mitkina
commented
5 years ago I should have been using tf-gpu 1.9.0!
Closed mitkina closed 5 years ago
mitkina
commented
5 years ago I should have been using tf-gpu 1.9.0!
Hello,
This may be a silly question - but I've been trying to run the model with CUDA 9.0, CUDNN 7.0.5, tf-gpu-1.5.0, python 3.5, and I get the below error when running train.py with the kth dataset. Do you have any insight as to what might be causing the error?
lib/python3.5/site-packages/tensorflow/python/framework/tensor_util.py", line 498, in make_tensor_proto str_values = [compat.as_bytes(x) for x in proto_values] File "/home/video_prediction/python3_env/lib/python3.5/site-packages/tensorflow/python/framework/tensor_util.py", line 498, in str_values = [compat.as_bytes(x) for x in proto_values] File "/home/video_prediction/python3_env/lib/python3.5/site-packages/tensorflow/python/util/compat.py", line 65, in as_bytes (bytes_or_text,)) TypeError: Expected binary or unicode string, got {'eval_gen_images_ssim/min': <tf.Tensor 'eval_outputs_and_metrics/zeros_like_6:0' shape=(10, 16, 64, 64, 3) dtype=float32$, 'eval_vgg_csim/sum': <tf.Tensor 'eval_outputs_and_metrics/zeros_5:0' shape=(10, 16) dtype=float32>, 'eval_ssim_finn/max': <tf.Tensor 'eval_outputs_and_metrics/Fill_9:0' $hape=(10, 16) dtype=float32>, 'eval_gen_images_ssim_scikit/sum': <tf.Tensor 'eval_outputs_and_metrics/zeros_like_10:0' shape=(10, 16, 64, 64, 3) dtype=float32>, 'eval_ssim$finn/sum': <tf.Tensor 'eval_outputs_and_metrics/zeros_4:0' shape=(10, 16) dtype=float32>, 'eval_ssim_finn/min': <tf.Tensor 'eval_outputs_and_metrics/Fill_8:0' shape=(10, 1$) dtype=float32>, 'eval_gen_images_ssim_finn/min': <tf.Tensor 'eval_outputs_and_metrics/zeros_like_12:0' shape=(10, 16, 64, 64, 3) dtype=float32>, 'eval_ssim_scikit/sum': $tf.Tensor 'eval_outputs_and_metrics/zeros_3:0' shape=(10, 16) dtype=float32>, 'eval_mse/max': <tf.Tensor 'eval_outputs_and_metrics/Fill_3:0' shape=(10, 16) dtype=float32>, 'eval_gen_images_ssim_finn/sum': <tf.Tensor 'eval_outputs_and_metrics/zeros_like_13:0' shape=(10, 16, 64, 64, 3) dtype=float32>, 'eval_ssim_scikit/max': <tf.Tensor 'eval_o$tputs_and_metrics/Fill_7:0' shape=(10, 16) dtype=float32>, 'eval_ssim/min': <tf.Tensor 'eval_outputs_and_metrics/Fill_4:0' shape=(10, 16) dtype=float32>, 'eval_genimages$gg_cdist/sum': <tf.Tensor 'eval_outputs_and_metrics/zeros_like_19:0' shape=(10, 16, 64, 64, 3) dtype=float32>, 'eval_gen_images_ssim/max': <tf.Tensor 'eval_outputs_and_met$ics/zeros_like_8:0' shape=(10, 16, 64, 64, 3) dtype=float32>, 'eval_gen_images_ssim_scikit/min': <tf.Tensor 'eval_outputs_and_metrics/zeros_like_9:0' shape=(10, 16, 64, 64$ 3) dtype=float32>, 'eval_gen_images_ssim_finn/max': <tf.Tensor 'eval_outputs_and_metrics/zeros_like_14:0' shape=(10, 16, 64, 64, 3) dtype=float32>, 'eval_gen_images_mse/m$x': <tf.Tensor 'eval_outputs_and_metrics/zeros_like_5:0' shape=(10, 16, 64, 64, 3) dtype=float32>, 'eval_vgg_cdist/max': <tf.Tensor 'eval_outputs_and_metrics/Fill_13:0' sh$pe=(10, 16) dtype=float32>, 'eval_vgg_csim/min': <tf.Tensor 'eval_outputs_and_metrics/Fill_10:0' shape=(10, 16) dtype=float32>, 'eval_psnr/sum': <tf.Tensor 'eval_outputs_a$d_metrics/zeros:0' shape=(10, 16) dtype=float32>, 'eval_vgg_cdist/sum': <tf.Tensor 'eval_outputs_and_metrics/zeros_6:0' shape=(10, 16) dtype=float32>, 'eval_mse/sum': <tf.$ensor 'eval_outputs_and_metrics/zeros_1:0' shape=(10, 16) dtype=float32>, 'eval_gen_images_vgg_csim/min': <tf.Tensor 'eval_outputs_and_metrics/zeros_like_15:0' shape=(10, $6, 64, 64, 3) dtype=float32>, 'eval_gen_images_ssim/sum': <tf.Tensor 'eval_outputs_and_metrics/zeros_like_7:0' shape=(10, 16, 64, 64, 3) dtype=float32>, 'eval_gen_images_p$nr/sum': <tf.Tensor 'eval_outputs_and_metrics/zeros_like_1:0' shape=(10, 16, 64, 64, 3) dtype=float32>, 'eval_gen_images_vgg_csim/max': <tf.Tensor 'eval_outputs_and_metric$/zeros_like_17:0' shape=(10, 16, 64, 64, 3) dtype=float32>, 'eval_gen_images_psnr/max': <tf.Tensor 'eval_outputs_and_metrics/zeros_like_2:0' shape=(10, 16, 64, 64, 3) dtyp$=float32>, 'eval_ssim/sum': <tf.Tensor 'eval_outputs_and_metrics/zeros_2:0' shape=(10, 16) dtype=float32>, 'eval_gen_images_vgg_cdist/max': <tf.Tensor 'eval_outputs_and_me$rics/zeros_like_20:0' shape=(10, 16, 64, 64, 3) dtype=float32>, 'eval_gen_images_mse/min': <tf.Tensor 'eval_outputs_and_metrics/zeros_like_3:0' shape=(10, 16, 64, 64, 3) d$ype=float32>, 'eval_psnr/min': <tf.Tensor 'eval_outputs_and_metrics/Fill:0' shape=(10, 16) dtype=float32>, 'eval_psnr/max': <tf.Tensor 'eval_outputs_and_metrics/Fill_1:0' $hape=(10, 16) dtype=float32>, 'eval_vgg_cdist/min': <tf.Tensor 'eval_outputs_and_metrics/Fill_12:0' shape=(10, 16) dtype=float32>, 'eval_ssim/max': <tf.Tensor 'eval_outputs_and_metrics/Fill_5:0' shape=(10, 16) dtype=float32>, 'eval_gen_images_vgg_csim/sum': <tf.Tensor 'eval_outputs_and_metrics/zeros_like_16:0' shape=(10, 16, 64, 64, 3) dtype=float32>, 'eval_gen_images_ssim_scikit/max': <tf.Tensor 'eval_outputs_and_metrics/zeros_like_11:0' shape=(10, 16, 64, 64, 3) dtype=float32>, 'eval_gen_images_vgg_cdist/min': <tf.Tensor 'eval_outputs_and_metrics/zeros_like_18:0' shape=(10, 16, 64, 64, 3) dtype=float32>, 'eval_vgg_csim/max': <tf.Tensor 'eval_outputs_and_metrics/Fill_11:0' shape=(10, 16) dtype=float32>, 'eval_gen_images_psnr/min': <tf.Tensor 'eval_outputs_and_metrics/zeros_like:0' shape=(10, 16, 64, 64, 3) dtype=float32>, 'eval_mse/min': <tf.Tens$r 'eval_outputs_and_metrics/Fill_2:0' shape=(10, 16) dtype=float32>, 'eval_gen_images_mse/sum': <tf.Tensor 'eval_outputs_and_metrics/zeros_like_4:0' shape=(10, 16, 64, 64, 3) dtype=float32>, 'eval_ssim_scikit/min': <tf.Tensor 'eval_outputs_and_metrics/Fill_6:0' shape=(10, 16) dtype=float32>}