# Search

curl -X POST "https://search.dria.co/hnsw/search" \
-H "x-api-key: <YOUR_API_KEY>" \
-H "Content-Type: application/json" \
-d '{"rerank": true, "top_n": 10, "contract_id": "ZfIoBp3reUHM8Ndcg3-HZ3RkRzfPXxVpt-SWsxzJMX8", "query": "What is alexanDRIA library?"}'
        
# Query

curl -X POST "https://search.dria.co/hnsw/query" \
-H "x-api-key: <YOUR_API_KEY>" \
-H "Content-Type: application/json" \
-d '{"vector": [0.123, 0.5236], "top_n": 10, "contract_id": "ZfIoBp3reUHM8Ndcg3-HZ3RkRzfPXxVpt-SWsxzJMX8", "level": 2}'
        

Bash

# Search

response = requests.post(
    "https://search.dria.co/hnsw/search",
    headers={'x-api-key': '<YOUR_API_KEY>', 'Content-Type': 'application/json'},
    json={
    "rerank": True,
    "top_n": 10,
    "contract_id": "ZfIoBp3reUHM8Ndcg3-HZ3RkRzfPXxVpt-SWsxzJMX8",
    "query": "What is alexanDRIA library?"
}
)
print(response.status_code)
print(response.json())
        
# Query

response = requests.post(
    "https://search.dria.co/hnsw/query",
    headers={'x-api-key': '<YOUR_API_KEY>', 'Content-Type': 'application/json'},
    json={
    "vector": [
        0.123,
        0.5236
    ],
    "top_n": 10,
    "contract_id": "ZfIoBp3reUHM8Ndcg3-HZ3RkRzfPXxVpt-SWsxzJMX8",
    "level": 2
}
)
print(response.status_code)
print(response.json())
        

Python

Presentations from 2 of his talks, at Sobornne and MIT

Computer Perception With Deep Learning - Yann LeCun

Scene Parsing/Labeling No post-processing Frame-by-frame ConvNet runs at 50ms/frame on Virtex-6 FPGA hardware But communicating the features over ethernet limits system performance Y LeCun Scene Parsing/Labeling: Temporal Consistency Causal method for temporal consistency [Couprie, Farabet, Najman, LeCun ICLR 2013, ICIP 2013] Y LeCun Temporal Consistency Spatio-Temporal Super-Pixel segmentation [Couprie et al ICIP 2013] [Couprie et al JMLR under review] Majority vote over super-pixels Y LeCun Y LeCun NYU RGB-Depth Indoor Scenes Dataset 407024 RGB-D images of apartments [Silberman et al. 2012] 1449 labeled frames, 894 object categories NYU RGB-D Dataset Captured with a Kinect on a steadycam Y LeCun Results Y LeCun Results Depth helps a bit Helps a lot for floor and props

Helps surprisingly little for structures, and hurts for furniture [C. Cadena, J. Kosecka Semantic Parsing for Priming Object Detection in RGB-D Scenes Semantic Perception Mapping and Exploration (SPME), Karlsruhe 2013] Y LeCun Architecture for indoor RGB-D Semantic Segmentation Similar to outdoors semantic segmentation method Convnet with 4 input channels Vote over superpixels Y LeCun Scene Parsing/Labeling on RGB+Depth Images With temporal consistency [Couprie, Farabet, Najman, LeCun ICLR 2013, ICIP 2013] Y LeCun Scene Parsing/Labeling on RGB+Depth Images With temporal consistency [Couprie, Farabet, Najman, LeCun ICLR 2013, ICIP 2013] Y LeCun Labeling Videos Temporal consistency [Couprie, Farabet, Najman, LeCun ICLR 2013] [Couprie, Farabet, Najman, LeCun ICIP 2013] [Couprie, Farabet, Najman, LeCun submitted to JMLR] Y LeCun Semantic Segmentation on RGB+D Images and Videos [Couprie, Farabet, Najman, LeCun ICLR 2013, ICIP 2013] Y LeCun

ConvNet Scene Labeling For Vision-Based Navigation Of Off-Road Robots Y LeCun LAGR project: vision-based navigation for off-road robot Getting a robot to drive autonomously in unknown terrain solely from vision (camera input). Our team (NYU/Net-Scale Technologies Inc.) was one of 8 participants funded by DARPA All teams received identical robots and can only modify the software (not the hardware) The robot is given the GPS coordinates of a goal, and must drive to the goal as fast as possible. The terrain is unknown in advance. The robot is run 3 times through the same course. Long-Range Obstacle Detection with on-line, self-trained ConvNet Uses temporal consistency! Y LeCun Obstacle Detection at Short Range: Stereovision Obstacles overlaid with camera image Camera image Detected obstacles (red) Y LeCun But Stereovision Doesn't work at long range Stereo is only good up to about 10 meters. But not seeing past 10 meters is like driving in a fog or a snowstorm! Y LeCun

Long Range Vision with a Convolutional Net Pre-processing (125 ms) Ground plane estimation Horizon leveling Conversion to YUV + local contrast normalization Scale invariant pyramid of distance-normalized image bands Y LeCun Convolutional Net Architecture 100 features per 3x12x25 input window 10025x121 . . . `` CONVOLUTIONS (6x5) YUV image band 20-36 pixels tall, 36-500 pixels wide 2030x125 . . . MAX SUBSAMPLING (1x4) . . . 2030x484 CONVOLUTIONS (7x6) 336x484 YUV input Y LeCun Scene Labeling with ConvNet + online learning Image Labeling for Off-Road Robots [Hadsell JFR 2008] ConvNet labels pixels as one of 3 categories Traversible/flat (green), non traversible (red), foot of obstacle (purple) Labels obtained from stereo vision and SLAM Input image Input image Stereo Labels Stereo Labels Classifier Output Classifier Output Input image Input image Stereo Labels Stereo Labels Classifier Output Classifier Output Y LeCun Input image Input image