Full-text search
Search in
Scope to owner or repo
4,139 results
deepmind / language-perceiver
README.md
model
21 matches
tags: transformers, pytorch, safetensors, perceiver, fill-mask, en, dataset:wikipedia, dataset:c4, arxiv:1810.04805, arxiv:2107.14795, arxiv:2004.03720, license:apache-2.0, deploy:azure, region:us
11
# Perceiver IO for language
12
13
Perceiver IO model pre-trained on the Masked Language Modeling (MLM) task proposed in [BERT](https://arxiv.org/abs/1810....
15
Disclaimer: The team releasing Perceiver IO did not write a model card for this model so this model card has been writte...
⋯
19
Perceiver IO is a transformer encoder model that can be applied on any modality (text, images, audio, video, ...). The c...
⋯
23
...datasets/huggingface/documentation-images/resolve/main/perceiver_architecture.jpg" alt="drawing" width="600"/>
24
25
<small> Perceiver IO architecture.</small>
27
... mechanism don't depend on the size of the inputs, the Perceiver IO authors train the model directly on raw UTF-8 bytes,...
⋯
30
classifier using the features produced by the Perceiver model as inputs.
⋯
34
...del hub](https://huggingface.co/models?search=deepmind/perceiver) to look for fine-tuned versions on a task that interes...
deepmind / vision-perceiver-conv
README.md
model
20 matches
tags: transformers, pytorch, perceiver, image-classification, dataset:imagenet, arxiv:2107.14795, license:apache-2.0, autotrain_compatible, endpoints_compatible, region:us, deploy:azure
8
# Perceiver IO for vision (convolutional processing)
9
10
Perceiver IO model pre-trained on ImageNet (14 million images, 1,000 classes) at resolution 224x224. It was introduced i...
12
Disclaimer: The team releasing Perceiver IO did not write a model card for this model so this model card has been writte...
⋯
16
Perceiver IO is a transformer encoder model that can be applied on any modality (text, images, audio, video, ...). The c...
⋯
20
...datasets/huggingface/documentation-images/resolve/main/perceiver_architecture.jpg" alt="drawing" width="600"/>
21
22
<small> Perceiver IO architecture.</small>
24
... mechanism don't depend on the size of the inputs, the Perceiver IO authors can train the model directly on raw pixel va...
⋯
30
...del hub](https://huggingface.co/models?search=deepmind/perceiver) to look for other fine-tuned versions on a task that m...
⋯
37
from transformers import PerceiverFeatureExtractor, PerceiverForImageClassificationConvProcessing
deepmind / vision-perceiver-fourier
README.md
model
20 matches
tags: transformers, pytorch, perceiver, image-classification, dataset:imagenet, arxiv:2107.14795, license:apache-2.0, autotrain_compatible, endpoints_compatible, region:us, deploy:azure
7
# Perceiver IO for vision (fixed Fourier position embeddings)
8
9
Perceiver IO model pre-trained on ImageNet (14 million images, 1,000 classes) at resolution 224x224. It was introduced i...
11
Disclaimer: The team releasing Perceiver IO did not write a model card for this model so this model card has been writte...
⋯
15
Perceiver IO is a transformer encoder model that can be applied on any modality (text, images, audio, video, ...). The c...
⋯
19
...datasets/huggingface/documentation-images/resolve/main/perceiver_architecture.jpg" alt="drawing" width="600"/>
20
21
<small> Perceiver IO architecture.</small>
23
... mechanism don't depend on the size of the inputs, the Perceiver IO authors can train the model directly on raw pixel va...
⋯
29
...del hub](https://huggingface.co/models?search=deepmind/perceiver) to look for other fine-tuned versions on a task that m...
⋯
36
from transformers import PerceiverImageProcessor, PerceiverForImageClassificationFourier
deepmind / vision-perceiver-learned
README.md
model
20 matches
tags: transformers, pytorch, perceiver, image-classification, dataset:imagenet, arxiv:2107.14795, license:apache-2.0, endpoints_compatible, deploy:azure, region:us
7
# Perceiver IO for vision (learned position embeddings)
8
9
Perceiver IO model pre-trained on ImageNet (14 million images, 1,000 classes) at resolution 224x224. It was introduced i...
11
Disclaimer: The team releasing Perceiver IO did not write a model card for this model so this model card has been writte...
⋯
15
Perceiver IO is a transformer encoder model that can be applied on any modality (text, images, audio, video, ...). The c...
⋯
19
...datasets/huggingface/documentation-images/resolve/main/perceiver_architecture.jpg" alt="drawing" width="600"/>
20
21
<small> Perceiver IO architecture.</small>
23
... mechanism don't depend on the size of the inputs, the Perceiver IO authors can train the model directly on raw pixel va...
⋯
29
...del hub](https://huggingface.co/models?search=deepmind/perceiver) to look for other fine-tuned versions on a task that m...
⋯
36
from transformers import PerceiverFeatureExtractor, PerceiverForImageClassificationLearned
expertlearning / vision-perceiver-fourier
README.md
model
19 matches
tags: transformers, pytorch, perceiver, image-classification, dataset:imagenet, arxiv:2107.14795, license:apache-2.0, autotrain_compatible, endpoints_compatible, region:us
7
# Perceiver IO for vision (fixed Fourier position embeddings)
8
9
Perceiver IO model pre-trained on ImageNet (14 million images, 1,000 classes) at resolution 224x224. It was introduced i...
11
Disclaimer: The team releasing Perceiver IO did not write a model card for this model so this model card has been writte...
⋯
15
Perceiver IO is a transformer encoder model that can be applied on any modality (text, images, audio, video, ...). The c...
⋯
19
...datasets/huggingface/documentation-images/resolve/main/perceiver_architecture.jpg" alt="drawing" width="600"/>
20
21
<small> Perceiver IO architecture.</small>
23
... mechanism don't depend on the size of the inputs, the Perceiver IO authors can train the model directly on raw pixel va...
⋯
29
...del hub](https://huggingface.co/models?search=deepmind/perceiver) to look for other fine-tuned versions on a task that m...
⋯
36
from transformers import PerceiverImageProcessor, PerceiverForImageClassificationFourier
krasserm / perceiver-io-mlm
README.md
model
27 matches
tags: transformers, pytorch, safetensors, perceiver-io-masked-language-model, fill-mask, en, dataset:c4, dataset:wikipedia, arxiv:2107.14795, license:apache-2.0, region:us
12
# Perceiver IO masked language model
13
14
This model is a Perceiver IO model pretrained on the masked language modeling (MLM) task using a text corpus created
15
from [C4](https://huggingface.co/datasets/c4) and [English Wikipedia](https://huggingface.co/datasets/wikipedia). It
16
is weight-equivalent to the [deepmind/language-perceiver](https://huggingface.co/deepmind/language-perceiver) model
17
but based on implementation classes of the [perceiver-io](https://github.com/krasserm/perceiver-io) library. It can
18
be created from the `deepmind/language-perceiver` model with a library-specific [conversion utility](#model-conversion)....
⋯
21
Content of the `deepmind/language-perceiver` [model card](https://huggingface.co/deepmind/language-perceiver)
⋯
27
... Section 4 (Table 1) and Appendix F (Table 11) of the [Perceiver IO paper](https://arxiv.org/abs/2107.14795)
⋯
34
([example](https://huggingface.co/krasserm/perceiver-io-mlm-imdb)) or fine-tuning on a labeled dataset using the
35
pretrained encoder of this model ([example](https://huggingface.co/krasserm/perceiver-io-txt-clf-imdb)) for weight
⋯
40
...ou first need to [install](https://github.com/krasserm/perceiver-io/blob/main/README.md#installation)
41
the `perceiver-io` library with extension `text`.
krasserm / perceiver-io-img-clf
README.md
model
25 matches
tags: transformers, pytorch, safetensors, perceiver-io-image-classifier, image-classification, dataset:imagenet-1k, arxiv:2107.14795, license:apache-2.0, region:us
9
# Perceiver IO image classifier
10
11
This model is a Perceiver IO model pretrained on ImageNet (14 million images, 1,000 classes). It is weight-equivalent
12
to the [deepmind/vision-perceiver-fourier](https://huggingface.co/deepmind/vision-perceiver-fourier) model but based on
13
implementation classes of the [perceiver-io](https://github.com/krasserm/perceiver-io) library. It can be created from
14
the `deepmind/vision-perceiver-fourier` model with a library-specific [conversion utility](#model-conversion). Both
⋯
17
Content of the `deepmind/vision-perceiver-fourier` [model card](https://huggingface.co/deepmind/vision-perceiver-fourier...
⋯
23
The model is specif in Appendix A of the [Perceiver IO paper](https://arxiv.org/abs/2107.14795) (2D Fourier features).
⋯
31
...ou first need to [install](https://github.com/krasserm/perceiver-io/blob/main/README.md#installation)
32
the `perceiver-io` library with extension `text`.
⋯
35
pip install perceiver-io[text]
krasserm / perceiver-io-optical-flow
README.md
model
34 matches
tags: transformers, pytorch, perceiver-io-optical-flow, dataset:autoflow, arxiv:2107.14795, license:apache-2.0, region:us
8
# Perceiver IO optical flow model
9
10
This model is a Perceiver IO optical flow model pretrained on [AutoFlow](https://autoflow-google.github.io/).
11
It is weight-equivalent to the [deepmind/optical-flow-perceiver](https://huggingface.co/deepmind/optical-flow-perceiver)
12
model but based on implementation classes of the [perceiver-io](https://github.com/krasserm/perceiver-io) library. It
13
can be created from the `deepmind/optical-flow-perceiver` model with a library-specific [conversion utility](#model-conv...
⋯
16
Content of the `deepmind/optical-flow-perceiver` [model card](https://huggingface.co/deepmind/optical-flow-perceiver)
⋯
22
The model is specified in Appendix H (Table 16) of the [Perceiver IO paper](https://arxiv.org/abs/2107.14795).
⋯
30
...ou first need to [install](https://github.com/krasserm/perceiver-io/blob/main/README.md#installation)
31
the `perceiver-io` library with extension `vision`.
⋯
34
pip install perceiver-io[vision]
deepmind / multimodal-perceiver
README.md
model
17 matches
tags: transformers, pytorch, perceiver, dataset:kinetics-700-2020, arxiv:2010.10864, arxiv:2107.14795, license:apache-2.0, endpoints_compatible, region:us
7
# Perceiver IO for multimodal autoencoding
8
9
Perceiver IO model trained on [Kinetics-700-2020](https://arxiv.org/abs/2010.10864) for auto-encoding videos that consis...
⋯
13
Disclaimer: The team releasing Perceiver IO did not write a model card for this model so this model card has been writte...
⋯
17
Perceiver IO is a transformer encoder model that can be applied on any modality (text, images, audio, video, ...). The c...
⋯
21
...datasets/huggingface/documentation-images/resolve/main/perceiver_architecture.jpg" alt="drawing" width="600"/>
22
23
<small> Perceiver IO architecture.</small>
25
... mechanism don't depend on the size of the inputs, the Perceiver IO authors can train the model by padding the inputs (i...
⋯
31
...del hub](https://huggingface.co/models search=deepmind/perceiver) to look for other versions on a task that may interest...
deepmind / optical-flow-perceiver
README.md
model
17 matches
tags: transformers, pytorch, safetensors, perceiver, dataset:autoflow, arxiv:2107.14795, license:apache-2.0, endpoints_compatible, region:us
7
# Perceiver IO for optical flow
8
9
Perceiver IO model trained on [AutoFlow](https://autoflow-google.github.io/). It was introduced in the paper [Perceiver ...
⋯
13
Disclaimer: The team releasing Perceiver IO did not write a model card for this model so this model card has been writte...
⋯
17
Perceiver IO is a transformer encoder model that can be applied on any modality (text, images, audio, video, ...). The c...
⋯
21
...datasets/huggingface/documentation-images/resolve/main/perceiver_architecture.jpg" alt="drawing" width="600"/>
22
23
<small> Perceiver IO architecture.</small>
25
... mechanism don't depend on the size of the inputs, the Perceiver IO authors can train the model on raw pixel values, by ...
⋯
31
...del hub](https://huggingface.co/models?search=deepmind/perceiver) to look for other versions on a task that may interest...
BEE-spoke-data / language-perceiver-title2genre
README.md
model
3 matches
tags: transformers, safetensors, perceiver, text-classification, book, genre, book title, base_model:deepmind/language-perceiver, base_model:finetune:deepmind/language-perceiver, license:apache-2.0, endpoints_compatible, region:us
35
# language-perceiver for title-genre classification
36
37
This model is a fine-tuned version of [deepmind/language-perceiver](https://huggingface.co/deepmind/language-perceiver) ...
oandreae / financial_sentiment_model
README.md
model
3 matches
tags: transformers, pytorch, tensorboard, perceiver, text-classification, generated_from_trainer, dataset:financial_phrasebank, license:apache-2.0, model-index, autotrain_compatible, endpoints_compatible, region:us
36
# financial_sentiment_model
37
38
This model is a fine-tuned version of [deepmind/language-perceiver](https://huggingface.co/deepmind/language-perceiver) ...
openflamingo / OpenFlamingo-9B-deprecated
README.md
model
2 matches
gurumurthy3 / gpt2vl-stackformer-v2
README.md
model
5 matches
tags: image-captioning, gpt2, clip, vision-language, dataset:AKCIT/coco2017-captioning, arxiv:2204.14198, license:mit, region:us
27
**CLIP ViT-B/16** vision tower through a **Flamingo-style Perceiver Resampler** and
⋯
44
Perceiver Resampler (3 layers, 8 heads)
⋯
63
- **Perceiver Resampler** — compresses the CLIP token sequence down to 64 fixed latent
⋯
80
| Visual tokens | 64 (Perceiver Resampler, depth 3, 8 heads) |
⋯
125
- Resampler design: [Flamingo](https://arxiv.org/abs/2204.14198) (DeepMind)
HuggingFaceM4 / idefics-80b
README.md
model
4 matches
tags: transformers, pytorch, safetensors, idefics, image-text-to-text, multimodal, text, image, image-to-text, text-generation, en, dataset:HuggingFaceM4/OBELICS, dataset:wikipedia, dataset:facebook/pmd, dataset:laion/laion2B-en, arxiv:2204.14198, arxiv:2306.16527, arxiv:2303.12733, arxiv:2302.05442, arxiv:1910.07467, arxiv:2204.02311, arxiv:2306.05425, arxiv:1808.10584, arxiv:2109.05014, arxiv:2307.06281, license:other, text-generation-inference, deploy:azure, region:us
26
...98), a closed-source visual language model developed by Deepmind. Like GPT-4, the multimodal model accepts arbitrary seq...
⋯
211
...r and vision hidden states are pooled with Transformer Perceiver blocks and then fused into the text sequence through th...
213
...lization on the projected queries and keys of both the Perceiver and cross-attention blocks, which improved training sta...
⋯
220
| Perceiver Resampler | Number of Layers | 6 | 6 |
HuggingFaceM4 / idefics-9b
README.md
model
4 matches
tags: transformers, pytorch, safetensors, idefics, image-text-to-text, multimodal, text, image, image-to-text, text-generation, en, dataset:HuggingFaceM4/OBELICS, dataset:wikipedia, dataset:facebook/pmd, dataset:laion/laion2B-en, arxiv:2204.14198, arxiv:2306.16527, arxiv:2303.12733, arxiv:2302.05442, arxiv:1910.07467, arxiv:2204.02311, arxiv:2306.05425, arxiv:1808.10584, arxiv:2109.05014, arxiv:2307.06281, license:other, text-generation-inference, deploy:azure, region:us
26
...98), a closed-source visual language model developed by Deepmind. Like GPT-4, the multimodal model accepts arbitrary seq...
⋯
211
...r and vision hidden states are pooled with Transformer Perceiver blocks and then fused into the text sequence through th...
213
...lization on the projected queries and keys of both the Perceiver and cross-attention blocks, which improved training sta...
⋯
220
| Perceiver Resampler | Number of Layers | 6 | 6 |
HuggingFaceM4 / idefics-9b-instruct
README.md
model
4 matches
tags: transformers, pytorch, safetensors, idefics, image-text-to-text, multimodal, text, image, image-to-text, text-generation, en, dataset:HuggingFaceM4/OBELICS, dataset:wikipedia, dataset:facebook/pmd, dataset:laion/laion2B-en, arxiv:2204.14198, arxiv:2306.16527, arxiv:2303.12733, arxiv:2302.05442, arxiv:1910.07467, arxiv:2204.02311, arxiv:2306.05425, arxiv:1808.10584, arxiv:2109.05014, arxiv:2307.06281, license:other, text-generation-inference, deploy:azure, region:us
26
...98), a closed-source visual language model developed by Deepmind. Like GPT-4, the multimodal model accepts arbitrary seq...
⋯
211
...r and vision hidden states are pooled with Transformer Perceiver blocks and then fused into the text sequence through th...
213
...lization on the projected queries and keys of both the Perceiver and cross-attention blocks, which improved training sta...
⋯
220
| Perceiver Resampler | Number of Layers | 6 | 6 |
HuggingFaceM4 / idefics-80b-instruct
README.md
model
4 matches
tags: transformers, pytorch, safetensors, idefics, image-text-to-text, multimodal, text, image, image-to-text, text-generation, en, dataset:HuggingFaceM4/OBELICS, dataset:wikipedia, dataset:facebook/pmd, dataset:laion/laion2B-en, arxiv:2204.14198, arxiv:2306.16527, arxiv:2303.12733, arxiv:2302.05442, arxiv:1910.07467, arxiv:2204.02311, arxiv:2306.05425, arxiv:1808.10584, arxiv:2109.05014, arxiv:2307.06281, license:other, text-generation-inference, deploy:azure, region:us
26
...98), a closed-source visual language model developed by Deepmind. Like GPT-4, the multimodal model accepts arbitrary seq...
⋯
211
...r and vision hidden states are pooled with Transformer Perceiver blocks and then fused into the text sequence through th...
213
...lization on the projected queries and keys of both the Perceiver and cross-attention blocks, which improved training sta...
⋯
220
| Perceiver Resampler | Number of Layers | 6 | 6 |
areegtarek / idefics-9b-instruct-all
README.md
model
4 matches
tags: transformers, safetensors, idefics, any-to-any, multimodal, text, image, image-to-text, text-generation, en, dataset:HuggingFaceM4/OBELICS, dataset:wikipedia, dataset:facebook/pmd, dataset:laion/laion2B-en, arxiv:2204.14198, arxiv:2306.16527, arxiv:2303.12733, arxiv:2302.05442, arxiv:1910.07467, arxiv:2204.02311, arxiv:2306.05425, arxiv:1808.10584, arxiv:2109.05014, arxiv:2307.06281, license:other, text-generation-inference, 4-bit, bitsandbytes, deploy:azure, region:us
24
...98), a closed-source visual language model developed by Deepmind. Like GPT-4, the multimodal model accepts arbitrary seq...
⋯
190
...r and vision hidden states are pooled with Transformer Perceiver blocks and then fused into the text sequence through th...
192
...lization on the projected queries and keys of both the Perceiver and cross-attention blocks, which improved training sta...
⋯
199
| Perceiver Resampler | Number of Layers | 6 | 6 |
georgeanton / sifta-living-os
README.md
model
25 matches
tags: arxiv:2403.12031, arxiv:2106.09685, arxiv:2305.18290, arxiv:2108.13264, arxiv:1312.6114, arxiv:1606.06565, arxiv:1912.01603, region:us
46
🦅 **Apex Predator Perceiver** — Cross-modal attention bottleneck (Perceiver IO × Native Sparse Attention × MAIN-VLA pru...
⋯
107
> ║ ✅ ALIVE Apex Predator Perceiver (Event 71) ║
108
> ║ └─ Perceiver IO × NSA × MAIN-VLA ║
⋯
1069
## 🧠 Chapter III — The DeepMind Cognitive Suite (April 18, 2026)
⋯
1077
| Layer | Biology | DeepMind / RL canon |
⋯
1087
Each layer maps to one Python module on disk. Together they form the **DeepMind Cognitive Suite**.