English image_classifier_vit_gtsrb_model ViTForImageClassification from bazyl

Description

Pretrained VIT model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP.image_classifier_vit_gtsrb_model is a English model originally trained by bazyl.

Predicted Entities

Children crossing, Double curve, Road work, Yield, Beware of ice/snow, Speed limit (70km/h), Bicycles crossing, Roundabout mandatory, Speed limit (30km/h), Keep left, Dangerous curve left, No vehicles, End of no passing, Bumpy road, Speed limit (50km/h), Turn left ahead, Speed limit (20km/h), General caution, Speed limit (100km/h), End speed + passing limits, Go straight or right, Dangerous curve right, Speed limit (80km/h), Slippery road, Turn right ahead, No passing veh over 3.5 tons, Speed limit (60km/h), Pedestrians, Right-of-way at intersection, Priority road, End of speed limit (80km/h), Road narrows on the right, No entry, Stop, Wild animals crossing, Veh > 3.5 tons prohibited, End no passing veh > 3.5 tons, Go straight or left, Speed limit (120km/h), Ahead only, Keep right, Traffic signals, No passing

Live Demo Open in Colab Download Copy S3 URI

How to use

image_assembler = ImageAssembler() \
    .setInputCol("image") \
    .setOutputCol("image_assembler")

imageClassifier = ViTForImageClassification \
    .pretrained("image_classifier_vit_gtsrb_model", "en")\
    .setInputCols("image_assembler") \
    .setOutputCol("class")

pipeline = Pipeline(stages=[
    image_assembler,
    imageClassifier,
])

pipelineModel = pipeline.fit(imageDF)

pipelineDF = pipelineModel.transform(imageDF)

val imageAssembler = new ImageAssembler()
.setInputCol("image")
.setOutputCol("image_assembler")

val imageClassifier = ViTForImageClassification
.pretrained("image_classifier_vit_gtsrb_model", "en")
.setInputCols("image_assembler")
.setOutputCol("class")

val pipeline = new Pipeline().setStages(Array(imageAssembler, imageClassifier))

val pipelineModel = pipeline.fit(imageDF)

val pipelineDF = pipelineModel.transform(imageDF)

import nlu
import requests
response = requests.get('https://raw.githubusercontent.com/JohnSnowLabs/spark-nlp/master/docs/assets/images/hen.JPEG')
with open('hen.JPEG', 'wb') as f:
    f.write(response.content)
nlu.load("en.classify_image.gtsrb_model").predict("hen.JPEG")