search

thtrieu / darkflow

Translate darknet to tensorflow. Load trained weights, retrain/fine-tune using tensorflow, export constant graph def to mobile devices
GNU General Public License v3.0
6.13k stars 2.08k forks source link

Forwarding 8 inputs ... after running this its completed and training/epoch not started. #1111

Closed ankitAMD closed 4 years ago

ankitAMD commented 4 years ago

I am using Ubuntu16.04 I take class=1 and filter=30 i execute this command ....................................

       python flow --model cfg/Yolo-2c.cfg --load bin/yolo.weights --annotation new_model_data/annotations --dataset new_model_data/images --epoch 40

But output comes at last paragraaph

    Running entirely on CPU
    WARNING:tensorflow:From /home/ankitgupta/Music/Darkflow-object-detection-master/darkflow/net/build.py:145: The name tf.Session is deprecated. Please use tf.compat.v1.Session instead.

    2019-12-27 15:02:28.819061: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
     2019-12-27 15:02:28.842552: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU           Frequency: 2394565000 Hz
    2019-12-27 15:02:28.842860: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x4f26f80 executing computations on platform Host. Devices:
   2019-12-27 15:02:28.842892: I tensorflow/compiler/xla/service/service.cc:175]     StreamExecutor device (0): <undefined>, <undefined>
   2019-12-27 15:02:29.039800: W tensorflow/compiler/jit/mark_for_compilation_pass.cc:1412]  (One-time warning): Not using XLA:CPU for cluster because envvar TF_XLA_FLAGS=--tf_xla_cpu_global_jit was not set.  If you want XLA:CPU, either set that envvar, or use experimental_jit_scope to enable XLA:CPU.  To confirm that XLA is active, pass --vmodule=xla_compilation_cache=1 (as a proper command-line flag, not via TF_XLA_FLAGS) or set the envvar XLA_FLAGS=--xla_hlo_profile.
       Finished in 1.8025741577148438s

    Forwarding 8 inputs ...
    Total time = 1.0950102806091309s / 8 inps = 7.305867480577233 ips
    Post processing 8 inputs ...
   Total time = 0.3878042697906494s / 8 inps = 20.628963173403648 ips

My main concerned is everything work fine but training or epoch not started .............but why ???

ankitAMD commented 4 years ago

Sorry i written wrong command .....................

         python flow --model cfg/Yolo-2c.cfg --load bin/yolo.weights --annotation new_model_data/annotations --dataset new_model_data/images --epoch 40

change to this (right command)......................

         python flow  --model cfg/yolo-1c.cfg --load bin/yolo.weights --train --annotation new_model_data/annotations --dataset new_model_data/images  --epoch 40 --savepb

"--savepb " is used for freezing the graph and saving the .pb extension file.