.Interface(
+            fn=test_input,
+            inputs=gr.Textbox(label="Input:word_1,word2,...,word_n"),
+            outputs=gr.Textbox(label="In training-distribution?"),
+            description=description_w,
+            examples=examples_w,
+)
+
+gr.TabbedInterface(
+    [interface_model, interface_words], ["Test Model", "Check if words in/out-distribution"]
+).launch()

requirements.txt

@@ -0,0 +1 @@
+pandas