initial commit

README.md

@@ -1,3 +1,31 @@
+# Handcrafted solution example for the S23DR competition
+
+This repo provides an example of a simple algorithm to reconstruct wireframe and submit to S23DR competition.
+
+
+The repo consistst of the following parts:
+
+- `script.py` - the main file, which is run by the competition space. It should produce `submission.parquet` as the result of the run.
+- `hoho.py` - the file for parsing the dataset at the inference time. Do NOT change it.
+- `handcrafted_solution.py` - contains the actual implementation of the algorithm
+- other `*.py` files - helper i/o and visualization utilities
+- `packages/` - the directory to put python wheels for the custom packages you want to install and use. 
+
+## Solution description
+
+The solution is is simple. 
+
+1. Using provided (but noisy) semantic segmentation called `gestalt`, it taks the centroids of the vertex classes - `apex` and `eave_end_point` and projects them to 3D using provided (also noisy) monocular depth. 
+2. The vertices are connected using the same segmentation, by checking for edges classes to be present - `['eave', 'ridge', 'rake', 'valley']`. 
+3. All the "per-image" vertex predictions are merged in 3D space if their distance is less than threshold.
+4. All vertices, which have zero connections, are removed. 
+
+
+## Example on the training set
+
+See in [notebooks/example_on_training.ipynb](notebooks/example_on_training.ipynb)
+
 ---
 license: apache-2.0
 ---
+

color_mappings.py

@@ -0,0 +1,182 @@
+gestalt_color_mapping = {
+    "unclassified": (215, 62, 138),
+    "apex": (235, 88, 48),
+    "eave_end_point": (248, 130, 228),
+    "flashing_end_point": (71, 11, 161),
+    "ridge": (214, 251, 248),
+    "rake": (13, 94, 47),
+    "eave": (54, 243, 63),
+    "post": (187, 123, 236),
+    "ground_line": (136, 206, 14),
+    "flashing": (162, 162, 32),
+    "step_flashing": (169, 255, 219),
+    "hip": (8, 89, 52),
+    "valley": (85, 27, 65),
+    "roof": (215, 232, 179),
+    "door": (110, 52, 23),
+    "garage": (50, 233, 171),
+    "window": (230, 249, 40),
+    "shutter": (122, 4, 233),
+    "fascia": (95, 230, 240),
+    "soffit": (2, 102, 197),
+    "horizontal_siding": (131, 88, 59),
+    "vertical_siding": (110, 187, 198),
+    "brick": (171, 252, 7),
+    "concrete": (32, 47, 246),
+    "other_wall": (112, 61, 240),
+    "trim": (151, 206, 58),
+    "unknown": (127, 127, 127),
+}
+
+ade20k_color_mapping = {
+    'wall': (120, 120, 120),
+    'building;edifice': (180, 120, 120),
+    'sky': (6, 230, 230),
+    'floor;flooring': (80, 50, 50),
+    'tree': (4, 200, 3),
+    'ceiling': (120, 120, 80),
+    'road;route': (140, 140, 140),
+    'bed': (204, 5, 255),
+    'windowpane;window': (230, 230, 230),
+    'grass': (4, 250, 7),
+    'cabinet': (224, 5, 255),
+    'sidewalk;pavement': (235, 255, 7),
+    'person;individual;someone;somebody;mortal;soul': (150, 5, 61),
+    'earth;ground': (120, 120, 70),
+    'door;double;door': (8, 255, 51),
+    'table': (255, 6, 82),
+    'mountain;mount': (143, 255, 140),
+    'plant;flora;plant;life': (204, 255, 4),
+    'curtain;drape;drapery;mantle;pall': (255, 51, 7),
+    'chair': (204, 70, 3),
+    'car;auto;automobile;machine;motorcar': (0, 102, 200),
+    'water': (61, 230, 250),
+    'painting;picture': (255, 6, 51),
+    'sofa;couch;lounge': (11, 102, 255),
+    'shelf': (255, 7, 71),
+    'house': (255, 9, 224),
+    'sea': (9, 7, 230),
+    'mirror': (220, 220, 220),
+    'rug;carpet;carpeting': (255, 9, 92),
+    'field': (112, 9, 255),
+    'armchair': (8, 255, 214),
+    'seat': (7, 255, 224),
+    'fence;fencing': (255, 184, 6),
+    'desk': (10, 255, 71),
+    'rock;stone': (255, 41, 10),
+    'wardrobe;closet;press': (7, 255, 255),
+    'lamp': (224, 255, 8),
+    'bathtub;bathing;tub;bath;tub': (102, 8, 255),
+    'railing;rail': (255, 61, 6),
+    'cushion': (255, 194, 7),
+    'base;pedestal;stand': (255, 122, 8),
+    'box': (0, 255, 20),
+    'column;pillar': (255, 8, 41),
+    'signboard;sign': (255, 5, 153),
+    'chest;of;drawers;chest;bureau;dresser': (6, 51, 255),
+    'counter': (235, 12, 255),
+    'sand': (160, 150, 20),
+    'sink': (0, 163, 255),
+    'skyscraper': (140, 140, 140),
+    'fireplace;hearth;open;fireplace': (250, 10, 15),
+    'refrigerator;icebox': (20, 255, 0),
+    'grandstand;covered;stand': (31, 255, 0),
+    'path': (255, 31, 0),
+    'stairs;steps': (255, 224, 0),
+    'runway': (153, 255, 0),
+    'case;display;case;showcase;vitrine': (0, 0, 255),
+    'pool;table;billiard;table;snooker;table': (255, 71, 0),
+    'pillow': (0, 235, 255),
+    'screen;door;screen': (0, 173, 255),
+    'stairway;staircase': (31, 0, 255),
+    'river': (11, 200, 200),
+    'bridge;span': (255 ,82, 0),
+    'bookcase': (0, 255, 245),
+    'blind;screen': (0, 61, 255),
+    'coffee;table;cocktail;table': (0, 255, 112),
+    'toilet;can;commode;crapper;pot;potty;stool;throne': (0, 255, 133),
+    'flower': (255, 0, 0),
+    'book': (255, 163, 0),
+    'hill': (255, 102, 0),
+    'bench': (194, 255, 0),
+    'countertop': (0, 143, 255),
+    'stove;kitchen;stove;range;kitchen;range;cooking;stove': (51, 255, 0),
+    'palm;palm;tree': (0, 82, 255),
+    'kitchen;island': (0, 255, 41),
+    'computer;computing;machine;computing;device;data;processor;electronic;computer;information;processing;system': (0, 255, 173),
+    'swivel;chair': (10, 0, 255),
+    'boat': (173, 255, 0),
+    'bar': (0, 255, 153),
+    'arcade;machine': (255, 92, 0),
+    'hovel;hut;hutch;shack;shanty': (255, 0, 255),
+    'bus;autobus;coach;charabanc;double-decker;jitney;motorbus;motorcoach;omnibus;passenger;vehicle': (255, 0, 245),
+    'towel': (255, 0, 102),
+    'light;light;source': (255, 173, 0),
+    'truck;motortruck': (255, 0, 20),
+    'tower': (255, 184, 184),
+    'chandelier;pendant;pendent': (0, 31, 255),
+    'awning;sunshade;sunblind': (0, 255, 61),
+    'streetlight;street;lamp': (0, 71, 255),
+    'booth;cubicle;stall;kiosk': (255, 0, 204),
+    'television;television;receiver;television;set;tv;tv;set;idiot;box;boob;tube;telly;goggle;box': (0, 255, 194),
+    'airplane;aeroplane;plane': (0, 255, 82),
+    'dirt;track': (0, 10, 255),
+    'apparel;wearing;apparel;dress;clothes': (0, 112, 255),
+    'pole': (51, 0, 255),
+    'land;ground;soil': (0, 194, 255),
+    'bannister;banister;balustrade;balusters;handrail': (0, 122, 255),
+    'escalator;moving;staircase;moving;stairway': (0, 255, 163),
+    'ottoman;pouf;pouffe;puff;hassock': (255, 153, 0),
+    'bottle': (0, 255, 10),
+    'buffet;counter;sideboard': (255, 112, 0),
+    'poster;posting;placard;notice;bill;card': (143, 255, 0),
+    'stage': (82, 0, 255),
+    'van': (163, 255, 0),
+    'ship': (255, 235, 0),
+    'fountain': (8, 184, 170),
+    'conveyer;belt;conveyor;belt;conveyer;conveyor;transporter': (133, 0, 255),
+    'canopy': (0, 255, 92),
+    'washer;automatic;washer;washing;machine': (184, 0, 255),
+    'plaything;toy': (255, 0, 31),
+    'swimming;pool;swimming;bath;natatorium': (0, 184, 255),
+    'stool': (0, 214, 255),
+    'barrel;cask': (255, 0, 112),
+    'basket;handbasket': (92, 255, 0),
+    'waterfall;falls': (0, 224, 255),
+    'tent;collapsible;shelter': (112, 224, 255),
+    'bag': (70, 184, 160),
+    'minibike;motorbike': (163, 0, 255),
+    'cradle': (153, 0, 255),
+    'oven': (71, 255, 0),
+    'ball': (255, 0, 163),
+    'food;solid;food': (255, 204, 0),
+    'step;stair': (255, 0, 143),
+    'tank;storage;tank': (0, 255, 235),
+    'trade;name;brand;name;brand;marque': (133, 255, 0),
+    'microwave;microwave;oven': (255, 0, 235),
+    'pot;flowerpot': (245, 0, 255),
+    'animal;animate;being;beast;brute;creature;fauna': (255, 0, 122),
+    'bicycle;bike;wheel;cycle': (255, 245, 0),
+    'lake': (10, 190, 212),
+    'dishwasher;dish;washer;dishwashing;machine': (214, 255, 0),
+    'screen;silver;screen;projection;screen': (0, 204, 255),
+    'blanket;cover': (20, 0, 255),
+    'sculpture': (255, 255, 0),
+    'hood;exhaust;hood': (0, 153, 255),
+    'sconce': (0, 41, 255),
+    'vase': (0, 255, 204),
+    'traffic;light;traffic;signal;stoplight': (41, 0, 255),
+    'tray': (41, 255, 0),
+    'ashcan;trash;can;garbage;can;wastebin;ash;bin;ash-bin;ashbin;dustbin;trash;barrel;trash;bin': (173, 0, 255),
+    'fan': (0, 245, 255),
+    'pier;wharf;wharfage;dock': (71, 0, 255),
+    'crt;screen': (122, 0, 255),
+    'plate': (0, 255, 184),
+    'monitor;monitoring;device': (0, 92, 255),
+    'bulletin;board;notice;board': (184, 255, 0),
+    'shower': (0, 133, 255),
+    'radiator': (255, 214, 0),
+    'glass;drinking;glass': (25, 194, 194),
+    'clock': (102, 255, 0),
+    'flag': (92, 0, 255),
+}

handcrafted_solution.py

@@ -0,0 +1,242 @@
+# Description: This file contains the handcrafted solution for the task of wireframe reconstruction 
+
+import io
+from read_write_colmap import read_cameras_binary, read_images_binary, read_points3D_binary
+from PIL import Image as PImage
+import numpy as np
+from color_mappings import gestalt_color_mapping, ade20k_color_mapping
+from collections import defaultdict
+import cv2
+from typing import Tuple, List
+from scipy.spatial.distance import cdist
+
+
+def empty_solution():
+    '''Return a minimal valid solution, i.e. 2 vertices and 1 edge.'''
+    return np.zeros((2,3)), [(0, 1)], [0]
+    
+
+def convert_entry_to_human_readable(entry):
+    out = {}
+    already_good = ['__key__', 'wf_vertices', 'wf_edges', 'edge_semantics', 'mesh_vertices', 'mesh_faces', 'face_semantics', 'K', 'R', 't']
+    for k, v in entry.items():
+        if k in already_good:
+            out[k] = v
+            continue
+        if k == 'points3d':
+            out[k] = read_points3D_binary(fid=io.BytesIO(v))
+        if k == 'cameras':
+            out[k] = read_cameras_binary(fid=io.BytesIO(v))
+        if k == 'images':
+            out[k] = read_images_binary(fid=io.BytesIO(v))
+        if k in ['ade20k', 'gestalt']:
+            out[k] =  [PImage.open(io.BytesIO(x)).convert('RGB') for x in v]
+        if k == 'depthcm':
+            out[k] = [PImage.open(io.BytesIO(x)) for x in entry['depthcm']]
+    return out
+
+
+def get_vertices_and_edges_from_segmentation(gest_seg_np, edge_th = 50.0):
+    '''Get the vertices and edges from the gestalt segmentation mask of the house'''
+    vertices = []
+    connections = []
+    # Apex
+    apex_color = np.array(gestalt_color_mapping['apex'])
+    apex_mask = cv2.inRange(gest_seg_np,  apex_color-0.5, apex_color+0.5)
+    if apex_mask.sum() > 0:
+        output = cv2.connectedComponentsWithStats(apex_mask, 8, cv2.CV_32S)
+        (numLabels, labels, stats, centroids) = output
+        stats, centroids = stats[1:], centroids[1:]
+        
+        for i in range(numLabels-1):
+            vert = {"xy": centroids[i], "type": "apex"}
+            vertices.append(vert)
+    
+    eave_end_color = np.array(gestalt_color_mapping['eave_end_point'])
+    eave_end_mask = cv2.inRange(gest_seg_np,  eave_end_color-0.5, eave_end_color+0.5)
+    if eave_end_mask.sum() > 0:
+        output = cv2.connectedComponentsWithStats(eave_end_mask, 8, cv2.CV_32S)
+        (numLabels, labels, stats, centroids) = output
+        stats, centroids = stats[1:], centroids[1:]
+        
+        for i in range(numLabels-1):
+            vert = {"xy": centroids[i], "type": "eave_end_point"}
+            vertices.append(vert) 
+    # Connectivity
+    apex_pts = []
+    apex_pts_idxs = []
+    for j, v in enumerate(vertices):
+        apex_pts.append(v['xy'])
+        apex_pts_idxs.append(j)
+    apex_pts = np.array(apex_pts)
+            
+    # Ridge connects two apex points
+    for edge_class in ['eave', 'ridge', 'rake', 'valley']:
+        edge_color = np.array(gestalt_color_mapping[edge_class])
+        mask = cv2.morphologyEx(cv2.inRange(gest_seg_np,
+                                            edge_color-0.5,
+                                            edge_color+0.5),
+                                cv2.MORPH_DILATE, np.ones((11, 11)))
+        line_img = np.copy(gest_seg_np) * 0
+        if mask.sum() > 0:
+            output = cv2.connectedComponentsWithStats(mask, 8, cv2.CV_32S)
+            (numLabels, labels, stats, centroids) = output
+            stats, centroids = stats[1:], centroids[1:]
+            edges = []
+            for i in range(1, numLabels):
+                y,x = np.where(labels == i)
+                xleft_idx = np.argmin(x)
+                x_left = x[xleft_idx]
+                y_left = y[xleft_idx]
+                xright_idx = np.argmax(x)
+                x_right = x[xright_idx]
+                y_right = y[xright_idx]
+                edges.append((x_left, y_left, x_right, y_right))
+                cv2.line(line_img, (x_left, y_left), (x_right, y_right), (255, 255, 255), 2)
+            edges = np.array(edges)
+            if (len(apex_pts) < 2) or len(edges) <1:
+                continue
+            pts_to_edges_dist = np.minimum(cdist(apex_pts, edges[:,:2]), cdist(apex_pts, edges[:,2:]))
+            connectivity_mask = pts_to_edges_dist <= edge_th
+            edge_connects = connectivity_mask.sum(axis=0)
+            for edge_idx, edgesum in enumerate(edge_connects):
+                if edgesum>=2:
+                    connected_verts = np.where(connectivity_mask[:,edge_idx])[0]
+                    for a_i, a in enumerate(connected_verts):
+                        for b in connected_verts[a_i+1:]:
+                            connections.append((a, b))
+    return vertices, connections
+
+def get_uv_depth(vertices, depth):
+    '''Get the depth of the vertices from the depth image'''
+    uv = []
+    for v in vertices:
+        uv.append(v['xy'])
+    uv = np.array(uv)
+    uv_int = uv.astype(np.int32)
+    H, W = depth.shape[:2]
+    uv_int[:, 0] = np.clip( uv_int[:, 0], 0, W-1)
+    uv_int[:, 1] = np.clip( uv_int[:, 1], 0, H-1)
+    vertex_depth = depth[(uv_int[:, 1] , uv_int[:, 0])]
+    return uv, vertex_depth
+
+
+def merge_vertices_3d(vert_edge_per_image, th=0.1):
+    '''Merge vertices that are close to each other in 3D space and are of same types'''
+    all_3d_vertices = []
+    connections_3d = []
+    all_indexes = []
+    cur_start = 0
+    types = []
+    for cimg_idx, (vertices, connections, vertices_3d) in vert_edge_per_image.items():
+        types += [int(v['type']=='apex') for v in vertices]
+        all_3d_vertices.append(vertices_3d)
+        connections_3d+=[(x+cur_start,y+cur_start) for (x,y) in connections]
+        cur_start+=len(vertices_3d)
+    all_3d_vertices = np.concatenate(all_3d_vertices, axis=0)
+    #print (connections_3d)
+    distmat = cdist(all_3d_vertices, all_3d_vertices)
+    types = np.array(types).reshape(-1,1)
+    same_types = cdist(types, types)
+    mask_to_merge = (distmat <= th) & (same_types==0)
+    new_vertices = []
+    new_connections = []
+    to_merge = sorted(list(set([tuple(a.nonzero()[0].tolist()) for a in mask_to_merge])))
+    to_merge_final = defaultdict(list)
+    for i in range(len(all_3d_vertices)):
+        for j in to_merge:
+            if i in j:
+                to_merge_final[i]+=j
+    for k, v in to_merge_final.items():
+        to_merge_final[k] = list(set(v))
+    already_there = set() 
+    merged = []
+    for k, v in to_merge_final.items():
+        if k in already_there:
+            continue
+        merged.append(v)
+        for vv in v:
+            already_there.add(vv)
+    old_idx_to_new = {}
+    count=0
+    for idxs in merged:
+        new_vertices.append(all_3d_vertices[idxs].mean(axis=0))
+        for idx in idxs:
+            old_idx_to_new[idx] = count
+        count +=1
+    #print (connections_3d)
+    new_vertices=np.array(new_vertices)
+    #print (connections_3d)
+    for conn in connections_3d:
+        new_con = sorted((old_idx_to_new[conn[0]], old_idx_to_new[conn[1]]))
+        if new_con[0] == new_con[1]:
+            continue
+        if new_con not in new_connections:
+            new_connections.append(new_con)
+    #print (f'{len(new_vertices)} left after merging {len(all_3d_vertices)} with {th=}')
+    return new_vertices, new_connections
+
+def prune_not_connected(all_3d_vertices, connections_3d):
+    '''Prune vertices that are not connected to any other vertex'''
+    connected = defaultdict(list)
+    for c in connections_3d:
+        connected[c[0]].append(c)
+        connected[c[1]].append(c)
+    new_indexes = {}
+    new_verts = []
+    connected_out = []
+    for k,v in connected.items():
+        vert = all_3d_vertices[k]
+        if tuple(vert) not in new_verts:
+            new_verts.append(tuple(vert))
+            new_indexes[k]=len(new_verts) -1
+    for k,v in connected.items():
+        for vv in v:
+            connected_out.append((new_indexes[vv[0]],new_indexes[vv[1]]))
+    connected_out=list(set(connected_out))                   
+    
+    return np.array(new_verts), connected_out
+
+
+def predict(entry, visualize=False) -> Tuple[np.ndarray, List[int]]:
+    good_entry = convert_entry_to_human_readable(entry)
+    vert_edge_per_image = {}
+    for i, (gest, depth, K, R, t) in enumerate(zip(good_entry['gestalt'],
+                                                good_entry['depthcm'], 
+                                                good_entry['K'],
+                                                good_entry['R'],
+                                                good_entry['t'] 
+                                                )):
+        gest_seg = gest.resize(depth.size)
+        gest_seg_np = np.array(gest_seg).astype(np.uint8)
+        # Metric3D
+        depth_np = np.array(depth) / 2.5 # 2.5 is the scale estimation coefficient
+        vertices, connections = get_vertices_and_edges_from_segmentation(gest_seg_np, edge_th = 20.)
+        if (len(vertices) < 2) or (len(connections) < 1):
+            print (f'Not enough vertices or connections in image {i}')
+            vert_edge_per_image[i] = np.empty((0, 2)), [], np.empty((0, 3))
+            continue
+        uv, depth_vert = get_uv_depth(vertices, depth_np)
+        # Normalize the uv to the camera intrinsics
+        xy_local = np.ones((len(uv), 3))
+        xy_local[:, 0] = (uv[:, 0] - K[0,2]) / K[0,0]
+        xy_local[:, 1] = (uv[:, 1] - K[1,2]) / K[1,1]
+        # Get the 3D vertices
+        vertices_3d_local = depth_vert[...,None] * (xy_local/np.linalg.norm(xy_local, axis=1)[...,None])
+        world_to_cam = np.eye(4)
+        world_to_cam[:3, :3] = R
+        world_to_cam[:3, 3] = t.reshape(-1)
+        cam_to_world =  np.linalg.inv(world_to_cam)
+        vertices_3d = cv2.transform(cv2.convertPointsToHomogeneous(vertices_3d_local), cam_to_world)
+        vertices_3d = cv2.convertPointsFromHomogeneous(vertices_3d).reshape(-1, 3)
+        vert_edge_per_image[i] = vertices, connections, vertices_3d
+    all_3d_vertices, connections_3d = merge_vertices_3d(vert_edge_per_image, 3.0)
+    all_3d_vertices_clean, connections_3d_clean  = prune_not_connected(all_3d_vertices, connections_3d)
+    if (len(all_3d_vertices_clean) < 2) or len(connections_3d_clean) < 1:
+        print (f'Not enough vertices or connections in the 3D vertices')
+        return empty_solution()
+    if visualize:
+        from viz3d import plot_estimate_and_gt
+        plot_estimate_and_gt(all_3d_vertices_clean, connections_3d_clean, good_entry['wf_vertices'],
+                            good_entry['wf_edges'])
+    return  all_3d_vertices_clean, connections_3d_clean, [0 for i in range(len(connections_3d_clean))]

hoho.py

@@ -0,0 +1,261 @@
+import os 
+import json
+import shutil
+from pathlib import Path
+from typing import Dict
+
+from PIL import ImageFile
+ImageFile.LOAD_TRUNCATED_IMAGES = True
+
+LOCAL_DATADIR = None
+
+def setup(local_dir='./data/usm-training-data/data'):
+    
+    # If we are in the test environment, we need to link the data directory to the correct location
+    tmp_datadir = Path('/tmp/data/data')
+    local_test_datadir = Path('./data/usm-test-data-x/data')
+    local_val_datadir = Path(local_dir)
+    
+    os.system('pwd')
+    os.system('ls -lahtr .')
+    
+    if tmp_datadir.exists() and not local_test_datadir.exists():
+        global LOCAL_DATADIR
+        LOCAL_DATADIR = local_test_datadir
+        # shutil.move(datadir, './usm-test-data-x/data')
+        print(f"Linking {tmp_datadir} to {LOCAL_DATADIR} (we are in the test environment)")
+        LOCAL_DATADIR.parent.mkdir(parents=True, exist_ok=True)
+        LOCAL_DATADIR.symlink_to(tmp_datadir)
+    else:
+        LOCAL_DATADIR = local_val_datadir
+        print(f"Using {LOCAL_DATADIR} as the data directory (we are running locally)")
+        
+    # os.system("ls -lahtr")
+    
+    assert LOCAL_DATADIR.exists(), f"Data directory {LOCAL_DATADIR} does not exist"
+    return LOCAL_DATADIR
+    
+    
+    
+    
+import importlib
+from pathlib import Path
+import subprocess
+
+def download_package(package_name, path_to_save='packages'):
+    """
+    Downloads a package using pip and saves it to a specified directory.
+
+    Parameters:
+    package_name (str): The name of the package to download.
+    path_to_save (str): The path to the directory where the package will be saved.
+    """
+    try:
+        # pip download webdataset -d packages/webdataset --platform manylinux1_x86_64 --python-version 38 --only-binary=:all:
+        subprocess.check_call([subprocess.sys.executable, "-m", "pip", "download", package_name, 
+                               "-d", str(Path(path_to_save)/package_name),  # Download the package to the specified directory
+                               "--platform", "manylinux1_x86_64",  # Specify the platform
+                               "--python-version", "38",  # Specify the Python version
+                               "--only-binary=:all:"])  # Download only binary packages
+        print(f'Package "{package_name}" downloaded successfully')
+    except subprocess.CalledProcessError as e:
+        print(f'Failed to downloaded package "{package_name}". Error: {e}')
+              
+              
+def install_package_from_local_file(package_name, folder='packages'):
+    """
+    Installs a package from a local .whl file or a directory containing .whl files using pip.
+
+    Parameters:
+    path_to_file_or_directory (str): The path to the .whl file or the directory containing .whl files.
+    """
+    try:
+        pth = str(Path(folder) / package_name)
+        subprocess.check_call([subprocess.sys.executable, "-m", "pip", "install", 
+                               "--no-index",  # Do not use package index
+                               "--find-links", pth,  # Look for packages in the specified directory or at the file
+                               package_name])  # Specify the package to install
+        print(f"Package installed successfully from {pth}")
+    except subprocess.CalledProcessError as e:
+        print(f"Failed to install package from {pth}. Error: {e}")
+        
+        
+def importt(module_name, as_name=None):
+    """
+    Imports a module and returns it.
+
+    Parameters:
+    module_name (str): The name of the module to import.
+    as_name (str): The name to use for the imported module. If None, the original module name will be used.
+
+    Returns:
+    The imported module.
+    """
+    for _ in range(2):
+        try:
+            if as_name is None:
+                print(f'imported {module_name}')
+                return importlib.import_module(module_name)
+            else:
+                print(f'imported {module_name} as {as_name}')
+                return importlib.import_module(module_name, as_name)
+        except ModuleNotFoundError as e:
+            install_package_from_local_file(module_name)
+            print(f"Failed to import module {module_name}. Error: {e}")
+            
+    
+def prepare_submission():
+    # Download packages from requirements.txt 
+    if Path('requirements.txt').exists():
+        print('downloading packages from requirements.txt')
+        Path('packages').mkdir(exist_ok=True)
+        with open('requirements.txt') as f:
+            packages = f.readlines()
+            for p in packages:
+                download_package(p.strip())
+                
+                
+    print('all packages downloaded. Don\'t foget to include the packages in the submission by adding them with git lfs.')
+        
+
+def Rt_to_eye_target(im, K, R, t):
+    height = im.height
+    focal_length = K[0,0]
+    fov = 2.0 * np.arctan2((0.5 * height), focal_length) / (np.pi / 180.0)
+
+    x_axis, y_axis, z_axis = R
+
+    eye = -(R.T @ t).squeeze()
+    z_axis = z_axis.squeeze()
+    target = eye + z_axis
+    up = -y_axis
+    
+    return eye, target, up, fov        
+
+
+########## general utilities ##########
+import contextlib
+import tempfile 
+from pathlib import Path
+
+@contextlib.contextmanager
+def working_directory(path):
+    """Changes working directory and returns to previous on exit."""
+    prev_cwd = Path.cwd()
+    os.chdir(path)
+    try:
+        yield
+    finally:
+        os.chdir(prev_cwd)
+        
+@contextlib.contextmanager
+def temp_working_directory():
+    with tempfile.TemporaryDirectory(dir='.') as D:
+        with working_directory(D):
+            yield
+
+
+############# Dataset ############# 
+def proc(row, split='train'):
+    # column_names_train = ['ade20k', 'depthcm', 'gestalt', 'colmap', 'KRt', 'mesh', 'wireframe']
+    # column_names_test = ['ade20k', 'depthcm', 'gestalt', 'colmap', 'KRt', 'wireframe']
+    # cols = column_names_train if split == 'train' else column_names_test
+    out = {}
+    for k, v in row.items():
+        colname = k.split('.')[0]
+        if colname in {'ade20k', 'depthcm', 'gestalt'}:
+            if colname in out:
+                out[colname].append(v)
+            else:
+                out[colname] = [v]
+        elif colname in {'wireframe', 'mesh'}:
+            # out.update({a: b.tolist() for a,b in v.items()})
+            out.update({a: b for a,b in v.items()})
+        elif colname in 'kr':
+            out[colname.upper()] = v
+        else:
+            out[colname] = v
+            
+    return Sample(out)
+
+
+class Sample(Dict):
+    def __repr__(self):
+        return str({k: v.shape if hasattr(v, 'shape') else [type(v[0])] if isinstance(v, list) else type(v) for k,v in self.items()})
+
+    
+        
+def get_params():
+    exmaple_param_dict = {
+        "competition_id": "usm3d/S23DR",
+        "competition_type": "script",
+        "metric": "custom",
+        "token": "hf_**********************************",
+        "team_id": "local-test-team_id",
+        "submission_id": "local-test-submission_id",
+        "submission_id_col": "__key__",
+        "submission_cols": [
+            "__key__",
+            "wf_edges",
+            "wf_vertices",
+            "edge_semantics"
+        ],
+        "submission_rows": 180,
+        "output_path": ".",
+        "submission_repo": "<THE HF MODEL ID of THIS REPO",
+        "time_limit": 7200,
+        "dataset": "usm3d/usm-test-data-x",
+        "submission_filenames": [
+            "submission.parquet"
+        ]
+    }
+    
+    param_path = Path('params.json')
+    
+    if not param_path.exists():
+        print('params.json not found (this means we probably aren\'t in the test env). Using example params.')
+        params = exmaple_param_dict
+    else:
+        print('found params.json (this means we are probably in the test env). Using params from file.')
+        with param_path.open() as f:
+            params = json.load(f)
+    print(params)
+    return params
+
+
+
+import webdataset as wds 
+import numpy as np
+
+def get_dataset(decode='pil', proc=proc, split='train', dataset_type='webdataset'):
+    if LOCAL_DATADIR is None:
+        raise ValueError('LOCAL_DATADIR is not set. Please run setup() first.')
+        
+    local_dir = Path(LOCAL_DATADIR)
+    if split != 'all':
+        local_dir = local_dir / split
+    
+    paths = [str(p) for p in local_dir.rglob('*.tar.gz')]
+    
+    dataset = wds.WebDataset(paths)
+    if decode is not None:
+        dataset = dataset.decode(decode)
+    else:
+        dataset = dataset.decode()
+    
+    dataset = dataset.map(proc)
+    
+    if dataset_type == 'webdataset':
+        return dataset
+    
+    if dataset_type == 'hf':
+        import datasets
+        from datasets import Features, Value, Sequence, Image, Array2D
+       
+        if split == 'train':
+            return datasets.IterableDataset.from_generator(lambda: dataset.iterator())
+        elif split == 'val':
+            return datasets.IterableDataset.from_generator(lambda: dataset.iterator())
+
+    
+    

notebooks/.ipynb_checkpoints/example-checkpoint.ipynb

@@ -0,0 +1,166 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "503c6bcb-aa46-46c6-8b86-566b0a470b43",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import sys\n",
+    "sys.path.append('..')\n",
+    "from handcrafted_solution import *\n",
+    "from viz3d import *\n",
+    "from read_write_colmap import *"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "8bdcd910-bac0-44be-8344-cb901ea2f369",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Collecting webdataset\n",
+      "  Downloading webdataset-0.2.86-py3-none-any.whl.metadata (29 kB)\n",
+      "Collecting braceexpand (from webdataset)\n",
+      "  Downloading braceexpand-0.1.7-py2.py3-none-any.whl.metadata (3.0 kB)\n",
+      "Requirement already satisfied: numpy in /Users/dmytromishkin/miniconda3/envs/pytorch/lib/python3.9/site-packages (from webdataset) (1.24.4)\n",
+      "Requirement already satisfied: pyyaml in /Users/dmytromishkin/miniconda3/envs/pytorch/lib/python3.9/site-packages (from webdataset) (6.0)\n",
+      "Downloading webdataset-0.2.86-py3-none-any.whl (70 kB)\n",
+      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m70.4/70.4 kB\u001b[0m \u001b[31m3.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
+      "\u001b[?25hDownloading braceexpand-0.1.7-py2.py3-none-any.whl (5.9 kB)\n",
+      "Installing collected packages: braceexpand, webdataset\n",
+      "Successfully installed braceexpand-0.1.7 webdataset-0.2.86\n"
+     ]
+    }
+   ],
+   "source": [
+    "!pip install webdataset"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "88f4fc8f-efa9-404b-9073-c7d4a73f9075",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import webdataset as wds \n",
+    "import numpy as np\n",
+    "import datasets\n",
+    "from datasets import Features, Value, Sequence, Image, Array2D\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "080f1a12-06bf-4b97-8a52-d7cf416adede",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "809ae1d7cc0e48718433b6896bb84067",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Resolving data files:   0%|          | 0/25 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "86f66f66049746eeb98c9a15972c2ca2",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Downloading data:   0%|          | 0.00/1.01G [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "ename": "KeyboardInterrupt",
+     "evalue": "",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
+      "Cell \u001b[0;32mIn[7], line 2\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mdatasets\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m load_dataset\n\u001b[0;32m----> 2\u001b[0m ds \u001b[38;5;241m=\u001b[39m \u001b[43mload_dataset\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mtest-org-usm3d/usm-training-data\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/datasets/load.py:2574\u001b[0m, in \u001b[0;36mload_dataset\u001b[0;34m(path, name, data_dir, data_files, split, cache_dir, features, download_config, download_mode, verification_mode, ignore_verifications, keep_in_memory, save_infos, revision, token, use_auth_token, task, streaming, num_proc, storage_options, trust_remote_code, **config_kwargs)\u001b[0m\n\u001b[1;32m   2571\u001b[0m try_from_hf_gcs \u001b[38;5;241m=\u001b[39m path \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;129;01min\u001b[39;00m _PACKAGED_DATASETS_MODULES\n\u001b[1;32m   2573\u001b[0m \u001b[38;5;66;03m# Download and prepare data\u001b[39;00m\n\u001b[0;32m-> 2574\u001b[0m \u001b[43mbuilder_instance\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mdownload_and_prepare\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   2575\u001b[0m \u001b[43m    \u001b[49m\u001b[43mdownload_config\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdownload_config\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   2576\u001b[0m \u001b[43m    \u001b[49m\u001b[43mdownload_mode\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdownload_mode\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   2577\u001b[0m \u001b[43m    \u001b[49m\u001b[43mverification_mode\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mverification_mode\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   2578\u001b[0m \u001b[43m    \u001b[49m\u001b[43mtry_from_hf_gcs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mtry_from_hf_gcs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   2579\u001b[0m \u001b[43m    \u001b[49m\u001b[43mnum_proc\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mnum_proc\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   2580\u001b[0m \u001b[43m    \u001b[49m\u001b[43mstorage_options\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mstorage_options\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   2581\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   2583\u001b[0m \u001b[38;5;66;03m# Build dataset for splits\u001b[39;00m\n\u001b[1;32m   2584\u001b[0m keep_in_memory \u001b[38;5;241m=\u001b[39m (\n\u001b[1;32m   2585\u001b[0m     keep_in_memory \u001b[38;5;28;01mif\u001b[39;00m keep_in_memory \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;28;01melse\u001b[39;00m is_small_dataset(builder_instance\u001b[38;5;241m.\u001b[39minfo\u001b[38;5;241m.\u001b[39mdataset_size)\n\u001b[1;32m   2586\u001b[0m )\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/datasets/builder.py:1005\u001b[0m, in \u001b[0;36mDatasetBuilder.download_and_prepare\u001b[0;34m(self, output_dir, download_config, download_mode, verification_mode, ignore_verifications, try_from_hf_gcs, dl_manager, base_path, use_auth_token, file_format, max_shard_size, num_proc, storage_options, **download_and_prepare_kwargs)\u001b[0m\n\u001b[1;32m   1003\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m num_proc \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m   1004\u001b[0m         prepare_split_kwargs[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mnum_proc\u001b[39m\u001b[38;5;124m\"\u001b[39m] \u001b[38;5;241m=\u001b[39m num_proc\n\u001b[0;32m-> 1005\u001b[0m     \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_download_and_prepare\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   1006\u001b[0m \u001b[43m        \u001b[49m\u001b[43mdl_manager\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdl_manager\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1007\u001b[0m \u001b[43m        \u001b[49m\u001b[43mverification_mode\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mverification_mode\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1008\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mprepare_split_kwargs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1009\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mdownload_and_prepare_kwargs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1010\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1011\u001b[0m \u001b[38;5;66;03m# Sync info\u001b[39;00m\n\u001b[1;32m   1012\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39minfo\u001b[38;5;241m.\u001b[39mdataset_size \u001b[38;5;241m=\u001b[39m \u001b[38;5;28msum\u001b[39m(split\u001b[38;5;241m.\u001b[39mnum_bytes \u001b[38;5;28;01mfor\u001b[39;00m split \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39minfo\u001b[38;5;241m.\u001b[39msplits\u001b[38;5;241m.\u001b[39mvalues())\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/datasets/builder.py:1767\u001b[0m, in \u001b[0;36mGeneratorBasedBuilder._download_and_prepare\u001b[0;34m(self, dl_manager, verification_mode, **prepare_splits_kwargs)\u001b[0m\n\u001b[1;32m   1766\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m_download_and_prepare\u001b[39m(\u001b[38;5;28mself\u001b[39m, dl_manager, verification_mode, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mprepare_splits_kwargs):\n\u001b[0;32m-> 1767\u001b[0m     \u001b[38;5;28;43msuper\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_download_and_prepare\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   1768\u001b[0m \u001b[43m        \u001b[49m\u001b[43mdl_manager\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1769\u001b[0m \u001b[43m        \u001b[49m\u001b[43mverification_mode\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1770\u001b[0m \u001b[43m        \u001b[49m\u001b[43mcheck_duplicate_keys\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mverification_mode\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m==\u001b[39;49m\u001b[43m \u001b[49m\u001b[43mVerificationMode\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mBASIC_CHECKS\u001b[49m\n\u001b[1;32m   1771\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;129;43;01mor\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[43mverification_mode\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m==\u001b[39;49m\u001b[43m \u001b[49m\u001b[43mVerificationMode\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mALL_CHECKS\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1772\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mprepare_splits_kwargs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1773\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/datasets/builder.py:1078\u001b[0m, in \u001b[0;36mDatasetBuilder._download_and_prepare\u001b[0;34m(self, dl_manager, verification_mode, **prepare_split_kwargs)\u001b[0m\n\u001b[1;32m   1076\u001b[0m split_dict \u001b[38;5;241m=\u001b[39m SplitDict(dataset_name\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mdataset_name)\n\u001b[1;32m   1077\u001b[0m split_generators_kwargs \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_make_split_generators_kwargs(prepare_split_kwargs)\n\u001b[0;32m-> 1078\u001b[0m split_generators \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_split_generators\u001b[49m\u001b[43m(\u001b[49m\u001b[43mdl_manager\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43msplit_generators_kwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1080\u001b[0m \u001b[38;5;66;03m# Checksums verification\u001b[39;00m\n\u001b[1;32m   1081\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m verification_mode \u001b[38;5;241m==\u001b[39m VerificationMode\u001b[38;5;241m.\u001b[39mALL_CHECKS \u001b[38;5;129;01mand\u001b[39;00m dl_manager\u001b[38;5;241m.\u001b[39mrecord_checksums:\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/datasets/packaged_modules/webdataset/webdataset.py:47\u001b[0m, in \u001b[0;36mWebDataset._split_generators\u001b[0;34m(self, dl_manager)\u001b[0m\n\u001b[1;32m     45\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mconfig\u001b[38;5;241m.\u001b[39mdata_files:\n\u001b[1;32m     46\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mAt least one data file must be specified, but got data_files=\u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mconfig\u001b[38;5;241m.\u001b[39mdata_files\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m---> 47\u001b[0m data_files \u001b[38;5;241m=\u001b[39m \u001b[43mdl_manager\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mdownload\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mconfig\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mdata_files\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     48\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(data_files, (\u001b[38;5;28mstr\u001b[39m, \u001b[38;5;28mlist\u001b[39m, \u001b[38;5;28mtuple\u001b[39m)):\n\u001b[1;32m     49\u001b[0m     tar_paths \u001b[38;5;241m=\u001b[39m data_files\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/datasets/download/download_manager.py:434\u001b[0m, in \u001b[0;36mDownloadManager.download\u001b[0;34m(self, url_or_urls)\u001b[0m\n\u001b[1;32m    432\u001b[0m start_time \u001b[38;5;241m=\u001b[39m datetime\u001b[38;5;241m.\u001b[39mnow()\n\u001b[1;32m    433\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m stack_multiprocessing_download_progress_bars():\n\u001b[0;32m--> 434\u001b[0m     downloaded_path_or_paths \u001b[38;5;241m=\u001b[39m \u001b[43mmap_nested\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    435\u001b[0m \u001b[43m        \u001b[49m\u001b[43mdownload_func\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    436\u001b[0m \u001b[43m        \u001b[49m\u001b[43murl_or_urls\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    437\u001b[0m \u001b[43m        \u001b[49m\u001b[43mmap_tuple\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m,\u001b[49m\n\u001b[1;32m    438\u001b[0m \u001b[43m        \u001b[49m\u001b[43mnum_proc\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdownload_config\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mnum_proc\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    439\u001b[0m \u001b[43m        \u001b[49m\u001b[43mdesc\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mDownloading data files\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m,\u001b[49m\n\u001b[1;32m    440\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    441\u001b[0m duration \u001b[38;5;241m=\u001b[39m datetime\u001b[38;5;241m.\u001b[39mnow() \u001b[38;5;241m-\u001b[39m start_time\n\u001b[1;32m    442\u001b[0m logger\u001b[38;5;241m.\u001b[39minfo(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mDownloading took \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mduration\u001b[38;5;241m.\u001b[39mtotal_seconds()\u001b[38;5;250m \u001b[39m\u001b[38;5;241m/\u001b[39m\u001b[38;5;241m/\u001b[39m\u001b[38;5;250m \u001b[39m\u001b[38;5;241m60\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m min\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/datasets/utils/py_utils.py:466\u001b[0m, in \u001b[0;36mmap_nested\u001b[0;34m(function, data_struct, dict_only, map_list, map_tuple, map_numpy, num_proc, parallel_min_length, types, disable_tqdm, desc)\u001b[0m\n\u001b[1;32m    464\u001b[0m     num_proc \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m1\u001b[39m\n\u001b[1;32m    465\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28many\u001b[39m(\u001b[38;5;28misinstance\u001b[39m(v, types) \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;28mlen\u001b[39m(v) \u001b[38;5;241m>\u001b[39m \u001b[38;5;28mlen\u001b[39m(iterable) \u001b[38;5;28;01mfor\u001b[39;00m v \u001b[38;5;129;01min\u001b[39;00m iterable):\n\u001b[0;32m--> 466\u001b[0m     mapped \u001b[38;5;241m=\u001b[39m [\n\u001b[1;32m    467\u001b[0m         map_nested(\n\u001b[1;32m    468\u001b[0m             function\u001b[38;5;241m=\u001b[39mfunction,\n\u001b[1;32m    469\u001b[0m             data_struct\u001b[38;5;241m=\u001b[39mobj,\n\u001b[1;32m    470\u001b[0m             num_proc\u001b[38;5;241m=\u001b[39mnum_proc,\n\u001b[1;32m    471\u001b[0m             parallel_min_length\u001b[38;5;241m=\u001b[39mparallel_min_length,\n\u001b[1;32m    472\u001b[0m             types\u001b[38;5;241m=\u001b[39mtypes,\n\u001b[1;32m    473\u001b[0m         )\n\u001b[1;32m    474\u001b[0m         \u001b[38;5;28;01mfor\u001b[39;00m obj \u001b[38;5;129;01min\u001b[39;00m iterable\n\u001b[1;32m    475\u001b[0m     ]\n\u001b[1;32m    476\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m num_proc \u001b[38;5;241m!=\u001b[39m \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m1\u001b[39m \u001b[38;5;129;01mand\u001b[39;00m num_proc \u001b[38;5;241m<\u001b[39m\u001b[38;5;241m=\u001b[39m \u001b[38;5;241m1\u001b[39m \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mlen\u001b[39m(iterable) \u001b[38;5;241m<\u001b[39m parallel_min_length:\n\u001b[1;32m    477\u001b[0m     mapped \u001b[38;5;241m=\u001b[39m [\n\u001b[1;32m    478\u001b[0m         _single_map_nested((function, obj, types, \u001b[38;5;28;01mNone\u001b[39;00m, \u001b[38;5;28;01mTrue\u001b[39;00m, \u001b[38;5;28;01mNone\u001b[39;00m))\n\u001b[1;32m    479\u001b[0m         \u001b[38;5;28;01mfor\u001b[39;00m obj \u001b[38;5;129;01min\u001b[39;00m hf_tqdm(iterable, disable\u001b[38;5;241m=\u001b[39mdisable_tqdm, desc\u001b[38;5;241m=\u001b[39mdesc)\n\u001b[1;32m    480\u001b[0m     ]\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/datasets/utils/py_utils.py:467\u001b[0m, in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m    464\u001b[0m     num_proc \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m1\u001b[39m\n\u001b[1;32m    465\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28many\u001b[39m(\u001b[38;5;28misinstance\u001b[39m(v, types) \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;28mlen\u001b[39m(v) \u001b[38;5;241m>\u001b[39m \u001b[38;5;28mlen\u001b[39m(iterable) \u001b[38;5;28;01mfor\u001b[39;00m v \u001b[38;5;129;01min\u001b[39;00m iterable):\n\u001b[1;32m    466\u001b[0m     mapped \u001b[38;5;241m=\u001b[39m [\n\u001b[0;32m--> 467\u001b[0m         \u001b[43mmap_nested\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    468\u001b[0m \u001b[43m            \u001b[49m\u001b[43mfunction\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mfunction\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    469\u001b[0m \u001b[43m            \u001b[49m\u001b[43mdata_struct\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mobj\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    470\u001b[0m \u001b[43m            \u001b[49m\u001b[43mnum_proc\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mnum_proc\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    471\u001b[0m \u001b[43m            \u001b[49m\u001b[43mparallel_min_length\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mparallel_min_length\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    472\u001b[0m \u001b[43m            \u001b[49m\u001b[43mtypes\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mtypes\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    473\u001b[0m \u001b[43m        \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    474\u001b[0m         \u001b[38;5;28;01mfor\u001b[39;00m obj \u001b[38;5;129;01min\u001b[39;00m iterable\n\u001b[1;32m    475\u001b[0m     ]\n\u001b[1;32m    476\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m num_proc \u001b[38;5;241m!=\u001b[39m \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m1\u001b[39m \u001b[38;5;129;01mand\u001b[39;00m num_proc \u001b[38;5;241m<\u001b[39m\u001b[38;5;241m=\u001b[39m \u001b[38;5;241m1\u001b[39m \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mlen\u001b[39m(iterable) \u001b[38;5;241m<\u001b[39m parallel_min_length:\n\u001b[1;32m    477\u001b[0m     mapped \u001b[38;5;241m=\u001b[39m [\n\u001b[1;32m    478\u001b[0m         _single_map_nested((function, obj, types, \u001b[38;5;28;01mNone\u001b[39;00m, \u001b[38;5;28;01mTrue\u001b[39;00m, \u001b[38;5;28;01mNone\u001b[39;00m))\n\u001b[1;32m    479\u001b[0m         \u001b[38;5;28;01mfor\u001b[39;00m obj \u001b[38;5;129;01min\u001b[39;00m hf_tqdm(iterable, disable\u001b[38;5;241m=\u001b[39mdisable_tqdm, desc\u001b[38;5;241m=\u001b[39mdesc)\n\u001b[1;32m    480\u001b[0m     ]\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/datasets/utils/py_utils.py:477\u001b[0m, in \u001b[0;36mmap_nested\u001b[0;34m(function, data_struct, dict_only, map_list, map_tuple, map_numpy, num_proc, parallel_min_length, types, disable_tqdm, desc)\u001b[0m\n\u001b[1;32m    466\u001b[0m     mapped \u001b[38;5;241m=\u001b[39m [\n\u001b[1;32m    467\u001b[0m         map_nested(\n\u001b[1;32m    468\u001b[0m             function\u001b[38;5;241m=\u001b[39mfunction,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    474\u001b[0m         \u001b[38;5;28;01mfor\u001b[39;00m obj \u001b[38;5;129;01min\u001b[39;00m iterable\n\u001b[1;32m    475\u001b[0m     ]\n\u001b[1;32m    476\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m num_proc \u001b[38;5;241m!=\u001b[39m \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m1\u001b[39m \u001b[38;5;129;01mand\u001b[39;00m num_proc \u001b[38;5;241m<\u001b[39m\u001b[38;5;241m=\u001b[39m \u001b[38;5;241m1\u001b[39m \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mlen\u001b[39m(iterable) \u001b[38;5;241m<\u001b[39m parallel_min_length:\n\u001b[0;32m--> 477\u001b[0m     mapped \u001b[38;5;241m=\u001b[39m [\n\u001b[1;32m    478\u001b[0m         _single_map_nested((function, obj, types, \u001b[38;5;28;01mNone\u001b[39;00m, \u001b[38;5;28;01mTrue\u001b[39;00m, \u001b[38;5;28;01mNone\u001b[39;00m))\n\u001b[1;32m    479\u001b[0m         \u001b[38;5;28;01mfor\u001b[39;00m obj \u001b[38;5;129;01min\u001b[39;00m hf_tqdm(iterable, disable\u001b[38;5;241m=\u001b[39mdisable_tqdm, desc\u001b[38;5;241m=\u001b[39mdesc)\n\u001b[1;32m    480\u001b[0m     ]\n\u001b[1;32m    481\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m    482\u001b[0m     \u001b[38;5;28;01mwith\u001b[39;00m warnings\u001b[38;5;241m.\u001b[39mcatch_warnings():\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/datasets/utils/py_utils.py:478\u001b[0m, in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m    466\u001b[0m     mapped \u001b[38;5;241m=\u001b[39m [\n\u001b[1;32m    467\u001b[0m         map_nested(\n\u001b[1;32m    468\u001b[0m             function\u001b[38;5;241m=\u001b[39mfunction,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    474\u001b[0m         \u001b[38;5;28;01mfor\u001b[39;00m obj \u001b[38;5;129;01min\u001b[39;00m iterable\n\u001b[1;32m    475\u001b[0m     ]\n\u001b[1;32m    476\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m num_proc \u001b[38;5;241m!=\u001b[39m \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m1\u001b[39m \u001b[38;5;129;01mand\u001b[39;00m num_proc \u001b[38;5;241m<\u001b[39m\u001b[38;5;241m=\u001b[39m \u001b[38;5;241m1\u001b[39m \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mlen\u001b[39m(iterable) \u001b[38;5;241m<\u001b[39m parallel_min_length:\n\u001b[1;32m    477\u001b[0m     mapped \u001b[38;5;241m=\u001b[39m [\n\u001b[0;32m--> 478\u001b[0m         \u001b[43m_single_map_nested\u001b[49m\u001b[43m(\u001b[49m\u001b[43m(\u001b[49m\u001b[43mfunction\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mobj\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mtypes\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    479\u001b[0m         \u001b[38;5;28;01mfor\u001b[39;00m obj \u001b[38;5;129;01min\u001b[39;00m hf_tqdm(iterable, disable\u001b[38;5;241m=\u001b[39mdisable_tqdm, desc\u001b[38;5;241m=\u001b[39mdesc)\n\u001b[1;32m    480\u001b[0m     ]\n\u001b[1;32m    481\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m    482\u001b[0m     \u001b[38;5;28;01mwith\u001b[39;00m warnings\u001b[38;5;241m.\u001b[39mcatch_warnings():\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/datasets/utils/py_utils.py:370\u001b[0m, in \u001b[0;36m_single_map_nested\u001b[0;34m(args)\u001b[0m\n\u001b[1;32m    368\u001b[0m \u001b[38;5;66;03m# Singleton first to spare some computation\u001b[39;00m\n\u001b[1;32m    369\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(data_struct, \u001b[38;5;28mdict\u001b[39m) \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(data_struct, types):\n\u001b[0;32m--> 370\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mfunction\u001b[49m\u001b[43m(\u001b[49m\u001b[43mdata_struct\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    372\u001b[0m \u001b[38;5;66;03m# Reduce logging to keep things readable in multiprocessing with tqdm\u001b[39;00m\n\u001b[1;32m    373\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m rank \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;129;01mand\u001b[39;00m logging\u001b[38;5;241m.\u001b[39mget_verbosity() \u001b[38;5;241m<\u001b[39m logging\u001b[38;5;241m.\u001b[39mWARNING:\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/datasets/download/download_manager.py:459\u001b[0m, in \u001b[0;36mDownloadManager._download\u001b[0;34m(self, url_or_filename, download_config)\u001b[0m\n\u001b[1;32m    456\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m is_relative_path(url_or_filename):\n\u001b[1;32m    457\u001b[0m     \u001b[38;5;66;03m# append the relative path to the base_path\u001b[39;00m\n\u001b[1;32m    458\u001b[0m     url_or_filename \u001b[38;5;241m=\u001b[39m url_or_path_join(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_base_path, url_or_filename)\n\u001b[0;32m--> 459\u001b[0m out \u001b[38;5;241m=\u001b[39m \u001b[43mcached_path\u001b[49m\u001b[43m(\u001b[49m\u001b[43murl_or_filename\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdownload_config\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdownload_config\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    460\u001b[0m out \u001b[38;5;241m=\u001b[39m tracked_str(out)\n\u001b[1;32m    461\u001b[0m out\u001b[38;5;241m.\u001b[39mset_origin(url_or_filename)\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/datasets/utils/file_utils.py:190\u001b[0m, in \u001b[0;36mcached_path\u001b[0;34m(url_or_filename, download_config, **download_kwargs)\u001b[0m\n\u001b[1;32m    186\u001b[0m     url_or_filename \u001b[38;5;241m=\u001b[39m strip_protocol(url_or_filename)\n\u001b[1;32m    188\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m is_remote_url(url_or_filename):\n\u001b[1;32m    189\u001b[0m     \u001b[38;5;66;03m# URL, so get it from the cache (downloading if necessary)\u001b[39;00m\n\u001b[0;32m--> 190\u001b[0m     output_path \u001b[38;5;241m=\u001b[39m \u001b[43mget_from_cache\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    191\u001b[0m \u001b[43m        \u001b[49m\u001b[43murl_or_filename\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    192\u001b[0m \u001b[43m        \u001b[49m\u001b[43mcache_dir\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcache_dir\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    193\u001b[0m \u001b[43m        \u001b[49m\u001b[43mforce_download\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdownload_config\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mforce_download\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    194\u001b[0m \u001b[43m        \u001b[49m\u001b[43mproxies\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdownload_config\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mproxies\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    195\u001b[0m \u001b[43m        \u001b[49m\u001b[43mresume_download\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdownload_config\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mresume_download\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    196\u001b[0m \u001b[43m        \u001b[49m\u001b[43muser_agent\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdownload_config\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43muser_agent\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    197\u001b[0m \u001b[43m        \u001b[49m\u001b[43mlocal_files_only\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdownload_config\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mlocal_files_only\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    198\u001b[0m \u001b[43m        \u001b[49m\u001b[43muse_etag\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdownload_config\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43muse_etag\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    199\u001b[0m \u001b[43m        \u001b[49m\u001b[43mmax_retries\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdownload_config\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mmax_retries\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    200\u001b[0m \u001b[43m        \u001b[49m\u001b[43mtoken\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdownload_config\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mtoken\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    201\u001b[0m \u001b[43m        \u001b[49m\u001b[43mignore_url_params\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdownload_config\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mignore_url_params\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    202\u001b[0m \u001b[43m        \u001b[49m\u001b[43mstorage_options\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdownload_config\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mstorage_options\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    203\u001b[0m \u001b[43m        \u001b[49m\u001b[43mdownload_desc\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdownload_config\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mdownload_desc\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    204\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    205\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m os\u001b[38;5;241m.\u001b[39mpath\u001b[38;5;241m.\u001b[39mexists(url_or_filename):\n\u001b[1;32m    206\u001b[0m     \u001b[38;5;66;03m# File, and it exists.\u001b[39;00m\n\u001b[1;32m    207\u001b[0m     output_path \u001b[38;5;241m=\u001b[39m url_or_filename\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/datasets/utils/file_utils.py:632\u001b[0m, in \u001b[0;36mget_from_cache\u001b[0;34m(url, cache_dir, force_download, proxies, etag_timeout, resume_download, user_agent, local_files_only, use_etag, max_retries, token, use_auth_token, ignore_url_params, storage_options, download_desc)\u001b[0m\n\u001b[1;32m    630\u001b[0m     ftp_get(url, temp_file)\n\u001b[1;32m    631\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m scheme \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;129;01min\u001b[39;00m (\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mhttp\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mhttps\u001b[39m\u001b[38;5;124m\"\u001b[39m):\n\u001b[0;32m--> 632\u001b[0m     \u001b[43mfsspec_get\u001b[49m\u001b[43m(\u001b[49m\u001b[43murl\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mtemp_file\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mstorage_options\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mstorage_options\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdesc\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdownload_desc\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    633\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m    634\u001b[0m     http_get(\n\u001b[1;32m    635\u001b[0m         url,\n\u001b[1;32m    636\u001b[0m         temp_file\u001b[38;5;241m=\u001b[39mtemp_file,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    642\u001b[0m         desc\u001b[38;5;241m=\u001b[39mdownload_desc,\n\u001b[1;32m    643\u001b[0m     )\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/datasets/utils/file_utils.py:352\u001b[0m, in \u001b[0;36mfsspec_get\u001b[0;34m(url, temp_file, storage_options, desc)\u001b[0m\n\u001b[1;32m    340\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mGET can be called with at most one path but was called with \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mpaths\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m    341\u001b[0m callback \u001b[38;5;241m=\u001b[39m TqdmCallback(\n\u001b[1;32m    342\u001b[0m     tqdm_kwargs\u001b[38;5;241m=\u001b[39m{\n\u001b[1;32m    343\u001b[0m         \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mdesc\u001b[39m\u001b[38;5;124m\"\u001b[39m: desc \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mDownloading\u001b[39m\u001b[38;5;124m\"\u001b[39m,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    350\u001b[0m     }\n\u001b[1;32m    351\u001b[0m )\n\u001b[0;32m--> 352\u001b[0m \u001b[43mfs\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget_file\u001b[49m\u001b[43m(\u001b[49m\u001b[43mpaths\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;241;43m0\u001b[39;49m\u001b[43m]\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mtemp_file\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mname\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcallback\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcallback\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/fsspec/spec.py:914\u001b[0m, in \u001b[0;36mAbstractFileSystem.get_file\u001b[0;34m(self, rpath, lpath, callback, outfile, **kwargs)\u001b[0m\n\u001b[1;32m    912\u001b[0m data \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mTrue\u001b[39;00m\n\u001b[1;32m    913\u001b[0m \u001b[38;5;28;01mwhile\u001b[39;00m data:\n\u001b[0;32m--> 914\u001b[0m     data \u001b[38;5;241m=\u001b[39m \u001b[43mf1\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mread\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mblocksize\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    915\u001b[0m     segment_len \u001b[38;5;241m=\u001b[39m outfile\u001b[38;5;241m.\u001b[39mwrite(data)\n\u001b[1;32m    916\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m segment_len \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/fsspec/spec.py:1856\u001b[0m, in \u001b[0;36mAbstractBufferedFile.read\u001b[0;34m(self, length)\u001b[0m\n\u001b[1;32m   1853\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m length \u001b[38;5;241m==\u001b[39m \u001b[38;5;241m0\u001b[39m:\n\u001b[1;32m   1854\u001b[0m     \u001b[38;5;66;03m# don't even bother calling fetch\u001b[39;00m\n\u001b[1;32m   1855\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;124mb\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m-> 1856\u001b[0m out \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mcache\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_fetch\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mloc\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mloc\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m+\u001b[39;49m\u001b[43m \u001b[49m\u001b[43mlength\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1857\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mloc \u001b[38;5;241m+\u001b[39m\u001b[38;5;241m=\u001b[39m \u001b[38;5;28mlen\u001b[39m(out)\n\u001b[1;32m   1858\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m out\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/fsspec/caching.py:189\u001b[0m, in \u001b[0;36mReadAheadCache._fetch\u001b[0;34m(self, start, end)\u001b[0m\n\u001b[1;32m    187\u001b[0m     part \u001b[38;5;241m=\u001b[39m \u001b[38;5;124mb\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m    188\u001b[0m end \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mmin\u001b[39m(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39msize, end \u001b[38;5;241m+\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mblocksize)\n\u001b[0;32m--> 189\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mcache \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfetcher\u001b[49m\u001b[43m(\u001b[49m\u001b[43mstart\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mend\u001b[49m\u001b[43m)\u001b[49m  \u001b[38;5;66;03m# new block replaces old\u001b[39;00m\n\u001b[1;32m    190\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mstart \u001b[38;5;241m=\u001b[39m start\n\u001b[1;32m    191\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mend \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mstart \u001b[38;5;241m+\u001b[39m \u001b[38;5;28mlen\u001b[39m(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mcache)\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/huggingface_hub/hf_file_system.py:625\u001b[0m, in \u001b[0;36mHfFileSystemFile._fetch_range\u001b[0;34m(self, start, end)\u001b[0m\n\u001b[1;32m    614\u001b[0m headers \u001b[38;5;241m=\u001b[39m {\n\u001b[1;32m    615\u001b[0m     \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mrange\u001b[39m\u001b[38;5;124m\"\u001b[39m: \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mbytes=\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mstart\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m-\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mend\u001b[38;5;250m \u001b[39m\u001b[38;5;241m-\u001b[39m\u001b[38;5;250m \u001b[39m\u001b[38;5;241m1\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m,\n\u001b[1;32m    616\u001b[0m     \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mfs\u001b[38;5;241m.\u001b[39m_api\u001b[38;5;241m.\u001b[39m_build_hf_headers(),\n\u001b[1;32m    617\u001b[0m }\n\u001b[1;32m    618\u001b[0m url \u001b[38;5;241m=\u001b[39m hf_hub_url(\n\u001b[1;32m    619\u001b[0m     repo_id\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mresolved_path\u001b[38;5;241m.\u001b[39mrepo_id,\n\u001b[1;32m    620\u001b[0m     revision\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mresolved_path\u001b[38;5;241m.\u001b[39mrevision,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    623\u001b[0m     endpoint\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mfs\u001b[38;5;241m.\u001b[39mendpoint,\n\u001b[1;32m    624\u001b[0m )\n\u001b[0;32m--> 625\u001b[0m r \u001b[38;5;241m=\u001b[39m \u001b[43mhttp_backoff\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mGET\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43murl\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mheaders\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mheaders\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    626\u001b[0m hf_raise_for_status(r)\n\u001b[1;32m    627\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m r\u001b[38;5;241m.\u001b[39mcontent\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/huggingface_hub/utils/_http.py:281\u001b[0m, in \u001b[0;36mhttp_backoff\u001b[0;34m(method, url, max_retries, base_wait_time, max_wait_time, retry_on_exceptions, retry_on_status_codes, **kwargs)\u001b[0m\n\u001b[1;32m    278\u001b[0m     kwargs[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mdata\u001b[39m\u001b[38;5;124m\"\u001b[39m]\u001b[38;5;241m.\u001b[39mseek(io_obj_initial_pos)\n\u001b[1;32m    280\u001b[0m \u001b[38;5;66;03m# Perform request and return if status_code is not in the retry list.\u001b[39;00m\n\u001b[0;32m--> 281\u001b[0m response \u001b[38;5;241m=\u001b[39m \u001b[43msession\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mrequest\u001b[49m\u001b[43m(\u001b[49m\u001b[43mmethod\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mmethod\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43murl\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43murl\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    282\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m response\u001b[38;5;241m.\u001b[39mstatus_code \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;129;01min\u001b[39;00m retry_on_status_codes:\n\u001b[1;32m    283\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m response\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/requests/sessions.py:589\u001b[0m, in \u001b[0;36mSession.request\u001b[0;34m(self, method, url, params, data, headers, cookies, files, auth, timeout, allow_redirects, proxies, hooks, stream, verify, cert, json)\u001b[0m\n\u001b[1;32m    584\u001b[0m send_kwargs \u001b[38;5;241m=\u001b[39m {\n\u001b[1;32m    585\u001b[0m     \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mtimeout\u001b[39m\u001b[38;5;124m\"\u001b[39m: timeout,\n\u001b[1;32m    586\u001b[0m     \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mallow_redirects\u001b[39m\u001b[38;5;124m\"\u001b[39m: allow_redirects,\n\u001b[1;32m    587\u001b[0m }\n\u001b[1;32m    588\u001b[0m send_kwargs\u001b[38;5;241m.\u001b[39mupdate(settings)\n\u001b[0;32m--> 589\u001b[0m resp \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43msend\u001b[49m\u001b[43m(\u001b[49m\u001b[43mprep\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43msend_kwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    591\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m resp\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/requests/sessions.py:725\u001b[0m, in \u001b[0;36mSession.send\u001b[0;34m(self, request, **kwargs)\u001b[0m\n\u001b[1;32m    722\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m allow_redirects:\n\u001b[1;32m    723\u001b[0m     \u001b[38;5;66;03m# Redirect resolving generator.\u001b[39;00m\n\u001b[1;32m    724\u001b[0m     gen \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mresolve_redirects(r, request, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n\u001b[0;32m--> 725\u001b[0m     history \u001b[38;5;241m=\u001b[39m [resp \u001b[38;5;28;01mfor\u001b[39;00m resp \u001b[38;5;129;01min\u001b[39;00m gen]\n\u001b[1;32m    726\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m    727\u001b[0m     history \u001b[38;5;241m=\u001b[39m []\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/requests/sessions.py:725\u001b[0m, in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m    722\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m allow_redirects:\n\u001b[1;32m    723\u001b[0m     \u001b[38;5;66;03m# Redirect resolving generator.\u001b[39;00m\n\u001b[1;32m    724\u001b[0m     gen \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mresolve_redirects(r, request, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n\u001b[0;32m--> 725\u001b[0m     history \u001b[38;5;241m=\u001b[39m [resp \u001b[38;5;28;01mfor\u001b[39;00m resp \u001b[38;5;129;01min\u001b[39;00m gen]\n\u001b[1;32m    726\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m    727\u001b[0m     history \u001b[38;5;241m=\u001b[39m []\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/requests/sessions.py:266\u001b[0m, in \u001b[0;36mSessionRedirectMixin.resolve_redirects\u001b[0;34m(self, resp, req, stream, timeout, verify, cert, proxies, yield_requests, **adapter_kwargs)\u001b[0m\n\u001b[1;32m    263\u001b[0m     \u001b[38;5;28;01myield\u001b[39;00m req\n\u001b[1;32m    264\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m--> 266\u001b[0m     resp \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43msend\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    267\u001b[0m \u001b[43m        \u001b[49m\u001b[43mreq\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    268\u001b[0m \u001b[43m        \u001b[49m\u001b[43mstream\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mstream\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    269\u001b[0m \u001b[43m        \u001b[49m\u001b[43mtimeout\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mtimeout\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    270\u001b[0m \u001b[43m        \u001b[49m\u001b[43mverify\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mverify\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    271\u001b[0m \u001b[43m        \u001b[49m\u001b[43mcert\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcert\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    272\u001b[0m \u001b[43m        \u001b[49m\u001b[43mproxies\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mproxies\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    273\u001b[0m \u001b[43m        \u001b[49m\u001b[43mallow_redirects\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mFalse\u001b[39;49;00m\u001b[43m,\u001b[49m\n\u001b[1;32m    274\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43madapter_kwargs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    275\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    277\u001b[0m     extract_cookies_to_jar(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mcookies, prepared_request, resp\u001b[38;5;241m.\u001b[39mraw)\n\u001b[1;32m    279\u001b[0m     \u001b[38;5;66;03m# extract redirect url, if any, for the next loop\u001b[39;00m\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/requests/sessions.py:747\u001b[0m, in \u001b[0;36mSession.send\u001b[0;34m(self, request, **kwargs)\u001b[0m\n\u001b[1;32m    744\u001b[0m         \u001b[38;5;28;01mpass\u001b[39;00m\n\u001b[1;32m    746\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m stream:\n\u001b[0;32m--> 747\u001b[0m     \u001b[43mr\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mcontent\u001b[49m\n\u001b[1;32m    749\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m r\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/requests/models.py:899\u001b[0m, in \u001b[0;36mResponse.content\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    897\u001b[0m         \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_content \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[1;32m    898\u001b[0m     \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m--> 899\u001b[0m         \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_content \u001b[38;5;241m=\u001b[39m \u001b[38;5;124;43mb\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mjoin\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43miter_content\u001b[49m\u001b[43m(\u001b[49m\u001b[43mCONTENT_CHUNK_SIZE\u001b[49m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;124mb\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m    901\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_content_consumed \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mTrue\u001b[39;00m\n\u001b[1;32m    902\u001b[0m \u001b[38;5;66;03m# don't need to release the connection; that's been handled by urllib3\u001b[39;00m\n\u001b[1;32m    903\u001b[0m \u001b[38;5;66;03m# since we exhausted the data.\u001b[39;00m\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/requests/models.py:816\u001b[0m, in \u001b[0;36mResponse.iter_content.<locals>.generate\u001b[0;34m()\u001b[0m\n\u001b[1;32m    814\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mhasattr\u001b[39m(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mraw, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mstream\u001b[39m\u001b[38;5;124m\"\u001b[39m):\n\u001b[1;32m    815\u001b[0m     \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 816\u001b[0m         \u001b[38;5;28;01myield from\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mraw\u001b[38;5;241m.\u001b[39mstream(chunk_size, decode_content\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mTrue\u001b[39;00m)\n\u001b[1;32m    817\u001b[0m     \u001b[38;5;28;01mexcept\u001b[39;00m ProtocolError \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[1;32m    818\u001b[0m         \u001b[38;5;28;01mraise\u001b[39;00m ChunkedEncodingError(e)\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/urllib3/response.py:628\u001b[0m, in \u001b[0;36mHTTPResponse.stream\u001b[0;34m(self, amt, decode_content)\u001b[0m\n\u001b[1;32m    626\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m    627\u001b[0m     \u001b[38;5;28;01mwhile\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m is_fp_closed(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_fp):\n\u001b[0;32m--> 628\u001b[0m         data \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mread\u001b[49m\u001b[43m(\u001b[49m\u001b[43mamt\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mamt\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdecode_content\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdecode_content\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    630\u001b[0m         \u001b[38;5;28;01mif\u001b[39;00m data:\n\u001b[1;32m    631\u001b[0m             \u001b[38;5;28;01myield\u001b[39;00m data\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/urllib3/response.py:567\u001b[0m, in \u001b[0;36mHTTPResponse.read\u001b[0;34m(self, amt, decode_content, cache_content)\u001b[0m\n\u001b[1;32m    564\u001b[0m fp_closed \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mgetattr\u001b[39m(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_fp, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mclosed\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;28;01mFalse\u001b[39;00m)\n\u001b[1;32m    566\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_error_catcher():\n\u001b[0;32m--> 567\u001b[0m     data \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_fp_read\u001b[49m\u001b[43m(\u001b[49m\u001b[43mamt\u001b[49m\u001b[43m)\u001b[49m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m fp_closed \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;124mb\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m    568\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m amt \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m    569\u001b[0m         flush_decoder \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mTrue\u001b[39;00m\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/site-packages/urllib3/response.py:533\u001b[0m, in \u001b[0;36mHTTPResponse._fp_read\u001b[0;34m(self, amt)\u001b[0m\n\u001b[1;32m    530\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m buffer\u001b[38;5;241m.\u001b[39mgetvalue()\n\u001b[1;32m    531\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m    532\u001b[0m     \u001b[38;5;66;03m# StringIO doesn't like amt=None\u001b[39;00m\n\u001b[0;32m--> 533\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_fp\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mread\u001b[49m\u001b[43m(\u001b[49m\u001b[43mamt\u001b[49m\u001b[43m)\u001b[49m \u001b[38;5;28;01mif\u001b[39;00m amt \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_fp\u001b[38;5;241m.\u001b[39mread()\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/http/client.py:463\u001b[0m, in \u001b[0;36mHTTPResponse.read\u001b[0;34m(self, amt)\u001b[0m\n\u001b[1;32m    460\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m amt \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m    461\u001b[0m     \u001b[38;5;66;03m# Amount is given, implement using readinto\u001b[39;00m\n\u001b[1;32m    462\u001b[0m     b \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mbytearray\u001b[39m(amt)\n\u001b[0;32m--> 463\u001b[0m     n \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mreadinto\u001b[49m\u001b[43m(\u001b[49m\u001b[43mb\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    464\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mmemoryview\u001b[39m(b)[:n]\u001b[38;5;241m.\u001b[39mtobytes()\n\u001b[1;32m    465\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m    466\u001b[0m     \u001b[38;5;66;03m# Amount is not given (unbounded read) so we must check self.length\u001b[39;00m\n\u001b[1;32m    467\u001b[0m     \u001b[38;5;66;03m# and self.chunked\u001b[39;00m\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/http/client.py:507\u001b[0m, in \u001b[0;36mHTTPResponse.readinto\u001b[0;34m(self, b)\u001b[0m\n\u001b[1;32m    502\u001b[0m         b \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mmemoryview\u001b[39m(b)[\u001b[38;5;241m0\u001b[39m:\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mlength]\n\u001b[1;32m    504\u001b[0m \u001b[38;5;66;03m# we do not use _safe_read() here because this may be a .will_close\u001b[39;00m\n\u001b[1;32m    505\u001b[0m \u001b[38;5;66;03m# connection, and the user is reading more bytes than will be provided\u001b[39;00m\n\u001b[1;32m    506\u001b[0m \u001b[38;5;66;03m# (for example, reading in 1k chunks)\u001b[39;00m\n\u001b[0;32m--> 507\u001b[0m n \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfp\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mreadinto\u001b[49m\u001b[43m(\u001b[49m\u001b[43mb\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    508\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m n \u001b[38;5;129;01mand\u001b[39;00m b:\n\u001b[1;32m    509\u001b[0m     \u001b[38;5;66;03m# Ideally, we would raise IncompleteRead if the content-length\u001b[39;00m\n\u001b[1;32m    510\u001b[0m     \u001b[38;5;66;03m# wasn't satisfied, but it might break compatibility.\u001b[39;00m\n\u001b[1;32m    511\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_close_conn()\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/socket.py:704\u001b[0m, in \u001b[0;36mSocketIO.readinto\u001b[0;34m(self, b)\u001b[0m\n\u001b[1;32m    702\u001b[0m \u001b[38;5;28;01mwhile\u001b[39;00m \u001b[38;5;28;01mTrue\u001b[39;00m:\n\u001b[1;32m    703\u001b[0m     \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 704\u001b[0m         \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_sock\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mrecv_into\u001b[49m\u001b[43m(\u001b[49m\u001b[43mb\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    705\u001b[0m     \u001b[38;5;28;01mexcept\u001b[39;00m timeout:\n\u001b[1;32m    706\u001b[0m         \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_timeout_occurred \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mTrue\u001b[39;00m\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/ssl.py:1242\u001b[0m, in \u001b[0;36mSSLSocket.recv_into\u001b[0;34m(self, buffer, nbytes, flags)\u001b[0m\n\u001b[1;32m   1238\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m flags \u001b[38;5;241m!=\u001b[39m \u001b[38;5;241m0\u001b[39m:\n\u001b[1;32m   1239\u001b[0m         \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\n\u001b[1;32m   1240\u001b[0m           \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mnon-zero flags not allowed in calls to recv_into() on \u001b[39m\u001b[38;5;132;01m%s\u001b[39;00m\u001b[38;5;124m\"\u001b[39m \u001b[38;5;241m%\u001b[39m\n\u001b[1;32m   1241\u001b[0m           \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__class__\u001b[39m)\n\u001b[0;32m-> 1242\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mread\u001b[49m\u001b[43m(\u001b[49m\u001b[43mnbytes\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mbuffer\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1243\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m   1244\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28msuper\u001b[39m()\u001b[38;5;241m.\u001b[39mrecv_into(buffer, nbytes, flags)\n",
+      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.9/ssl.py:1100\u001b[0m, in \u001b[0;36mSSLSocket.read\u001b[0;34m(self, len, buffer)\u001b[0m\n\u001b[1;32m   1098\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m   1099\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m buffer \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m-> 1100\u001b[0m         \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_sslobj\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mread\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mlen\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mbuffer\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1101\u001b[0m     \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m   1102\u001b[0m         \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_sslobj\u001b[38;5;241m.\u001b[39mread(\u001b[38;5;28mlen\u001b[39m)\n",
+      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
+     ]
+    }
+   ],
+   "source": [
+    "from datasets import load_dataset\n",
+    "ds = load_dataset('test-org-usm3d/usm-training-data')"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.16"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

read_write_colmap.py

@@ -0,0 +1,489 @@
+# Modified to read from bytes-like object by Dmytro Mishkin.
+# The original license is below:
+# Copyright (c) 2018, ETH Zurich and UNC Chapel Hill.
+# All rights reserved.
+#
+# Redistribution and use in source and binary forms, with or without
+# modification, are permitted provided that the following conditions are met:
+#
+#     * Redistributions of source code must retain the above copyright
+#       notice, this list of conditions and the following disclaimer.
+#
+#     * Redistributions in binary form must reproduce the above copyright
+#       notice, this list of conditions and the following disclaimer in the
+#       documentation and/or other materials provided with the distribution.
+#
+#     * Neither the name of ETH Zurich and UNC Chapel Hill nor the names of
+#       its contributors may be used to endorse or promote products derived
+#       from this software without specific prior written permission.
+#
+# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+# AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+# IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+# ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+# LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+# CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+# SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+# INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+# CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+# ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+# POSSIBILITY OF SUCH DAMAGE.
+#
+# Author: Johannes L. Schoenberger (jsch-at-demuc-dot-de)
+
+import os
+import collections
+import numpy as np
+import struct
+import argparse
+
+
+CameraModel = collections.namedtuple(
+    "CameraModel", ["model_id", "model_name", "num_params"])
+Camera = collections.namedtuple(
+    "Camera", ["id", "model", "width", "height", "params"])
+BaseImage = collections.namedtuple(
+    "Image", ["id", "qvec", "tvec", "camera_id", "name", "xys", "point3D_ids"])
+Point3D = collections.namedtuple(
+    "Point3D", ["id", "xyz", "rgb", "error", "image_ids", "point2D_idxs"])
+
+
+class Image(BaseImage):
+    def qvec2rotmat(self):
+        return qvec2rotmat(self.qvec)
+
+
+CAMERA_MODELS = {
+    CameraModel(model_id=0, model_name="SIMPLE_PINHOLE", num_params=3),
+    CameraModel(model_id=1, model_name="PINHOLE", num_params=4),
+    CameraModel(model_id=2, model_name="SIMPLE_RADIAL", num_params=4),
+    CameraModel(model_id=3, model_name="RADIAL", num_params=5),
+    CameraModel(model_id=4, model_name="OPENCV", num_params=8),
+    CameraModel(model_id=5, model_name="OPENCV_FISHEYE", num_params=8),
+    CameraModel(model_id=6, model_name="FULL_OPENCV", num_params=12),
+    CameraModel(model_id=7, model_name="FOV", num_params=5),
+    CameraModel(model_id=8, model_name="SIMPLE_RADIAL_FISHEYE", num_params=4),
+    CameraModel(model_id=9, model_name="RADIAL_FISHEYE", num_params=5),
+    CameraModel(model_id=10, model_name="THIN_PRISM_FISHEYE", num_params=12)
+}
+CAMERA_MODEL_IDS = dict([(camera_model.model_id, camera_model)
+                         for camera_model in CAMERA_MODELS])
+CAMERA_MODEL_NAMES = dict([(camera_model.model_name, camera_model)
+                           for camera_model in CAMERA_MODELS])
+
+
+def read_next_bytes(fid, num_bytes, format_char_sequence, endian_character="<"):
+    """Read and unpack the next bytes from a binary file.
+    :param fid:
+    :param num_bytes: Sum of combination of {2, 4, 8}, e.g. 2, 6, 16, 30, etc.
+    :param format_char_sequence: List of {c, e, f, d, h, H, i, I, l, L, q, Q}.
+    :param endian_character: Any of {@, =, <, >, !}
+    :return: Tuple of read and unpacked values.
+    """
+    data = fid.read(num_bytes)
+    return struct.unpack(endian_character + format_char_sequence, data)
+
+
+def write_next_bytes(fid, data, format_char_sequence, endian_character="<"):
+    """pack and write to a binary file.
+    :param fid:
+    :param data: data to send, if multiple elements are sent at the same time,
+    they should be encapsuled either in a list or a tuple
+    :param format_char_sequence: List of {c, e, f, d, h, H, i, I, l, L, q, Q}.
+    should be the same length as the data list or tuple
+    :param endian_character: Any of {@, =, <, >, !}
+    """
+    if isinstance(data, (list, tuple)):
+        bytes = struct.pack(endian_character + format_char_sequence, *data)
+    else:
+        bytes = struct.pack(endian_character + format_char_sequence, data)
+    fid.write(bytes)
+
+
+def read_cameras_text(path):
+    """
+    see: src/base/reconstruction.cc
+        void Reconstruction::WriteCamerasText(const std::string& path)
+        void Reconstruction::ReadCamerasText(const std::string& path)
+    """
+    cameras = {}
+    with open(path, "r") as fid:
+        while True:
+            line = fid.readline()
+            if not line:
+                break
+            line = line.strip()
+            if len(line) > 0 and line[0] != "#":
+                elems = line.split()
+                camera_id = int(elems[0])
+                model = elems[1]
+                width = int(elems[2])
+                height = int(elems[3])
+                params = np.array(tuple(map(float, elems[4:])))
+                cameras[camera_id] = Camera(id=camera_id, model=model,
+                                            width=width, height=height,
+                                            params=params)
+    return cameras
+
+
+def read_cameras_binary(path_to_model_file=None, fid=None):
+    """
+    see: src/base/reconstruction.cc
+        void Reconstruction::WriteCamerasBinary(const std::string& path)
+        void Reconstruction::ReadCamerasBinary(const std::string& path)
+    """
+    cameras = {}
+    if fid is None:
+        fid = open(path_to_model_file, "rb")
+    num_cameras = read_next_bytes(fid, 8, "Q")[0]
+    for _ in range(num_cameras):
+        camera_properties = read_next_bytes(
+            fid, num_bytes=24, format_char_sequence="iiQQ")
+        camera_id = camera_properties[0]
+        model_id = camera_properties[1]
+        model_name = CAMERA_MODEL_IDS[camera_properties[1]].model_name
+        width = camera_properties[2]
+        height = camera_properties[3]
+        num_params = CAMERA_MODEL_IDS[model_id].num_params
+        params = read_next_bytes(fid, num_bytes=8*num_params,
+                                    format_char_sequence="d"*num_params)
+        cameras[camera_id] = Camera(id=camera_id,
+                                    model=model_name,
+                                    width=width,
+                                    height=height,
+                                    params=np.array(params))
+    assert len(cameras) == num_cameras
+    if path_to_model_file is not None:
+        fid.close()
+    return cameras
+
+
+def write_cameras_text(cameras, path):
+    """
+    see: src/base/reconstruction.cc
+        void Reconstruction::WriteCamerasText(const std::string& path)
+        void Reconstruction::ReadCamerasText(const std::string& path)
+    """
+    HEADER = "# Camera list with one line of data per camera:\n" + \
+             "#   CAMERA_ID, MODEL, WIDTH, HEIGHT, PARAMS[]\n" + \
+             "# Number of cameras: {}\n".format(len(cameras))
+    with open(path, "w") as fid:
+        fid.write(HEADER)
+        for _, cam in cameras.items():
+            to_write = [cam.id, cam.model, cam.width, cam.height, *cam.params]
+            line = " ".join([str(elem) for elem in to_write])
+            fid.write(line + "\n")
+
+
+def write_cameras_binary(cameras, path_to_model_file):
+    """
+    see: src/base/reconstruction.cc
+        void Reconstruction::WriteCamerasBinary(const std::string& path)
+        void Reconstruction::ReadCamerasBinary(const std::string& path)
+    """
+    with open(path_to_model_file, "wb") as fid:
+        write_next_bytes(fid, len(cameras), "Q")
+        for _, cam in cameras.items():
+            model_id = CAMERA_MODEL_NAMES[cam.model].model_id
+            camera_properties = [cam.id,
+                                 model_id,
+                                 cam.width,
+                                 cam.height]
+            write_next_bytes(fid, camera_properties, "iiQQ")
+            for p in cam.params:
+                write_next_bytes(fid, float(p), "d")
+    return cameras
+
+
+def read_images_text(path):
+    """
+    see: src/base/reconstruction.cc
+        void Reconstruction::ReadImagesText(const std::string& path)
+        void Reconstruction::WriteImagesText(const std::string& path)
+    """
+    images = {}
+    with open(path, "r") as fid:
+        while True:
+            line = fid.readline()
+            if not line:
+                break
+            line = line.strip()
+            if len(line) > 0 and line[0] != "#":
+                elems = line.split()
+                image_id = int(elems[0])
+                qvec = np.array(tuple(map(float, elems[1:5])))
+                tvec = np.array(tuple(map(float, elems[5:8])))
+                camera_id = int(elems[8])
+                image_name = elems[9]
+                elems = fid.readline().split()
+                xys = np.column_stack([tuple(map(float, elems[0::3])),
+                                       tuple(map(float, elems[1::3]))])
+                point3D_ids = np.array(tuple(map(int, elems[2::3])))
+                images[image_id] = Image(
+                    id=image_id, qvec=qvec, tvec=tvec,
+                    camera_id=camera_id, name=image_name,
+                    xys=xys, point3D_ids=point3D_ids)
+    return images
+
+
+def read_images_binary(path_to_model_file=None, fid=None):
+    """
+    see: src/base/reconstruction.cc
+        void Reconstruction::ReadImagesBinary(const std::string& path)
+        void Reconstruction::WriteImagesBinary(const std::string& path)
+    """
+    images = {}
+    if fid is None:
+        fid = open(path_to_model_file, "rb")
+    num_reg_images = read_next_bytes(fid, 8, "Q")[0]
+    for _ in range(num_reg_images):
+        binary_image_properties = read_next_bytes(
+            fid, num_bytes=64, format_char_sequence="idddddddi")
+        image_id = binary_image_properties[0]
+        qvec = np.array(binary_image_properties[1:5])
+        tvec = np.array(binary_image_properties[5:8])
+        camera_id = binary_image_properties[8]
+        image_name = ""
+        current_char = read_next_bytes(fid, 1, "c")[0]
+        while current_char != b"\x00":   # look for the ASCII 0 entry
+            image_name += current_char.decode("utf-8")
+            current_char = read_next_bytes(fid, 1, "c")[0]
+        num_points2D = read_next_bytes(fid, num_bytes=8,
+                                        format_char_sequence="Q")[0]
+        x_y_id_s = read_next_bytes(fid, num_bytes=24*num_points2D,
+                                    format_char_sequence="ddq"*num_points2D)
+        xys = np.column_stack([tuple(map(float, x_y_id_s[0::3])),
+                                tuple(map(float, x_y_id_s[1::3]))])
+        point3D_ids = np.array(tuple(map(int, x_y_id_s[2::3])))
+        images[image_id] = Image(
+            id=image_id, qvec=qvec, tvec=tvec,
+            camera_id=camera_id, name=image_name,
+            xys=xys, point3D_ids=point3D_ids)
+    if path_to_model_file is not None:
+        fid.close()
+    return images
+
+
+def write_images_text(images, path):
+    """
+    see: src/base/reconstruction.cc
+        void Reconstruction::ReadImagesText(const std::string& path)
+        void Reconstruction::WriteImagesText(const std::string& path)
+    """
+    if len(images) == 0:
+        mean_observations = 0
+    else:
+        mean_observations = sum((len(img.point3D_ids) for _, img in images.items()))/len(images)
+    HEADER = "# Image list with two lines of data per image:\n" + \
+             "#   IMAGE_ID, QW, QX, QY, QZ, TX, TY, TZ, CAMERA_ID, NAME\n" + \
+             "#   POINTS2D[] as (X, Y, POINT3D_ID)\n" + \
+             "# Number of images: {}, mean observations per image: {}\n".format(len(images), mean_observations)
+
+    with open(path, "w") as fid:
+        fid.write(HEADER)
+        for _, img in images.items():
+            image_header = [img.id, *img.qvec, *img.tvec, img.camera_id, img.name]
+            first_line = " ".join(map(str, image_header))
+            fid.write(first_line + "\n")
+
+            points_strings = []
+            for xy, point3D_id in zip(img.xys, img.point3D_ids):
+                points_strings.append(" ".join(map(str, [*xy, point3D_id])))
+            fid.write(" ".join(points_strings) + "\n")
+
+
+def write_images_binary(images, path_to_model_file):
+    """
+    see: src/base/reconstruction.cc
+        void Reconstruction::ReadImagesBinary(const std::string& path)
+        void Reconstruction::WriteImagesBinary(const std::string& path)
+    """
+    with open(path_to_model_file, "wb") as fid:
+        write_next_bytes(fid, len(images), "Q")
+        for _, img in images.items():
+            write_next_bytes(fid, img.id, "i")
+            write_next_bytes(fid, img.qvec.tolist(), "dddd")
+            write_next_bytes(fid, img.tvec.tolist(), "ddd")
+            write_next_bytes(fid, img.camera_id, "i")
+            for char in img.name:
+                write_next_bytes(fid, char.encode("utf-8"), "c")
+            write_next_bytes(fid, b"\x00", "c")
+            write_next_bytes(fid, len(img.point3D_ids), "Q")
+            for xy, p3d_id in zip(img.xys, img.point3D_ids):
+                write_next_bytes(fid, [*xy, p3d_id], "ddq")
+
+
+def read_points3D_text(path):
+    """
+    see: src/base/reconstruction.cc
+        void Reconstruction::ReadPoints3DText(const std::string& path)
+        void Reconstruction::WritePoints3DText(const std::string& path)
+    """
+    points3D = {}
+    with open(path, "r") as fid:
+        while True:
+            line = fid.readline()
+            if not line:
+                break
+            line = line.strip()
+            if len(line) > 0 and line[0] != "#":
+                elems = line.split()
+                point3D_id = int(elems[0])
+                xyz = np.array(tuple(map(float, elems[1:4])))
+                rgb = np.array(tuple(map(int, elems[4:7])))
+                error = float(elems[7])
+                image_ids = np.array(tuple(map(int, elems[8::2])))
+                point2D_idxs = np.array(tuple(map(int, elems[9::2])))
+                points3D[point3D_id] = Point3D(id=point3D_id, xyz=xyz, rgb=rgb,
+                                               error=error, image_ids=image_ids,
+                                               point2D_idxs=point2D_idxs)
+    return points3D
+
+
+def read_points3D_binary(path_to_model_file=None, fid=None):
+    """
+    see: src/base/reconstruction.cc
+        void Reconstruction::ReadPoints3DBinary(const std::string& path)
+        void Reconstruction::WritePoints3DBinary(const std::string& path)
+    """
+    points3D = {}
+    if fid is None:
+        fid = open(path_to_model_file, "rb")
+    num_points = read_next_bytes(fid, 8, "Q")[0]
+    for _ in range(num_points):
+        binary_point_line_properties = read_next_bytes(
+            fid, num_bytes=43, format_char_sequence="QdddBBBd")
+        point3D_id = binary_point_line_properties[0]
+        xyz = np.array(binary_point_line_properties[1:4])
+        rgb = np.array(binary_point_line_properties[4:7])
+        error = np.array(binary_point_line_properties[7])
+        track_length = read_next_bytes(
+            fid, num_bytes=8, format_char_sequence="Q")[0]
+        track_elems = read_next_bytes(
+            fid, num_bytes=8*track_length,
+            format_char_sequence="ii"*track_length)
+        image_ids = np.array(tuple(map(int, track_elems[0::2])))
+        point2D_idxs = np.array(tuple(map(int, track_elems[1::2])))
+        points3D[point3D_id] = Point3D(
+            id=point3D_id, xyz=xyz, rgb=rgb,
+            error=error, image_ids=image_ids,
+            point2D_idxs=point2D_idxs)
+    if path_to_model_file is not None:
+        fid.close()
+    return points3D
+
+
+def write_points3D_text(points3D, path):
+    """
+    see: src/base/reconstruction.cc
+        void Reconstruction::ReadPoints3DText(const std::string& path)
+        void Reconstruction::WritePoints3DText(const std::string& path)
+    """
+    if len(points3D) == 0:
+        mean_track_length = 0
+    else:
+        mean_track_length = sum((len(pt.image_ids) for _, pt in points3D.items()))/len(points3D)
+    HEADER = "# 3D point list with one line of data per point:\n" + \
+             "#   POINT3D_ID, X, Y, Z, R, G, B, ERROR, TRACK[] as (IMAGE_ID, POINT2D_IDX)\n" + \
+             "# Number of points: {}, mean track length: {}\n".format(len(points3D), mean_track_length)
+
+    with open(path, "w") as fid:
+        fid.write(HEADER)
+        for _, pt in points3D.items():
+            point_header = [pt.id, *pt.xyz, *pt.rgb, pt.error]
+            fid.write(" ".join(map(str, point_header)) + " ")
+            track_strings = []
+            for image_id, point2D in zip(pt.image_ids, pt.point2D_idxs):
+                track_strings.append(" ".join(map(str, [image_id, point2D])))
+            fid.write(" ".join(track_strings) + "\n")
+
+
+def write_points3D_binary(points3D, path_to_model_file):
+    """
+    see: src/base/reconstruction.cc
+        void Reconstruction::ReadPoints3DBinary(const std::string& path)
+        void Reconstruction::WritePoints3DBinary(const std::string& path)
+    """
+    with open(path_to_model_file, "wb") as fid:
+        write_next_bytes(fid, len(points3D), "Q")
+        for _, pt in points3D.items():
+            write_next_bytes(fid, pt.id, "Q")
+            write_next_bytes(fid, pt.xyz.tolist(), "ddd")
+            write_next_bytes(fid, pt.rgb.tolist(), "BBB")
+            write_next_bytes(fid, pt.error, "d")
+            track_length = pt.image_ids.shape[0]
+            write_next_bytes(fid, track_length, "Q")
+            for image_id, point2D_id in zip(pt.image_ids, pt.point2D_idxs):
+                write_next_bytes(fid, [image_id, point2D_id], "ii")
+
+
+def detect_model_format(path, ext):
+    if os.path.isfile(os.path.join(path, "cameras"  + ext)) and \
+       os.path.isfile(os.path.join(path, "images"   + ext)) and \
+       os.path.isfile(os.path.join(path, "points3D" + ext)):
+        print("Detected model format: '" + ext + "'")
+        return True
+
+    return False
+
+
+def read_model(path, ext=""):
+    # try to detect the extension automatically
+    if ext == "":
+        if detect_model_format(path, ".bin"):
+            ext = ".bin"
+        elif detect_model_format(path, ".txt"):
+            ext = ".txt"
+        else:
+            print("Provide model format: '.bin' or '.txt'")
+            return
+
+    if ext == ".txt":
+        cameras = read_cameras_text(os.path.join(path, "cameras" + ext))
+        images = read_images_text(os.path.join(path, "images" + ext))
+        points3D = read_points3D_text(os.path.join(path, "points3D") + ext)
+    else:
+        cameras = read_cameras_binary(os.path.join(path, "cameras" + ext))
+        images = read_images_binary(os.path.join(path, "images" + ext))
+        points3D = read_points3D_binary(os.path.join(path, "points3D") + ext)
+    return cameras, images, points3D
+
+
+def write_model(cameras, images, points3D, path, ext=".bin"):
+    if ext == ".txt":
+        write_cameras_text(cameras, os.path.join(path, "cameras" + ext))
+        write_images_text(images, os.path.join(path, "images" + ext))
+        write_points3D_text(points3D, os.path.join(path, "points3D") + ext)
+    else:
+        write_cameras_binary(cameras, os.path.join(path, "cameras" + ext))
+        write_images_binary(images, os.path.join(path, "images" + ext))
+        write_points3D_binary(points3D, os.path.join(path, "points3D") + ext)
+    return cameras, images, points3D
+
+
+def qvec2rotmat(qvec):
+    return np.array([
+        [1 - 2 * qvec[2]**2 - 2 * qvec[3]**2,
+         2 * qvec[1] * qvec[2] - 2 * qvec[0] * qvec[3],
+         2 * qvec[3] * qvec[1] + 2 * qvec[0] * qvec[2]],
+        [2 * qvec[1] * qvec[2] + 2 * qvec[0] * qvec[3],
+         1 - 2 * qvec[1]**2 - 2 * qvec[3]**2,
+         2 * qvec[2] * qvec[3] - 2 * qvec[0] * qvec[1]],
+        [2 * qvec[3] * qvec[1] - 2 * qvec[0] * qvec[2],
+         2 * qvec[2] * qvec[3] + 2 * qvec[0] * qvec[1],
+         1 - 2 * qvec[1]**2 - 2 * qvec[2]**2]])
+
+
+def rotmat2qvec(R):
+    Rxx, Ryx, Rzx, Rxy, Ryy, Rzy, Rxz, Ryz, Rzz = R.flat
+    K = np.array([
+        [Rxx - Ryy - Rzz, 0, 0, 0],
+        [Ryx + Rxy, Ryy - Rxx - Rzz, 0, 0],
+        [Rzx + Rxz, Rzy + Ryz, Rzz - Rxx - Ryy, 0],
+        [Ryz - Rzy, Rzx - Rxz, Rxy - Ryx, Rxx + Ryy + Rzz]]) / 3.0
+    eigvals, eigvecs = np.linalg.eigh(K)
+    qvec = eigvecs[[3, 0, 1, 2], np.argmax(eigvals)]
+    if qvec[0] < 0:
+        qvec *= -1
+    return qvec
+

script.py

@@ -0,0 +1,145 @@
+### This is example of the script that will be run in the test environment.
+### Some parts of the code are compulsory and you should NOT CHANGE THEM.
+### They are between '''---compulsory---''' comments.
+### You can change the rest of the code to define and test your solution.
+### However, you should not change the signature of the provided function.
+### The script would save "submission.parquet" file in the current directory.
+### The actual logic of the solution is implemented in the `handcrafted_solution.py` file.
+### The `handcrafted_solution.py` file is a placeholder for your solution.
+### You should implement the logic of your solution in that file.
+### You can use any additional files and subdirectories to organize your code.
+
+'''---compulsory---'''
+import hoho; hoho.setup() # YOU MUST CALL hoho.setup() BEFORE ANYTHING ELSE
+import subprocess
+import importlib
+from pathlib import Path
+import subprocess
+
+
+### The function below is useful for installing additional python wheels.        
+def install_package_from_local_file(package_name, folder='packages'):
+    """
+    Installs a package from a local .whl file or a directory containing .whl files using pip.
+
+    Parameters:
+    path_to_file_or_directory (str): The path to the .whl file or the directory containing .whl files.
+    """
+    try:
+        pth = str(Path(folder) / package_name)
+        subprocess.check_call([subprocess.sys.executable, "-m", "pip", "install", 
+                               "--no-index",  # Do not use package index
+                               "--find-links", pth,  # Look for packages in the specified directory or at the file
+                               package_name])  # Specify the package to install
+        print(f"Package installed successfully from {pth}")
+    except subprocess.CalledProcessError as e:
+        print(f"Failed to install package from {pth}. Error: {e}")
+        
+
+# pip download webdataset -d packages/webdataset --platform manylinux1_x86_64 --python-version 38 --only-binary=:all:
+install_package_from_local_file('webdataset')
+# install_package_from_local_file('tqdm')
+
+### Here you can import any library or module you want.
+### The code below is used to read and parse the input dataset.
+### Please, do not modify it.
+
+import webdataset as wds
+from tqdm import tqdm
+from typing import Dict
+import pandas as pd
+from transformers import AutoTokenizer
+import os
+import time
+import io
+from read_write_colmap import read_cameras_binary, read_images_binary, read_points3D_binary
+from PIL import Image as PImage
+import numpy as np
+
+    
+def proc(row, split='train'):
+    out = {}
+    for k, v in row.items():
+        colname = k.split('.')[0]
+        if colname in {'ade20k', 'depthcm', 'gestalt'}:
+            if colname in out:
+                out[colname].append(v)
+            else:
+                out[colname] = [v]
+        elif colname in {'wireframe', 'mesh'}:
+            # out.update({a: b.tolist() for a,b in v.items()})
+            out.update({a: b for a,b in v.items()})
+        elif colname in 'kr':
+            out[colname.upper()] = v
+        else:
+            out[colname] = v
+            
+    return Sample(out)
+
+
+class Sample(Dict):
+    def __repr__(self):
+        return str({k: v.shape if hasattr(v, 'shape') else [type(v[0])] if isinstance(v, list) else type(v) for k,v in self.items()})
+
+def convert_entry_to_human_readable(entry):
+    out = {}
+    already_good = ['__key__', 'wf_vertices', 'wf_edges', 'edge_semantics', 'mesh_vertices', 'mesh_faces', 'face_semantics', 'K', 'R', 't']
+    for k, v in entry.items():
+        if k in already_good:
+            out[k] = v
+            continue
+        if k == 'points3d':
+            out[k] = read_points3D_binary(fid=io.BytesIO(v))
+        if k == 'cameras':
+            out[k] = read_cameras_binary(fid=io.BytesIO(v))
+        if k == 'images':
+            out[k] = read_images_binary(fid=io.BytesIO(v))
+        if k in ['ade20k', 'gestalt']:
+            out[k] =  [PImage.open(io.BytesIO(x)).convert('RGB') for x in v]
+        if k == 'depthcm':
+            out[k] = [PImage.open(io.BytesIO(x)) for x in entry['depthcm']]
+    return out
+
+'''---end of compulsory---'''
+
+def download_package(package_name, path_to_save='packages'):
+    """
+    Downloads a package using pip and saves it to a specified directory.
+
+    Parameters:
+    package_name (str): The name of the package to download.
+    path_to_save (str): The path to the directory where the package will be saved.
+    """
+    try:
+        # pip download webdataset -d packages/webdataset --platform manylinux1_x86_64 --python-version 38 --only-binary=:all:
+        subprocess.check_call([subprocess.sys.executable, "-m", "pip", "download", package_name, 
+                               "-d", str(Path(path_to_save)/package_name),  # Download the package to the specified directory
+                               "--platform", "manylinux1_x86_64",  # Specify the platform
+                               "--python-version", "38",  # Specify the Python version
+                               "--only-binary=:all:"])  # Download only binary packages
+        print(f'Package "{package_name}" downloaded successfully')
+    except subprocess.CalledProcessError as e:
+        print(f'Failed to downloaded package "{package_name}". Error: {e}')
+
+
+### The part below is used to define and test your solution.
+
+if __name__ == "__main__":
+    from handcrafted_solution import predict
+    print ("------------ Loading dataset------------ ")
+    params = hoho.get_params()
+    dataset = hoho.get_dataset(decode=None, split='all', dataset_type='webdataset')
+    print('------------ Now you can do your solution ---------------')
+    solution = []
+    for i, sample in enumerate(tqdm(dataset)):
+        pred_vertices, pred_edges, semantics = predict(sample, visualize=False)
+        solution.append({
+                        '__key__': sample['__key__'], 
+                        'wf_vertices': pred_vertices.tolist(),
+                        'wf_edges': pred_edges,
+                        'edge_semantics': semantics,
+                    })
+    print('------------ Saving results ---------------')
+    sub = pd.DataFrame(solution, columns=["__key__", "wf_vertices", "wf_edges", "edge_semantics"])
+    sub.to_parquet(Path(params['output_path']) / "submission.parquet")
+    print("------------ Done ------------ ")

viz3d.py

@@ -0,0 +1,302 @@
+
+"""
+Copyright [2022] [Paul-Edouard Sarlin and Philipp Lindenberger]
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License.
+
+3D visualization based on plotly.
+Works for a small number of points and cameras, might be slow otherwise.
+
+1) Initialize a figure with `init_figure`
+2) Add 3D points, camera frustums, or both as a pycolmap.Reconstruction
+
+Written by Paul-Edouard Sarlin and Philipp Lindenberger.
+"""
+# Slightly modified by Dmytro Mishkin
+
+from typing import Optional
+import numpy as np
+import pycolmap
+import plotly.graph_objects as go
+
+
+### Some helper functions for geometry
+def qvec2rotmat(qvec):
+    return np.array([
+        [1 - 2 * qvec[2]**2 - 2 * qvec[3]**2,
+         2 * qvec[1] * qvec[2] - 2 * qvec[0] * qvec[3],
+         2 * qvec[3] * qvec[1] + 2 * qvec[0] * qvec[2]],
+        [2 * qvec[1] * qvec[2] + 2 * qvec[0] * qvec[3],
+         1 - 2 * qvec[1]**2 - 2 * qvec[3]**2,
+         2 * qvec[2] * qvec[3] - 2 * qvec[0] * qvec[1]],
+        [2 * qvec[3] * qvec[1] - 2 * qvec[0] * qvec[2],
+         2 * qvec[2] * qvec[3] + 2 * qvec[0] * qvec[1],
+         1 - 2 * qvec[1]**2 - 2 * qvec[2]**2]])
+
+
+def to_homogeneous(points):
+    pad = np.ones((points.shape[:-1]+(1,)), dtype=points.dtype)
+    return np.concatenate([points, pad], axis=-1)
+
+def t_to_proj_center(qvec, tvec):
+    Rr = qvec2rotmat(qvec)
+    tt = (-Rr.T) @ tvec
+    return tt
+
+def calib(params):
+    out = np.eye(3)
+    if len(params) == 3:
+        out[0,0] = params[0]
+        out[1,1] = params[0]
+        out[0,2] = params[1]
+        out[1,2] = params[2]
+    else:
+        out[0,0] = params[0]
+        out[1,1] = params[1]
+        out[0,2] = params[2]
+        out[1,2] = params[3]
+    return out
+
+
+### Plotting functions
+
+def init_figure(height: int = 800) -> go.Figure:
+    """Initialize a 3D figure."""
+    fig = go.Figure()
+    axes = dict(
+        visible=False,
+        showbackground=False,
+        showgrid=False,
+        showline=False,
+        showticklabels=True,
+        autorange=True,
+    )
+    fig.update_layout(
+        template="plotly_dark",
+        height=height,
+        scene_camera=dict(
+            eye=dict(x=0., y=-.1, z=-2),
+            up=dict(x=0, y=-1., z=0),
+            projection=dict(type="orthographic")),
+        scene=dict(
+            xaxis=axes,
+            yaxis=axes,
+            zaxis=axes,
+            aspectmode='data',
+            dragmode='orbit',
+        ),
+        margin=dict(l=0, r=0, b=0, t=0, pad=0),
+        legend=dict(
+            orientation="h",
+            yanchor="top",
+            y=0.99,
+            xanchor="left",
+            x=0.1
+        ),
+    )
+    return fig
+
+
+def plot_lines_3d(
+        fig: go.Figure,
+        pts: np.ndarray,
+        color: str = 'rgba(255, 255, 255, 1)',
+        ps: int = 2,
+        colorscale: Optional[str] = None,
+        name: Optional[str] = None):
+    """Plot a set of 3D points."""
+    x = pts[..., 0]
+    y = pts[..., 1]
+    z = pts[..., 2]
+    traces = [go.Scatter3d(x=x1, y=y1, z=z1,
+                        mode='lines',
+                        line=dict(color=color, width=2)) for x1, y1, z1 in zip(x,y,z)]
+    for t in traces:
+        fig.add_trace(t)
+    fig.update_traces(showlegend=False)
+
+
+def plot_points(
+        fig: go.Figure,
+        pts: np.ndarray,
+        color: str = 'rgba(255, 0, 0, 1)',
+        ps: int = 2,
+        colorscale: Optional[str] = None,
+        name: Optional[str] = None):
+    """Plot a set of 3D points."""
+    x, y, z = pts.T
+    tr = go.Scatter3d(
+        x=x, y=y, z=z, mode='markers', name=name, legendgroup=name,
+        marker=dict(
+            size=ps, color=color, line_width=0.0, colorscale=colorscale))
+    fig.add_trace(tr)
+
+def plot_camera(
+        fig: go.Figure,
+        R: np.ndarray,
+        t: np.ndarray,
+        K: np.ndarray,
+        color: str = 'rgb(0, 0, 255)',
+        name: Optional[str] = None,
+        legendgroup: Optional[str] = None,
+        size: float = 1.0):
+    """Plot a camera frustum from pose and intrinsic matrix."""
+    W, H = K[0, 2]*2, K[1, 2]*2
+    corners = np.array([[0, 0], [W, 0], [W, H], [0, H], [0, 0]])
+    if size is not None:
+        image_extent = max(size * W / 1024.0, size * H / 1024.0)
+        world_extent = max(W, H) / (K[0, 0] + K[1, 1]) / 0.5
+        scale = 0.5 * image_extent / world_extent
+    else:
+        scale = 1.0
+    corners = to_homogeneous(corners) @ np.linalg.inv(K).T
+    corners = (corners / 2 * scale) @ R.T + t
+
+    x, y, z = corners.T
+    rect = go.Scatter3d(
+        x=x, y=y, z=z, line=dict(color=color), legendgroup=legendgroup,
+        name=name, marker=dict(size=0.0001), showlegend=False)
+    fig.add_trace(rect)
+
+    x, y, z = np.concatenate(([t], corners)).T
+    i = [0, 0, 0, 0]
+    j = [1, 2, 3, 4]
+    k = [2, 3, 4, 1]
+
+    pyramid = go.Mesh3d(
+        x=x, y=y, z=z, color=color, i=i, j=j, k=k,
+        legendgroup=legendgroup, name=name, showlegend=False)
+    fig.add_trace(pyramid)
+    triangles = np.vstack((i, j, k)).T
+    vertices = np.concatenate(([t], corners))
+    tri_points = np.array([
+        vertices[i] for i in triangles.reshape(-1)
+    ])
+
+    x, y, z = tri_points.T
+    pyramid = go.Scatter3d(
+        x=x, y=y, z=z, mode='lines', legendgroup=legendgroup,
+        name=name, line=dict(color=color, width=1), showlegend=False)
+    fig.add_trace(pyramid)
+
+
+def plot_camera_colmap(
+        fig: go.Figure,
+        image: pycolmap.Image,
+        camera: pycolmap.Camera,
+        name: Optional[str] = None,
+        **kwargs):
+    """Plot a camera frustum from PyCOLMAP objects"""
+    intr = calib(camera.params)
+    if intr[0][0] > 10000:
+        print("Bad camera")
+        return
+    plot_camera(
+        fig,
+        qvec2rotmat(image.qvec).T,
+        t_to_proj_center(image.qvec, image.tvec),
+        intr,#calibration_matrix(),
+        name=name or str(image.id),
+        **kwargs)
+
+
+def plot_cameras(
+        fig: go.Figure,
+        reconstruction,#: pycolmap.Reconstruction,
+        **kwargs):
+    """Plot a camera as a cone with camera frustum."""
+    for image_id, image in reconstruction["images"].items():
+        plot_camera_colmap(
+            fig, image, reconstruction["cameras"][image.camera_id], **kwargs)
+
+
+def plot_reconstruction(
+        fig: go.Figure,
+        rec,
+        color: str = 'rgb(0, 0, 255)',
+        name: Optional[str] = None,
+        points: bool = True,
+        cameras: bool = True,
+        cs: float = 1.0,
+        single_color_points=False,
+        camera_color='rgba(0, 255, 0, 0.5)'):
+    # rec is result of loading reconstruction from "read_write_colmap.py"
+    # Filter outliers
+    xyzs = []
+    rgbs = []
+    for k, p3D in rec['points'].items():
+        xyzs.append(p3D.xyz)
+        rgbs.append(p3D.rgb)
+
+    if points:
+        plot_points(fig, np.array(xyzs), color=color if single_color_points else np.array(rgbs), ps=1, name=name)
+    if cameras:
+        plot_cameras(fig, rec, color=camera_color, legendgroup=name, size=cs)
+
+
+def plot_pointcloud(
+        fig: go.Figure,
+        pts: np.ndarray,
+        colors: np.ndarray,
+        ps: int = 2,
+        name: Optional[str] = None):
+    """Plot a set of 3D points."""
+    plot_points(fig, np.array(pts), color=colors, ps=ps, name=name)
+
+
+def plot_triangle_mesh(
+        fig: go.Figure,
+        vert: np.ndarray,
+        colors: np.ndarray,
+        triangles: np.ndarray,
+        name: Optional[str] = None):
+    """Plot a triangle mesh."""
+    tr = go.Mesh3d(
+        x=vert[:,0],
+        y=vert[:,1],
+        z=vert[:,2],
+        vertexcolor = np.clip(255*colors, 0, 255),
+        # i, j and k give the vertices of triangles
+        # here we represent the 4 triangles of the tetrahedron surface
+        i=triangles[:,0],
+        j=triangles[:,1],
+        k=triangles[:,2],
+        name=name,
+        showscale=False
+    )
+    fig.add_trace(tr)
+
+def plot_estimate_and_gt(pred_vertices, pred_connections, gt_vertices=None, gt_connections=None):
+    fig3d = init_figure()
+    c1 = (30, 20, 255)
+    img_color = [c1 for _ in range(len(pred_vertices))]
+    plot_points(fig3d, pred_vertices, color = img_color, ps = 10)  
+    lines = []
+    for c in pred_connections:
+        v1 = pred_vertices[c[0]]
+        v2 = pred_vertices[c[1]]
+        lines.append(np.stack([v1, v2], axis=0))
+    plot_lines_3d(fig3d, np.array(lines), img_color, ps=4)  
+    if gt_vertices is not None:
+        c2 = (30, 255, 20)
+        img_color2 = [c2 for _ in range(len(gt_vertices))]
+        plot_points(fig3d, gt_vertices, color = img_color2, ps = 10)  
+        if gt_connections is not None:
+            gt_lines = []
+            for c in gt_connections:
+                v1 = gt_vertices[c[0]]
+                v2 = gt_vertices[c[1]]
+                gt_lines.append(np.stack([v1, v2], axis=0))
+        plot_lines_3d(fig3d, np.array(gt_lines), img_color2, ps=4)        
+    fig3d.show()
+    return fig3d