> For the complete documentation index, see [llms.txt](https://docs.deepdrive.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.deepdrive.io/vendor/tensorflow/models/research.md). # TensorFlow Research Models This folder contains machine learning models implemented by researchers in [TensorFlow](https://tensorflow.org). The models are maintained by their respective authors. To propose a model for inclusion, please submit a pull request. Currently, the models are compatible with TensorFlow 1.0 or later. If you are running TensorFlow 0.12 or earlier, please [upgrade your installation](https://www.tensorflow.org/install). ## Models * [adversarial\_crypto](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/adversarial_crypto/README.md): protecting communications with adversarial neural cryptography. * [adversarial\_text](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/adversarial_text/README.md): semi-supervised sequence learning with adversarial training. * [attention\_ocr](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/attention_ocr/README.md): a model for real-world image text extraction. * [audioset](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/audioset/README.md): Models and supporting code for use with [AudioSet](http://g.co/audioset). * [autoencoder](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/autoencoder/README.md): various autoencoders. * [brain\_coder](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/brain_coder/README.md): Program synthesis with reinforcement learning. * [cognitive\_mapping\_and\_planning](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/cognitive_mapping_and_planning/README.md): implementation of a spatial memory based mapping and planning architecture for visual navigation. * [compression](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/compression/README.md): compressing and decompressing images using a pre-trained Residual GRU network. * [deeplab](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/deeplab/README.md): deep labelling for semantic image segmentation. * [delf](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/delf/README.md): deep local features for image matching and retrieval. * [differential\_privacy](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/differential_privacy/README.md): differential privacy for training data. * [domain\_adaptation](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/domain_adaptation/README.md): domain separation networks. * [gan](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/gan/README.md): generative adversarial networks. * [im2txt](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/im2txt/README.md): image-to-text neural network for image captioning. * [inception](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/inception/README.md): deep convolutional networks for computer vision. * [learning\_to\_remember\_rare\_events](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/learning_to_remember_rare_events/README.md): a large-scale life-long memory module for use in deep learning. * [learning\_unsupervised\_learning](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/learning_unsupervised_learning/README.md): a meta-learned unsupervised learning update rule. * [lexnet\_nc](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/lexnet_nc/README.md): a distributed model for noun compound relationship classification. * [lfads](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/lfads/README.md): sequential variational autoencoder for analyzing neuroscience data. * [lm\_1b](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/lm_1b/README.md): language modeling on the one billion word benchmark. * [maskgan](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/maskgan/README.md): text generation with GANs. * [namignizer](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/namignizer/README.md): recognize and generate names. * [neural\_gpu](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/neural_gpu/README.md): highly parallel neural computer. * [neural\_programmer](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/neural_programmer/README.md): neural network augmented with logic and mathematic operations. * [next\_frame\_prediction](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/next_frame_prediction/README.md): probabilistic future frame synthesis via cross convolutional networks. * [object\_detection](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/object_detection/README.md): localizing and identifying multiple objects in a single image. * [pcl\_rl](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/pcl_rl/README.md): code for several reinforcement learning algorithms, including Path Consistency Learning. * [ptn](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/ptn/README.md): perspective transformer nets for 3D object reconstruction. * [qa\_kg](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/qa_kg/README.md): module networks for question answering on knowledge graphs. * [real\_nvp](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/real_nvp/README.md): density estimation using real-valued non-volume preserving (real NVP) transformations. * [rebar](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/rebar/README.md): low-variance, unbiased gradient estimates for discrete latent variable models. * [resnet](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/resnet/README.md): deep and wide residual networks. * [skip\_thoughts](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/skip_thoughts/README.md): recurrent neural network sentence-to-vector encoder. * [slim](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/slim/README.md): image classification models in TF-Slim. * [street](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/street/README.md): identify the name of a street (in France) from an image using a Deep RNN. * [swivel](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/swivel/README.md): the Swivel algorithm for generating word embeddings. * [syntaxnet](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/syntaxnet/README.md): neural models of natural language syntax. * [tcn](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/tcn/README.md): Self-supervised representation learning from multi-view video. * [textsum](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/textsum/README.md): sequence-to-sequence with attention model for text summarization. * [transformer](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/transformer/README.md): spatial transformer network, which allows the spatial manipulation of data within the network. * [video\_prediction](https://github.com/deepdrive/deepdrive/tree/c64371bd9b652dbaaa84f868a7e2ed025dce2cd9/vendor/tensorflow/models/research/video_prediction/README.md): predicting future video frames with neural advection.