Unable to download the pre-trained models

[t-naveenkumar]

I am unable to download trained models using download.sh. I am getting following error

`Resolving s3.us-east-2.amazonaws.com (s3.us-east-2.amazonaws.com)... 52.219.104.210 Connecting to s3.us-east-2.amazonaws.com (s3.us-east-2.amazonaws.com)|52.219.104.210|:443... connected. WARNING: cannot verify s3.us-east-2.amazonaws.com's certificate, issued by â/C=US/ST=California/O=Zscaler Inc./OU=Zscaler Inc./CN=Zscaler Intermediate Root CA (zscloud.net) (t) â: Unable to locally verify the issuer's authority. HTTP request sent, awaiting response... 403 Forbidden 2019-03-08 09:44:03 ERROR 403: Forbidden.

tar (child): weights.tar.gz: Cannot open: No such file or directory tar (child): Error is not recoverable: exiting now tar: Child returned status 2 tar: Error is not recoverable: exiting now`

please let me know, if you need further information to address this issue

[PiaCuk]

I get the same error:

~/flownet2-tf/checkpoints$ . download.sh 
--2019-03-14 14:33:33--  https://s3.us-east-2.amazonaws.com/flownetdata/weights.tar.gz
Resolving s3.us-east-2.amazonaws.com (s3.us-east-2.amazonaws.com)... 52.219.80.106
Connecting to s3.us-east-2.amazonaws.com (s3.us-east-2.amazonaws.com)|52.219.80.106|:443... connected.
HTTP request sent, awaiting response... 403 Forbidden
2019-03-14 14:33:33 ERROR 403: Forbidden.

tar (child): weights.tar.gz: Cannot open: No such file or directory
tar (child): Error is not recoverable: exiting now
tar: Child returned status 2
tar: Error is not recoverable: exiting now

[Iamanorange]

@t-naveenkumar @PiaCuk https://github.com/sampepose/flownet2-tf/issues/62

[PiaCuk]

Thank you for the quick response!

[PiaCuk]

Look at @Iamanorange 's answer, there is a backup for downloading the weights.

[lixiaolusunshine]

i got the same erro please tell me how to download the fiels?

[Iamanorange]

https://github.com/sampepose/flownet2-tf/issues/62

[lixiaolusunshine]

I am unable to download trained models using download.sh. I am getting following error

`Resolving s3.us-east-2.amazonaws.com (s3.us-east-2.amazonaws.com)... 52.219.104.210 Connecting to s3.us-east-2.amazonaws.com (s3.us-east-2.amazonaws.com)|52.219.104.210|:443... connected. WARNING: cannot verify s3.us-east-2.amazonaws.com's certificate, issued by â/C=US/ST=California/O=Zscaler Inc./OU=Zscaler Inc./CN=Zscaler Intermediate Root CA (zscloud.net) (t) â: Unable to locally verify the issuer's authority. HTTP request sent, awaiting response... 403 Forbidden 2019-03-08 09:44:03 ERROR 403: Forbidden.

tar (child): weights.tar.gz: Cannot open: No such file or directory tar (child): Error is not recoverable: exiting now tar: Child returned status 2 tar: Error is not recoverable: exiting now`

please let me know, if you need further information to address this issue I have encountered the same problem, have you solved it ? I can't run this code

[Iamanorange]

作者给的AWS挂掉了，我做了两个备份，你可以从这两个备份下载训练好的权重。 https://drive.google.com/file/d/1cft8EvnsBL5w4-REUeAaVWLeRx39hyHE/view?usp=sharing https://mega.nz/#!xk5DVYwB!hG2u_3ho8LWJl660Z5e0z-C0SlytB4SvlSkuoLEd568

[LiuzhuForFun]

@Iamanorange Could you please solve some bugs of my codes？Or Can you give me the code you test the performance of KITTI.Thanks a lot for your contribution of the research on optical flow.I modified the code from DFNet,but I can't get the same result the paper reported.My code like this:` from future import division import cv2 import tensorflow as tf import numpy as np import os import PIL.Image as pil import png import scipy from ..training_schedules import LONG_SCHEDULE

from .flownet_c import FlowNetC from ..flowlib import flow_to_image, write_flow from scipy.misc import imread, imsave import uuid flags = tf.app.flags flags.DEFINE_integer("batch_size", 1, "The size of of a sample batch") flags.DEFINE_integer("img_height", 384, "Image height") flags.DEFINE_integer("img_width", 1280, "Image width") flags.DEFINE_string("dataset_dir", '', "Dataset directory") flags.DEFINE_string("output_dir", None, "Output directory") flags.DEFINE_string("ckpt_file", '', "checkpoint file") FLAGS = flags.FLAGS

FLOW_SCALE = 5.0

if 'train' in FLAGS.dataset_dir: NUM = 194 elif 'test' in FLAGS.dataset_dir: NUM = 195

def get_flow(path): bgr = cv2.imread(path, cv2.IMREAD_ANYCOLOR | cv2.IMREAD_ANYDEPTH) invalid = bgr[:, :, 0] == 0 out_flow = (bgr[:, :, 2:0:-1].astype('f4') - 2**15) / 64. out_flow[invalid] = 0 return out_flow, bgr[:, :, 0]

def compute_flow_error(gt_flow, predflow, mask): H, W, = gt_flow.shape print(pred_flow.shape) old_H, oldW, = pred_flow.shape

# Reshape predicted flow to have same size as ground truth
pred0 = cv2.resize(pred_flow[:,:,0], (W, H), interpolation=cv2.INTER_LINEAR) * (1.0*W/old_W)
pred1 = cv2.resize(pred_flow[:,:,1], (W, H), interpolation=cv2.INTER_LINEAR) * (1.0*H/old_H)
pred = np.stack((pred0, pred1), axis=-1) * FLOW_SCALE

err = np.sqrt(np.sum(np.square(gt_flow - pred), axis=-1))
err_valid = np.sum(err * mask) / np.sum(mask)
return err_valid, pred

def write_flowpng(name, flow): H, W, = flow.shape out = np.ones((H, W, 3), dtype=np.uint64) out[:,:,1] = np.minimum(np.maximum(flow[:,:,1]64.+215, 0), 216).astype(np.uint64) out[:,:,0] = np.minimum(np.maximum(flow[:,:,0]64.+215, 0), 216).astype(np.uint64) with open(name, 'wb') as f: writer = png.Writer(width=W, height=H, bitdepth=16) im2list = out.reshape(-1, out.shape[1]*out.shape[2]).tolist() writer.write(f, im2list)

def pick_frame(path): new_files = []

flow2012 dataset only has 194 pairs

for i in range(NUM):
    frame1 = os.path.join(path, 'colored_0', '{:06d}'.format(i) + '_10.png')
    frame2 = os.path.join(path, 'colored_0', '{:06d}'.format(i) + '_11.png')
    new_files.append([frame1, frame2])
return new_files

def main(_): new_files = pick_frame(FLAGS.dataset_dir) basename = os.path.basename(FLAGS.ckpt_file)

im1_pl = tf.placeholder(dtype=tf.float32, shape=(1, FLAGS.img_height, FLAGS.img_width, 3))
im2_pl = tf.placeholder(dtype=tf.float32, shape=(1, FLAGS.img_height, FLAGS.img_width, 3))
model =FlowNetC()
inputs = {
    'input_a': im1_pl,
    'input_b': im2_pl,
}
pred_flows = model.model(inputs, training_schedule=LONG_SCHEDULE)

saver = tf.train.Saver([var for var in tf.all_variables()])
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
errs = np.zeros(NUM)

if not FLAGS.output_dir is None and not os.path.exists(FLAGS.output_dir):
    os.makedirs(FLAGS.output_dir)

with tf.Session(config=config) as sess:
    saver.restore(sess, FLAGS.ckpt_file)
    # For val set
    for t in range(0, len(new_files)):
        if t % 100 == 0:
            print('processing %s: %d/%d' % (basename, t, len(new_files)))
        raw_im0 = pil.open(new_files[t][0])
        raw_im1 = pil.open(new_files[t][1])
        scaled_im0 = raw_im0.resize((FLAGS.img_width, FLAGS.img_height), pil.ANTIALIAS)
        scaled_im1 = raw_im1.resize((FLAGS.img_width, FLAGS.img_height), pil.ANTIALIAS)

        scaled_im0 = (np.expand_dims(np.array(scaled_im0), axis=0).astype(np.float32))/255.
        scaled_im1 = (np.expand_dims(np.array(scaled_im1), axis=0).astype(np.float32))/255.
        feed_dict = {im1_pl: scaled_im0, im2_pl: scaled_im1}
        pred_flows_val = sess.run(pred_flows, feed_dict=feed_dict)
        pred_flow_val = pred_flows_val['flow']

        if 'train' in FLAGS.dataset_dir:
            # no occlusion
            #gt_flow, mask = get_flow(new_files[t][0].replace('colored_0', 'flow_noc'))
            # all
            gt_flow, mask = get_flow(new_files[t][0].replace('colored_0', 'flow_occ'))
            errs[t], scaled_pred = compute_flow_error(gt_flow, pred_flow_val[0,:,:,:], mask)
            unique_name = 'flow-' + str(uuid.uuid4())
            flow_img = flow_to_image(pred_flow_val[0,:,:,:])
            full_out_path = os.path.join(unique_name + '.png')
            imsave(full_out_path, flow_img)

        # Save for eval
        if 'test' in FLAGS.dataset_dir:
            _, scaled_pred = compute_flow_error(np.array(raw_im0)[:,:,:2], pred_flow_val[0,:,:,:], np.array(raw_im0)[:,:,0])
            png_name = os.path.join(FLAGS.output_dir, new_files[t][0].split('/')[-1])
            write_flow_png(png_name, scaled_pred)

        # Save for visual colormap
        if not 'test' in FLAGS.dataset_dir and not FLAGS.output_dir is None:
            # flow_im = flow_to_image(scaled_pred)
            png_name = os.path.join(FLAGS.output_dir, new_files[t][0].split('/')[-1]).replace('png', 'jpg')
            # cv2.imwrite(png_name, flow_im[:,:,::-1])

    print('{:>10}'.format('(valid) endpoint error'))
    print('{:10.4f}'.format(errs.mean()))

if name == 'main': tf.app.run()`

[XiaowanLi2018]

作者给的AWS挂掉了，我做了两个备份，你可以从这两个备份下载训练好的权重。 https://drive.google.com/file/d/1cft8EvnsBL5w4-REUeAaVWLeRx39hyHE/view?usp=sharing https://mega.nz/#!xk5DVYwB!hG2u_3ho8LWJl660Z5e0z-C0SlytB4SvlSkuoLEd568

您好，这两个权重备份是不是打包方式有问题，我下载之后显示无法解压，报错info如下 Archive: checkpoint.zip End-of-central-directory signature not found. Either this file is not a zipfile, or it constitutes one disk of a multi-part archive. In the latter case the central directory and zipfile comment will be found on the last disk(s) of this archive. unzip: cannot find zipfile directory in one of checkpoint.zip or checkpoint.zip.zip, and cannot find checkpoint.zip.ZIP, period.

[a726]

作者给的AWS挂掉了，我做了两个备份，你可以从这两个备份下载训练好的权重。 https://drive.google.com/file/d/1cft8EvnsBL5w4-REUeAaVWLeRx39hyHE/view?usp=sharing https://mega.nz/#!xk5DVYwB!hG2u_3ho8LWJl660Z5e0z-C0SlytB4SvlSkuoLEd568

您好，这两个权重备份好像失效了，您能不能给我分享一份，非常感谢。QQ306835421