CLIPImageClassifier wraps clip image model from transformers

Overview

CLIPImageClassifier

CLIPImageClassifier wraps clip image model from transformers.

CLIPImageClassifier is initialized with the argument classes, these are the texts that we want to classify an image to one of them The executor receives Documents with uri attribute. Each Document's uri represent the path to an image. The executor will read the image and classify it to one of the classes.

The result will be saved inside a new tag called class within the original document. The class tag is a dictionary that contains two things:

  • label: the chosen class from classes.
  • score: the confidence score in the chosen class given by the model.

Usage

Use the prebuilt images from Jina Hub in your Python code, add it to your Flow and classify your images according to chosen classes:

from jina import Flow
classes = ['this is a cat','this is a dog','this is a person']
f = Flow().add(
    uses='jinahub+docker://CLIPImageClassifier',
    uses_with={'classes':classes}
    )
docs = DocumentArray()
doc = Document(uri='/your/image/path')
docs.append(doc)

with f:
    f.post(on='/classify', inputs=docs, on_done=lambda resp: print(resp.docs[0].tags['class']['label']))

Returns

Document with class tag. This class tag which is a dict.It contains label which is an str and a float confidence score for the image.

GPU Usage

This executor also offers a GPU version. To use it, make sure to pass 'device'='cuda', as the initialization parameter, and gpus='all' when adding the containerized Executor to the Flow. See the Executor on GPU section of Jina documentation for more details.

Here's how you would modify the example above to use a GPU:

from jina import Flow

classes = ['this is a cat','this is a dog','this is a person']	
f = Flow().add(
    uses='jinahub+docker://CLIPImageClassifier',
    uses_with={
    'classes':classes,
    'device':'cuda',
    'gpus':'all'
    }
    )
docs = DocumentArray()
doc = Document(uri='/your/image/path')
docs.append(doc)

with f:
    f.post(on='/classify', inputs=docs, on_done=lambda resp: print(resp.docs[0].tags['class']['label']))

Reference

CLIP Image model

You might also like...
CLIP+FFT text-to-image
CLIP+FFT text-to-image

Aphantasia This is a text-to-image tool, part of the artwork of the same name. Based on CLIP model, with FFT parameterizer from Lucent library as a ge

A simple command line tool for text to image generation, using OpenAI's CLIP and a BigGAN.
A simple command line tool for text to image generation, using OpenAI's CLIP and a BigGAN.

Ryan Murdock has done it again, combining OpenAI's CLIP and the generator from a BigGAN! This repository wraps up his work so it is easily accessible to anyone who owns a GPU.

CLIP: Connecting Text and Image (Learning Transferable Visual Models From Natural Language Supervision)
CLIP: Connecting Text and Image (Learning Transferable Visual Models From Natural Language Supervision)

CLIP (Contrastive Language–Image Pre-training) Experiments (Evaluation) Model Dataset Acc (%) ViT-B/32 (Paper) CIFAR100 65.1 ViT-B/32 (Our) CIFAR100 6

Source code for models described in the paper "AudioCLIP: Extending CLIP to Image, Text and Audio" (https://arxiv.org/abs/2106.13043)

AudioCLIP Extending CLIP to Image, Text and Audio This repository contains implementation of the models described in the paper arXiv:2106.13043. This

improvement of CLIP features over the traditional resnet features on the visual question answering, image captioning, navigation and visual entailment tasks.

CLIP-ViL In our paper "How Much Can CLIP Benefit Vision-and-Language Tasks?", we show the improvement of CLIP features over the traditional resnet fea

 Segmentation in Style: Unsupervised Semantic Image Segmentation with Stylegan and CLIP
Segmentation in Style: Unsupervised Semantic Image Segmentation with Stylegan and CLIP

Segmentation in Style: Unsupervised Semantic Image Segmentation with Stylegan and CLIP Abstract: We introduce a method that allows to automatically se

Zero-Shot Text-to-Image Generation VQGAN+CLIP Dockerized
Zero-Shot Text-to-Image Generation VQGAN+CLIP Dockerized

VQGAN-CLIP-Docker About Zero-Shot Text-to-Image Generation VQGAN+CLIP Dockerized This is a stripped and minimal dependency repository for running loca

A Jupyter notebook to play with NVIDIA's StyleGAN3 and OpenAI's CLIP for a text-based guided image generation.

A Jupyter notebook to play with NVIDIA's StyleGAN3 and OpenAI's CLIP for a text-based guided image generation.

CLIP (Contrastive Language–Image Pre-training) trained on Indonesian data

CLIP-Indonesian CLIP (Radford et al., 2021) is a multimodal model that can connect images and text by training a vision encoder and a text encoder joi

Comments
  • CLIPImageClassifier error

    CLIPImageClassifier error

    I tried to run the following flow on "jinahub+sandbox" but I got the following error could you please share your insight with me? I am running the code from my Jupyter notebook.

    import warnings warnings.filterwarnings("ignore", category=DeprecationWarning) from jina import Flow classes = ['this is a cat','this is a dog','this is a person'] doc = Document(uri='image/dog.jpg') docs = DocumentArray() docs.append(doc) f = Flow().add( uses='jinahub://CLIPImageClassifier',name="classifier", uses_with={'classes':classes})

    with f: f.post(on='/classify', inputs=docs, on_done=lambda resp: print(resp.docs[0].tags['class']['label']))

    -----------------------error------------------ PkgResourcesDeprecationWarning: 1.1build1 is an invalid version and will not be supported in a future release (raised from /home/ubuntu/pyenv/lib/python3.10/site-packages/pkg_resources/init.py:116) PkgResourcesDeprecationWarning: 0.1.43ubuntu1 is an invalid version and will not be supported in a future release (raised from /home/ubuntu/pyenv/lib/python3.10/site-packages/pkg_resources/init.py:116) UserWarning: VersionConflict(torchvision 0.12.0+cpu (/usr/local/lib/python3.10/dist-packages), Requirement.parse('torchvision==0.10.0')) (raised from /home/ubuntu/pyenv/lib/python3.10/site-packages/jina/hubble/helper.py:483) ftfy or spacy is not installed using BERT BasicTokenizer instead of ftfy. ╭────── 🎉 Flow is ready to serve! ──────╮ │ 🔗 Protocol GRPC │ │ 🏠 Local 0.0.0.0:55600 │ │ 🔒 Private 172.31.17.247:55600 │ │ 🌍 Public 34.221.179.218:55600 │ ╰────────────────────────────────────────╯ ERROR classifier/[email protected] AttributeError("'DocumentArrayInMemory' [07/06/22 16:34:35] object has no attribute 'get_attributes'")
    add "--quiet-error" to suppress the exception details
    ╭────────────── Traceback (most recent call last) ───────────────╮
    │ /home/ubuntu/pyenv/lib/python3.10/site-packages/jina/serve/ru… │
    │ in process_data │
    │ │
    │ 162 │ │ │ │ if self.logger.debug_enabled: │
    │ 163 │ │ │ │ │ self._log_data_request(requests[0]) │
    │ 164 │ │ │ │ │
    │ ❱ 165 │ │ │ │ return await self._data_request_handler. │
    │ 166 │ │ │ except (RuntimeError, Exception) as ex: │
    │ 167 │ │ │ │ self.logger.error( │
    │ 168 │ │ │ │ │ f'{ex!r}' │
    │ │
    │ /home/ubuntu/pyenv/lib/python3.10/site-packages/jina/serve/ru… │
    │ in handle │
    │ │
    │ 147 │ │ ) │
    │ 148 │ │ │
    │ 149 │ │ # executor logic │
    │ ❱ 150 │ │ return_data = await self._executor.acall( │
    │ 151 │ │ │ req_endpoint=requests[0].header.exec_endpoin │
    │ 152 │ │ │ docs=docs, │
    │ 153 │ │ │ parameters=params, │
    │ │
    │ /home/ubuntu/pyenv/lib/python3.10/site-packages/jina/serve/ex… │
    │ in acall
    │ │
    │ 271 │ │ if req_endpoint in self.requests: │
    │ 272 │ │ │ return await self.acall_endpoint(req_end │
    │ 273 │ │ elif default_endpoint in self.requests: │
    │ ❱ 274 │ │ │ return await self.acall_endpoint(__defau │
    │ 275 │ │
    │ 276 │ async def acall_endpoint(self, req_endpoint, **k │
    │ 277 │ │ func = self.requests[req_endpoint] │
    │ │
    │ /home/ubuntu/pyenv/lib/python3.10/site-packages/jina/serve/ex… │
    │ in acall_endpoint
    │ │
    │ 292 │ │ │ if iscoroutinefunction(func): │
    │ 293 │ │ │ │ return await func(self, **kwargs) │
    │ 294 │ │ │ else: │
    │ ❱ 295 │ │ │ │ return func(self, **kwargs) │
    │ 296 │ │
    │ 297 │ @property │
    │ 298 │ def workspace(self) -> Optional[str]: │
    │ │
    │ /home/ubuntu/pyenv/lib/python3.10/site-packages/jina/serve/ex… │
    │ in arg_wrapper │
    │ │
    │ 177 │ │ │ │ def arg_wrapper( │
    │ 178 │ │ │ │ │ executor_instance, *args, **kwargs │
    │ 179 │ │ │ │ ): # we need to get the summary from th │
    │ the self │
    │ ❱ 180 │ │ │ │ │ return fn(executor_instance, *args, │
    │ 181 │ │ │ │ │
    │ 182 │ │ │ │ self.fn = arg_wrapper │
    │ 183 │
    │ │
    │ /home/ubuntu/.jina/hub-package/9k3zudzu/clip_image_classifier… │
    │ in classify │
    │ │
    │ 56 │ │ for docs_batch in docs.traverse_flat( │
    │ 57 │ │ │ parameters.get('traversal_paths', self.traver │
    │ 58 │ │ ).batch(batch_size=parameters.get('batch_size', s │
    │ ❱ 59 │ │ │ image_batch = docs_batch.get_attributes('blob │
    │ 60 │ │ │ with torch.inference_mode(): │
    │ 61 │ │ │ │ input = self._generate_input_features(cla │
    │ 62 │ │ │ │ outputs = self.model(**input) │
    ╰────────────────────────────────────────────────────────────────╯
    AttributeError: 'DocumentArrayInMemory' object has no attribute
    'get_attributes'
    Exception in thread Thread-107: Traceback (most recent call last): File "/home/ubuntu/pyenv/lib/python3.10/site-packages/jina/clients/base/grpc.py", line 86, in _get_results async for resp in stub.Call( File "/home/ubuntu/pyenv/lib/python3.10/site-packages/grpc/aio/_call.py", line 326, in _fetch_stream_responses await self._raise_for_status() File "/home/ubuntu/pyenv/lib/python3.10/site-packages/grpc/aio/_call.py", line 236, in _raise_for_status raise _create_rpc_error(await self.initial_metadata(), await grpc.aio._call.AioRpcError: <AioRpcError of RPC that terminated with: status = StatusCode.UNKNOWN details = "Unexpected <class 'grpc.aio._call.AioRpcError'>: <AioRpcError of RPC that terminated with: status = StatusCode.UNKNOWN details = "Unexpected <class 'TypeError'>: format_exception() got an unexpected keyword argument 'etype'" debug_error_string = "{"created":"@1657125275.618452649","description":"Error received from peer ipv4:0.0.0.0:58903","file":"src/core/lib/surface/call.cc","file_line":952,"grpc_message":"Unexpected <class 'TypeError'>: format_exception() got an unexpected keyword argument 'etype'","grpc_status":2}"

    " debug_error_string = "{"created":"@1657125275.619606817","description":"Error received from peer ipv4:0.0.0.0:55600","file":"src/core/lib/surface/call.cc","file_line":952,"grpc_message":"Unexpected <class 'grpc.aio._call.AioRpcError'>: <AioRpcError of RPC that terminated with:\n\tstatus = StatusCode.UNKNOWN\n\tdetails = "Unexpected <class 'TypeError'>: format_exception() got an unexpected keyword argument 'etype'"\n\tdebug_error_string = "{"created":"@1657125275.618452649","description":"Error received from peer ipv4:0.0.0.0:58903","file":"src/core/lib/surface/call.cc","file_line":952,"grpc_message":"Unexpected <class 'TypeError'>: format_exception() got an unexpected keyword argument 'etype'","grpc_status":2}"\n>","grpc_status":2}"

    The above exception was the direct cause of the following exception:

    Traceback (most recent call last): File "/usr/lib/python3.10/threading.py", line 1009, in _bootstrap_inner self.run() File "/home/ubuntu/pyenv/lib/python3.10/site-packages/jina/helper.py", line 1292, in run self.result = asyncio.run(func(*args, **kwargs)) File "/usr/lib/python3.10/asyncio/runners.py", line 44, in run return loop.run_until_complete(main) File "/usr/lib/python3.10/asyncio/base_events.py", line 646, in run_until_complete return future.result() File "/home/ubuntu/pyenv/lib/python3.10/site-packages/jina/clients/mixin.py", line 164, in _get_results async for resp in c._get_results(*args, **kwargs): File "/home/ubuntu/pyenv/lib/python3.10/site-packages/jina/clients/base/grpc.py", line 155, in _get_results raise e File "/home/ubuntu/pyenv/lib/python3.10/site-packages/jina/clients/base/grpc.py", line 135, in _get_results raise BadClient(msg) from err jina.excepts.BadClient: gRPC error: StatusCode.UNKNOWN Unexpected <class 'grpc.aio._call.AioRpcError'>: <AioRpcError of RPC that terminated with: status = StatusCode.UNKNOWN details = "Unexpected <class 'TypeError'>: format_exception() got an unexpected keyword argument 'etype'" debug_error_string = "{"created":"@1657125275.618452649","description":"Error received from peer ipv4:0.0.0.0:58903","file":"src/core/lib/surface/call.cc","file_line":952,"grpc_message":"Unexpected <class 'TypeError'>: format_exception() got an unexpected keyword argument 'etype'","grpc_status":2}"


    AttributeError Traceback (most recent call last) File ~/pyenv/lib/python3.10/site-packages/jina/helper.py:1307, in run_async(func, *args, **kwargs) 1306 try: -> 1307 return thread.result 1308 except AttributeError:

    AttributeError: '_RunThread' object has no attribute 'result'

    During handling of the above exception, another exception occurred:

    BadClient Traceback (most recent call last) Input In [15], in <cell line: 12>() 8 f = Flow().add( 9 uses='jinahub://CLIPImageClassifier',name="classifier", 10 uses_with={'classes':classes}) 12 with f: ---> 13 f.post(on='/classify', inputs=docs, on_done=lambda resp: print(resp.docs[0].tags['class']['label']))

    File ~/pyenv/lib/python3.10/site-packages/jina/clients/mixin.py:173, in PostMixin.post(self, on, inputs, on_done, on_error, on_always, parameters, target_executor, request_size, show_progress, continue_on_error, return_responses, **kwargs) 170 if return_results: 171 return result --> 173 return run_async( 174 _get_results, 175 inputs=inputs, 176 on_done=on_done, 177 on_error=on_error, 178 on_always=on_always, 179 exec_endpoint=on, 180 target_executor=target_executor, 181 parameters=parameters, 182 request_size=request_size, 183 **kwargs, 184 )

    File ~/pyenv/lib/python3.10/site-packages/jina/helper.py:1311, in run_async(func, *args, **kwargs) 1308 except AttributeError: 1309 from jina.excepts import BadClient -> 1311 raise BadClient( 1312 'something wrong when running the eventloop, result can not be retrieved' 1313 ) 1314 else: 1316 raise RuntimeError( 1317 'you have an eventloop running but not using Jupyter/ipython, ' 1318 'this may mean you are using Jina with other integration? if so, then you ' 1319 'may want to use Client/Flow(asyncio=True). If not, then ' 1320 'please report this issue here: https://github.com/jina-ai/jina' 1321 )

    BadClient: something wrong when running the eventloop, result can not be retrieved

    opened by sk-haghighi 4
Releases(v0.2)
Owner
Jina AI
A Neural Search Company. We provide the cloud-native neural search solution powered by state-of-the-art AI technology.
Jina AI
Civsim is a basic civilisation simulation and modelling system built in Python 3.8.

Civsim Introduction Civsim is a basic civilisation simulation and modelling system built in Python 3.8. It requires the following packages: perlin_noi

17 Aug 08, 2022
ML models implementation practice

Let's implement various ML algorithms with numpy/tf Vanilla Neural Network https://towardsdatascience.com/lets-code-a-neural-network-in-plain-numpy-ae

Jinsoo Heo 4 Jul 04, 2021
PED: DETR for Crowd Pedestrian Detection

PED: DETR for Crowd Pedestrian Detection Code for PED: DETR For (Crowd) Pedestrian Detection Paper PED: DETR for Crowd Pedestrian Detection Installati

36 Sep 13, 2022
Code for the paper "Query Embedding on Hyper-relational Knowledge Graphs"

Query Embedding on Hyper-Relational Knowledge Graphs This repository contains the code used for the experiments in the paper Query Embedding on Hyper-

DimitrisAlivas 19 Jul 26, 2022
Robust and Accurate Object Detection via Self-Knowledge Distillation

Robust and Accurate Object Detection via Self-Knowledge Distillation paper:https://arxiv.org/abs/2111.07239 Environments Python 3.7 Cuda 10.1 Prepare

Weipeng Xu 6 Jul 01, 2022
Research code for Arxiv paper "Camera Motion Agnostic 3D Human Pose Estimation"

GMR(Camera Motion Agnostic 3D Human Pose Estimation) This repo provides the source code of our arXiv paper: Seong Hyun Kim, Sunwon Jeong, Sungbum Park

Seong Hyun Kim 1 Feb 07, 2022
ZeroVL - The official implementation of ZeroVL

This repository contains source code necessary to reproduce the results presente

31 Nov 04, 2022
🔥🔥High-Performance Face Recognition Library on PaddlePaddle & PyTorch🔥🔥

face.evoLVe: High-Performance Face Recognition Library based on PaddlePaddle & PyTorch Evolve to be more comprehensive, effective and efficient for fa

Zhao Jian 3.1k Jan 04, 2023
PPO is a very popular Reinforcement Learning algorithm at present.

PPO is a very popular Reinforcement Learning algorithm at present. OpenAI takes PPO as the current baseline algorithm. We use the PPO algorithm to train a policy to give the best action in any situat

Rosefintech 11 Aug 23, 2021
Multiband spectro-radiometric satellite image analysis with K-means cluster algorithm

Multi-band Spectro Radiomertric Image Analysis with K-means Cluster Algorithm Overview Multi-band Spectro Radiomertric images are images comprising of

Chibueze Henry 6 Mar 16, 2022
This is the official code for the paper "Tracker Meets Night: A Transformer Enhancer for UAV Tracking".

SCT This is the official code for the paper "Tracker Meets Night: A Transformer Enhancer for UAV Tracking" The spatial-channel Transformer (SCT) enhan

Intelligent Vision for Robotics in Complex Environment 27 Nov 23, 2022
Public Code for NIPS submission SimiGrad: Fine-Grained Adaptive Batching for Large ScaleTraining using Gradient Similarity Measurement

Public code for NIPS submission "SimiGrad: Fine-Grained Adaptive Batching for Large Scale Training using Gradient Similarity Measurement" This repo co

Heyang Qin 0 Oct 13, 2021
Storchastic is a PyTorch library for stochastic gradient estimation in Deep Learning

Storchastic is a PyTorch library for stochastic gradient estimation in Deep Learning

Emile van Krieken 140 Dec 30, 2022
Reimplementation of NeurIPS'19: "Meta-Weight-Net: Learning an Explicit Mapping For Sample Weighting" by Shu et al.

[Re] Meta-Weight-Net: Learning an Explicit Mapping For Sample Weighting Reimplementation of NeurIPS'19: "Meta-Weight-Net: Learning an Explicit Mapping

Robert Cedergren 1 Mar 13, 2020
Disentangled Face Attribute Editing via Instance-Aware Latent Space Search, accepted by IJCAI 2021.

Instance-Aware Latent-Space Search This is a PyTorch implementation of the following paper: Disentangled Face Attribute Editing via Instance-Aware Lat

67 Dec 21, 2022
High-Resolution Image Synthesis with Latent Diffusion Models

Latent Diffusion Models Requirements A suitable conda environment named ldm can be created and activated with: conda env create -f environment.yaml co

CompVis Heidelberg 5.6k Jan 04, 2023
Generative Exploration and Exploitation - This is an improved version of GENE.

GENE This is an improved version of GENE. In the original version, the states are generated from the decoder of VAE. We have to check whether the gere

33 Mar 23, 2022
Research into Forex price prediction from price history using Deep Sequence Modeling with Stacked LSTMs.

Forex Data Prediction via Recurrent Neural Network Deep Sequence Modeling Research Paper Our research paper can be viewed here Installation Clone the

Alex Taradachuk 2 Aug 07, 2022
WRENCH: Weak supeRvision bENCHmark

🔧 What is it? Wrench is a benchmark platform containing diverse weak supervision tasks. It also provides a common and easy framework for development

Jieyu Zhang 176 Dec 28, 2022
Tensorflow implementation of DeepLabv2

TF-deeplab This is a Tensorflow implementation of DeepLab, compatible with Tensorflow 1.2.1. Currently it supports both training and testing the ResNe

Chenxi Liu 21 Sep 27, 2022