search

artyom-beilis / pytorch_dlprim

DLPrimitives/OpenCL out of tree backend for pytorch
http://blog.dlprimitives.org/
MIT License
227 stars 16 forks source link

python mnist.py --device ocl:0 runs on UHD Graphics 630, remains heavy load without progress after first epoch #34

Open rmast opened 1 year ago

rmast commented 1 year ago

The test with the following statement shows one epoch on the screen, and then seems to keep the GPU and CPU running whithout any further progress on the screen for at least 3 times the time of the first epoch:

Intel(R) Core(TM) i5-8500 CPU @ 3.00Ghz GPU 0 Intel(R) UHD Graphics 630 8GB RAM WSL2

python mnist.py --device ocl:0 

(dlprim) rmast@DESKTOP-9F227H8:/mnt/c/Users/nicor/pytorch_dlprim$ python mnist.py --device ocl:0
Downloading http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz
Downloading http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz to ../data/MNIST/raw/train-images-idx3-ubyte.gz
100.0%
Extracting ../data/MNIST/raw/train-images-idx3-ubyte.gz to ../data/MNIST/raw

Downloading http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz
Downloading http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz to ../data/MNIST/raw/train-labels-idx1-ubyte.gz
100.0%
Extracting ../data/MNIST/raw/train-labels-idx1-ubyte.gz to ../data/MNIST/raw

Downloading http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz
Downloading http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz to ../data/MNIST/raw/t10k-images-idx3-ubyte.gz
100.0%
Extracting ../data/MNIST/raw/t10k-images-idx3-ubyte.gz to ../data/MNIST/raw

Downloading http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz
Downloading http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz to ../data/MNIST/raw/t10k-labels-idx1-ubyte.gz
100.0%
Extracting ../data/MNIST/raw/t10k-labels-idx1-ubyte.gz to ../data/MNIST/raw

Using device: ocl:0
Accessing device #0:Intel(R) Graphics [0x3e92] on Intel(R) OpenCL HD Graphics
Train Epoch: 1 [0/60000 (0%)]   Loss: 2.326378
Train Epoch: 1 [640/60000 (1%)] Loss: 1.373419
Train Epoch: 1 [1280/60000 (2%)]        Loss: 0.674224
Train Epoch: 1 [1920/60000 (3%)]        Loss: 0.342615
Train Epoch: 1 [2560/60000 (4%)]        Loss: 0.282575
Train Epoch: 1 [3200/60000 (5%)]        Loss: 0.321835
Train Epoch: 1 [3840/60000 (6%)]        Loss: 0.117600
Train Epoch: 1 [4480/60000 (7%)]        Loss: 0.174954
Train Epoch: 1 [5120/60000 (9%)]        Loss: 0.295908
Train Epoch: 1 [5760/60000 (10%)]       Loss: 0.179200
Train Epoch: 1 [6400/60000 (11%)]       Loss: 0.148702
Train Epoch: 1 [7040/60000 (12%)]       Loss: 0.247692
Train Epoch: 1 [7680/60000 (13%)]       Loss: 0.097532
Train Epoch: 1 [8320/60000 (14%)]       Loss: 0.170208
Train Epoch: 1 [8960/60000 (15%)]       Loss: 0.098987
Train Epoch: 1 [9600/60000 (16%)]       Loss: 0.184404
Train Epoch: 1 [10240/60000 (17%)]      Loss: 0.095718
Train Epoch: 1 [10880/60000 (18%)]      Loss: 0.094362
Train Epoch: 1 [11520/60000 (19%)]      Loss: 0.054758
Train Epoch: 1 [12160/60000 (20%)]      Loss: 0.081427
Train Epoch: 1 [12800/60000 (21%)]      Loss: 0.137693
Train Epoch: 1 [13440/60000 (22%)]      Loss: 0.122686
Train Epoch: 1 [14080/60000 (23%)]      Loss: 0.039673
Train Epoch: 1 [14720/60000 (25%)]      Loss: 0.152234
Train Epoch: 1 [15360/60000 (26%)]      Loss: 0.080829
Train Epoch: 1 [16000/60000 (27%)]      Loss: 0.044828
Train Epoch: 1 [16640/60000 (28%)]      Loss: 0.146706
Train Epoch: 1 [17280/60000 (29%)]      Loss: 0.046113
Train Epoch: 1 [17920/60000 (30%)]      Loss: 0.054129
Train Epoch: 1 [18560/60000 (31%)]      Loss: 0.027078
Train Epoch: 1 [19200/60000 (32%)]      Loss: 0.034478
Train Epoch: 1 [19840/60000 (33%)]      Loss: 0.012288
Train Epoch: 1 [20480/60000 (34%)]      Loss: 0.026309
Train Epoch: 1 [21120/60000 (35%)]      Loss: 0.028055
Train Epoch: 1 [21760/60000 (36%)]      Loss: 0.127839
Train Epoch: 1 [22400/60000 (37%)]      Loss: 0.197114
Train Epoch: 1 [23040/60000 (38%)]      Loss: 0.065638
Train Epoch: 1 [23680/60000 (39%)]      Loss: 0.124214
Train Epoch: 1 [24320/60000 (41%)]      Loss: 0.109595
Train Epoch: 1 [24960/60000 (42%)]      Loss: 0.010281
Train Epoch: 1 [25600/60000 (43%)]      Loss: 0.033519
Train Epoch: 1 [26240/60000 (44%)]      Loss: 0.057112
Train Epoch: 1 [26880/60000 (45%)]      Loss: 0.102088
Train Epoch: 1 [27520/60000 (46%)]      Loss: 0.028558
Train Epoch: 1 [28160/60000 (47%)]      Loss: 0.049562
Train Epoch: 1 [28800/60000 (48%)]      Loss: 0.019591
Train Epoch: 1 [29440/60000 (49%)]      Loss: 0.062828
Train Epoch: 1 [30080/60000 (50%)]      Loss: 0.079009
Train Epoch: 1 [30720/60000 (51%)]      Loss: 0.082650
Train Epoch: 1 [31360/60000 (52%)]      Loss: 0.093156
Train Epoch: 1 [32000/60000 (53%)]      Loss: 0.101667
Train Epoch: 1 [32640/60000 (54%)]      Loss: 0.017085
Train Epoch: 1 [33280/60000 (55%)]      Loss: 0.053177
Train Epoch: 1 [33920/60000 (57%)]      Loss: 0.064748
Train Epoch: 1 [34560/60000 (58%)]      Loss: 0.090790
Train Epoch: 1 [35200/60000 (59%)]      Loss: 0.142080
Train Epoch: 1 [35840/60000 (60%)]      Loss: 0.063648
Train Epoch: 1 [36480/60000 (61%)]      Loss: 0.010347
Train Epoch: 1 [37120/60000 (62%)]      Loss: 0.126882
Train Epoch: 1 [37760/60000 (63%)]      Loss: 0.036586
Train Epoch: 1 [38400/60000 (64%)]      Loss: 0.011400
Train Epoch: 1 [39040/60000 (65%)]      Loss: 0.057160
Train Epoch: 1 [39680/60000 (66%)]      Loss: 0.040645
Train Epoch: 1 [40320/60000 (67%)]      Loss: 0.119776
Train Epoch: 1 [40960/60000 (68%)]      Loss: 0.106219
Train Epoch: 1 [41600/60000 (69%)]      Loss: 0.067625
Train Epoch: 1 [42240/60000 (70%)]      Loss: 0.053295
Train Epoch: 1 [42880/60000 (71%)]      Loss: 0.028393
Train Epoch: 1 [43520/60000 (72%)]      Loss: 0.007036
Train Epoch: 1 [44160/60000 (74%)]      Loss: 0.054127
Train Epoch: 1 [44800/60000 (75%)]      Loss: 0.130068
Train Epoch: 1 [45440/60000 (76%)]      Loss: 0.009325
Train Epoch: 1 [46080/60000 (77%)]      Loss: 0.046679
Train Epoch: 1 [46720/60000 (78%)]      Loss: 0.098600
Train Epoch: 1 [47360/60000 (79%)]      Loss: 0.007399
Train Epoch: 1 [48000/60000 (80%)]      Loss: 0.229577
Train Epoch: 1 [48640/60000 (81%)]      Loss: 0.099464
Train Epoch: 1 [49280/60000 (82%)]      Loss: 0.021609
Train Epoch: 1 [49920/60000 (83%)]      Loss: 0.018488
Train Epoch: 1 [50560/60000 (84%)]      Loss: 0.026598
Train Epoch: 1 [51200/60000 (85%)]      Loss: 0.052362
Train Epoch: 1 [51840/60000 (86%)]      Loss: 0.011622
Train Epoch: 1 [52480/60000 (87%)]      Loss: 0.066875
Train Epoch: 1 [53120/60000 (88%)]      Loss: 0.055762
Train Epoch: 1 [53760/60000 (90%)]      Loss: 0.100476
Train Epoch: 1 [54400/60000 (91%)]      Loss: 0.022489
Train Epoch: 1 [55040/60000 (92%)]      Loss: 0.006637
Train Epoch: 1 [55680/60000 (93%)]      Loss: 0.009578
Train Epoch: 1 [56320/60000 (94%)]      Loss: 0.148513
Train Epoch: 1 [56960/60000 (95%)]      Loss: 0.042047
Train Epoch: 1 [57600/60000 (96%)]      Loss: 0.126771
Train Epoch: 1 [58240/60000 (97%)]      Loss: 0.025688
Train Epoch: 1 [58880/60000 (98%)]      Loss: 0.007924
Train Epoch: 1 [59520/60000 (99%)]      Loss: 0.046730
Epoch in 101.9s

Schermafbeelding 2023-05-31 183531 image

What should I do to debug this?

(dlprim) rmast@DESKTOP-9F227H8:/mnt/c/Users/nicor/pytorch_dlprim/build$ pip list
Package            Version
------------------ ----------
certifi            2022.12.7
charset-normalizer 2.1.1
filelock           3.9.0
idna               3.4
Jinja2             3.1.2
MarkupSafe         2.1.2
mpmath             1.2.1
networkx           3.0
numpy              1.24.1
Pillow             9.3.0
pip                23.0.1
requests           2.28.1
setuptools         67.8.0
sympy              1.11.1
torch              2.0.1+cpu
torchaudio         2.0.2+cpu
torchvision        0.15.2+cpu
typing_extensions  4.4.0
urllib3            1.26.13
wheel              0.38.4
artyom-beilis commented 1 year ago

Interesting.

Is it consistent or once in a while?

rmast commented 1 year ago

Interesting.

Is it consistent or once in a while?

I repeated it, with same stall after the first epoch. By the way, the console is still responding, I can type CTRL-z to break up the process, and fg to make it continue. (dlprim) rmast@DESKTOP-9F227H8:/mnt/c/Users/nicor/pytorch_dlprim$ python mnist.py --device ocl:0 Using device: ocl:0 Accessing device #0:Intel(R) Graphics [0x3e92] on Intel(R) OpenCL HD Graphics Train Epoch: 1 [0/60000 (0%)] Loss: 2.326378 Train Epoch: 1 [640/60000 (1%)] Loss: 1.373419 Train Epoch: 1 [1280/60000 (2%)] Loss: 0.674224 Train Epoch: 1 [1920/60000 (3%)] Loss: 0.342615 Train Epoch: 1 [2560/60000 (4%)] Loss: 0.282575 Train Epoch: 1 [3200/60000 (5%)] Loss: 0.321835 Train Epoch: 1 [3840/60000 (6%)] Loss: 0.117600 Train Epoch: 1 [4480/60000 (7%)] Loss: 0.174962 Train Epoch: 1 [5120/60000 (9%)] Loss: 0.295924 Train Epoch: 1 [5760/60000 (10%)] Loss: 0.179236 Train Epoch: 1 [6400/60000 (11%)] Loss: 0.148775 Train Epoch: 1 [7040/60000 (12%)] Loss: 0.247576 Train Epoch: 1 [7680/60000 (13%)] Loss: 0.097563 Train Epoch: 1 [8320/60000 (14%)] Loss: 0.170751 Train Epoch: 1 [8960/60000 (15%)] Loss: 0.099251 Train Epoch: 1 [9600/60000 (16%)] Loss: 0.182786 Train Epoch: 1 [10240/60000 (17%)] Loss: 0.096495 Train Epoch: 1 [10880/60000 (18%)] Loss: 0.093181 Train Epoch: 1 [11520/60000 (19%)] Loss: 0.055978 Train Epoch: 1 [12160/60000 (20%)] Loss: 0.078291 Train Epoch: 1 [12800/60000 (21%)] Loss: 0.137493 Train Epoch: 1 [13440/60000 (22%)] Loss: 0.126942 Train Epoch: 1 [14080/60000 (23%)] Loss: 0.038443 Train Epoch: 1 [14720/60000 (25%)] Loss: 0.149018 Train Epoch: 1 [15360/60000 (26%)] Loss: 0.079035 Train Epoch: 1 [16000/60000 (27%)] Loss: 0.040937 Train Epoch: 1 [16640/60000 (28%)] Loss: 0.143583 Train Epoch: 1 [17280/60000 (29%)] Loss: 0.046299 Train Epoch: 1 [17920/60000 (30%)] Loss: 0.053709 Train Epoch: 1 [18560/60000 (31%)] Loss: 0.024080 Train Epoch: 1 [19200/60000 (32%)] Loss: 0.037789 Train Epoch: 1 [19840/60000 (33%)] Loss: 0.012120 Train Epoch: 1 [20480/60000 (34%)] Loss: 0.027062 Train Epoch: 1 [21120/60000 (35%)] Loss: 0.032783 Train Epoch: 1 [21760/60000 (36%)] Loss: 0.134422 Train Epoch: 1 [22400/60000 (37%)] Loss: 0.194787 Train Epoch: 1 [23040/60000 (38%)] Loss: 0.067409 Train Epoch: 1 [23680/60000 (39%)] Loss: 0.134988 Train Epoch: 1 [24320/60000 (41%)] Loss: 0.112529 Train Epoch: 1 [24960/60000 (42%)] Loss: 0.010880 Train Epoch: 1 [25600/60000 (43%)] Loss: 0.034358 Train Epoch: 1 [26240/60000 (44%)] Loss: 0.053919 Train Epoch: 1 [26880/60000 (45%)] Loss: 0.094899 Train Epoch: 1 [27520/60000 (46%)] Loss: 0.026722 Train Epoch: 1 [28160/60000 (47%)] Loss: 0.044900 Train Epoch: 1 [28800/60000 (48%)] Loss: 0.020568 Train Epoch: 1 [29440/60000 (49%)] Loss: 0.068913 Train Epoch: 1 [30080/60000 (50%)] Loss: 0.076643 Train Epoch: 1 [30720/60000 (51%)] Loss: 0.079894 Train Epoch: 1 [31360/60000 (52%)] Loss: 0.091840 Train Epoch: 1 [32000/60000 (53%)] Loss: 0.101678 Train Epoch: 1 [32640/60000 (54%)] Loss: 0.018253 Train Epoch: 1 [33280/60000 (55%)] Loss: 0.054471 Train Epoch: 1 [33920/60000 (57%)] Loss: 0.069801 Train Epoch: 1 [34560/60000 (58%)] Loss: 0.082842 Train Epoch: 1 [35200/60000 (59%)] Loss: 0.141856 Train Epoch: 1 [35840/60000 (60%)] Loss: 0.062433 Train Epoch: 1 [36480/60000 (61%)] Loss: 0.010398 Train Epoch: 1 [37120/60000 (62%)] Loss: 0.122779 Train Epoch: 1 [37760/60000 (63%)] Loss: 0.040562 Train Epoch: 1 [38400/60000 (64%)] Loss: 0.013098 Train Epoch: 1 [39040/60000 (65%)] Loss: 0.052796 Train Epoch: 1 [39680/60000 (66%)] Loss: 0.042598 Train Epoch: 1 [40320/60000 (67%)] Loss: 0.111036 Train Epoch: 1 [40960/60000 (68%)] Loss: 0.103284 Train Epoch: 1 [41600/60000 (69%)] Loss: 0.072419 Train Epoch: 1 [42240/60000 (70%)] Loss: 0.047269 Train Epoch: 1 [42880/60000 (71%)] Loss: 0.028031 Train Epoch: 1 [43520/60000 (72%)] Loss: 0.006540 Train Epoch: 1 [44160/60000 (74%)] Loss: 0.049276 Train Epoch: 1 [44800/60000 (75%)] Loss: 0.120560 Train Epoch: 1 [45440/60000 (76%)] Loss: 0.008151 Train Epoch: 1 [46080/60000 (77%)] Loss: 0.041024 Train Epoch: 1 [46720/60000 (78%)] Loss: 0.100522 Train Epoch: 1 [47360/60000 (79%)] Loss: 0.006663 Train Epoch: 1 [48000/60000 (80%)] Loss: 0.220928 Train Epoch: 1 [48640/60000 (81%)] Loss: 0.098380 Train Epoch: 1 [49280/60000 (82%)] Loss: 0.028299 Train Epoch: 1 [49920/60000 (83%)] Loss: 0.019357 Train Epoch: 1 [50560/60000 (84%)] Loss: 0.024031 Train Epoch: 1 [51200/60000 (85%)] Loss: 0.042200 Train Epoch: 1 [51840/60000 (86%)] Loss: 0.012437 Train Epoch: 1 [52480/60000 (87%)] Loss: 0.081118 Train Epoch: 1 [53120/60000 (88%)] Loss: 0.046438 Train Epoch: 1 [53760/60000 (90%)] Loss: 0.098449 Train Epoch: 1 [54400/60000 (91%)] Loss: 0.011161 Train Epoch: 1 [55040/60000 (92%)] Loss: 0.005813 Train Epoch: 1 [55680/60000 (93%)] Loss: 0.014051 Train Epoch: 1 [56320/60000 (94%)] Loss: 0.162412 Train Epoch: 1 [56960/60000 (95%)] Loss: 0.028086 Train Epoch: 1 [57600/60000 (96%)] Loss: 0.118508 Train Epoch: 1 [58240/60000 (97%)] Loss: 0.026278 Train Epoch: 1 [58880/60000 (98%)] Loss: 0.009992 Train Epoch: 1 [59520/60000 (99%)] Loss: 0.061088 Epoch in 98.4s ^C^Z [1]+ Stopped python mnist.py --device ocl:0 (dlprim) rmast@DESKTOP-9F227H8:/mnt/c/Users/nicor/pytorch_dlprim$ fg python mnist.py --device ocl:0 ^C^C^C^Z [1]+ Stopped python mnist.py --device ocl:0 (dlprim) rmast@DESKTOP-9F227H8:/mnt/c/Users/nicor/pytorch_dlprim$

skn123 commented 2 months ago

Here is what I have: Using device: privateuseone:0 failed to open /dev/dri/renderD128: Permission denied failed to open /dev/dri/renderD128: Permission denied Accessing device #0:Intel(R) Graphics [0x9a60] on Intel(R) OpenCL Graphics Train Epoch: 1 [0/60000 (0%)] Loss: 2.326378 Train Epoch: 1 [640/60000 (1%)] Loss: 1.373419 Train Epoch: 1 [1280/60000 (2%)] Loss: 0.674224 Train Epoch: 1 [1920/60000 (3%)] Loss: 0.342615 Train Epoch: 1 [2560/60000 (4%)] Loss: 0.282575 Train Epoch: 1 [3200/60000 (5%)] Loss: 0.321835 Train Epoch: 1 [3840/60000 (6%)] Loss: 0.117600 Train Epoch: 1 [4480/60000 (7%)] Loss: 0.174954 Train Epoch: 1 [5120/60000 (9%)] Loss: 0.295908 Train Epoch: 1 [5760/60000 (10%)] Loss: 0.179200 Train Epoch: 1 [6400/60000 (11%)] Loss: 0.148702 Train Epoch: 1 [7040/60000 (12%)] Loss: 0.247626 Train Epoch: 1 [7680/60000 (13%)] Loss: 0.097822 Train Epoch: 1 [8320/60000 (14%)] Loss: 0.170287 Train Epoch: 1 [8960/60000 (15%)] Loss: 0.099971 Train Epoch: 1 [9600/60000 (16%)] Loss: 0.182979 Train Epoch: 1 [10240/60000 (17%)] Loss: 0.097263 Train Epoch: 1 [10880/60000 (18%)] Loss: 0.094609 Train Epoch: 1 [11520/60000 (19%)] Loss: 0.054872 Train Epoch: 1 [12160/60000 (20%)] Loss: 0.079288 Train Epoch: 1 [12800/60000 (21%)] Loss: 0.141088 Train Epoch: 1 [13440/60000 (22%)] Loss: 0.122330 Train Epoch: 1 [14080/60000 (23%)] Loss: 0.039663 Train Epoch: 1 [14720/60000 (25%)] Loss: 0.150433 Train Epoch: 1 [15360/60000 (26%)] Loss: 0.078028 Train Epoch: 1 [16000/60000 (27%)] Loss: 0.040256 Train Epoch: 1 [16640/60000 (28%)] Loss: 0.144275 Train Epoch: 1 [17280/60000 (29%)] Loss: 0.044667 Train Epoch: 1 [17920/60000 (30%)] Loss: 0.056229 Train Epoch: 1 [18560/60000 (31%)] Loss: 0.024489 Train Epoch: 1 [19200/60000 (32%)] Loss: 0.035008 Train Epoch: 1 [19840/60000 (33%)] Loss: 0.013325 Train Epoch: 1 [20480/60000 (34%)] Loss: 0.029836 Train Epoch: 1 [21120/60000 (35%)] Loss: 0.028626 Train Epoch: 1 [21760/60000 (36%)] Loss: 0.129873 Train Epoch: 1 [22400/60000 (37%)] Loss: 0.201986 Train Epoch: 1 [23040/60000 (38%)] Loss: 0.072665 Train Epoch: 1 [23680/60000 (39%)] Loss: 0.133724 Train Epoch: 1 [24320/60000 (41%)] Loss: 0.106478 Train Epoch: 1 [24960/60000 (42%)] Loss: 0.009765 Train Epoch: 1 [25600/60000 (43%)] Loss: 0.035654 Train Epoch: 1 [26240/60000 (44%)] Loss: 0.059668 Train Epoch: 1 [26880/60000 (45%)] Loss: 0.098114 Train Epoch: 1 [27520/60000 (46%)] Loss: 0.026447 Train Epoch: 1 [28160/60000 (47%)] Loss: 0.046670 Train Epoch: 1 [28800/60000 (48%)] Loss: 0.016508 Train Epoch: 1 [29440/60000 (49%)] Loss: 0.068754 Train Epoch: 1 [30080/60000 (50%)] Loss: 0.087412 Train Epoch: 1 [30720/60000 (51%)] Loss: 0.077016 Train Epoch: 1 [31360/60000 (52%)] Loss: 0.087041 Train Epoch: 1 [32000/60000 (53%)] Loss: 0.102261 Train Epoch: 1 [32640/60000 (54%)] Loss: 0.017436 Train Epoch: 1 [33280/60000 (55%)] Loss: 0.058231 Train Epoch: 1 [33920/60000 (57%)] Loss: 0.058118 Train Epoch: 1 [34560/60000 (58%)] Loss: 0.089555 Train Epoch: 1 [35200/60000 (59%)] Loss: 0.139969 Train Epoch: 1 [35840/60000 (60%)] Loss: 0.057840 Train Epoch: 1 [36480/60000 (61%)] Loss: 0.010710 Train Epoch: 1 [37120/60000 (62%)] Loss: 0.129225 Train Epoch: 1 [37760/60000 (63%)] Loss: 0.039495 Train Epoch: 1 [38400/60000 (64%)] Loss: 0.012222 Train Epoch: 1 [39040/60000 (65%)] Loss: 0.062294 Train Epoch: 1 [39680/60000 (66%)] Loss: 0.041331 Train Epoch: 1 [40320/60000 (67%)] Loss: 0.108825 Train Epoch: 1 [40960/60000 (68%)] Loss: 0.106947 Train Epoch: 1 [41600/60000 (69%)] Loss: 0.062064 Train Epoch: 1 [42240/60000 (70%)] Loss: 0.054333 Train Epoch: 1 [42880/60000 (71%)] Loss: 0.032988 Train Epoch: 1 [43520/60000 (72%)] Loss: 0.006049 Train Epoch: 1 [44160/60000 (74%)] Loss: 0.056357 Train Epoch: 1 [44800/60000 (75%)] Loss: 0.131841 Train Epoch: 1 [45440/60000 (76%)] Loss: 0.009063 Train Epoch: 1 [46080/60000 (77%)] Loss: 0.044287 Train Epoch: 1 [46720/60000 (78%)] Loss: 0.100934 Train Epoch: 1 [47360/60000 (79%)] Loss: 0.009730 Train Epoch: 1 [48000/60000 (80%)] Loss: 0.234216 Train Epoch: 1 [48640/60000 (81%)] Loss: 0.106466 Train Epoch: 1 [49280/60000 (82%)] Loss: 0.022423 Train Epoch: 1 [49920/60000 (83%)] Loss: 0.019978 Train Epoch: 1 [50560/60000 (84%)] Loss: 0.025209 Train Epoch: 1 [51200/60000 (85%)] Loss: 0.045058 Train Epoch: 1 [51840/60000 (86%)] Loss: 0.010271 Train Epoch: 1 [52480/60000 (87%)] Loss: 0.063850 Train Epoch: 1 [53120/60000 (88%)] Loss: 0.063364 Train Epoch: 1 [53760/60000 (90%)] Loss: 0.093897 Train Epoch: 1 [54400/60000 (91%)] Loss: 0.017038 Train Epoch: 1 [55040/60000 (92%)] Loss: 0.008403 Train Epoch: 1 [55680/60000 (93%)] Loss: 0.008805 Train Epoch: 1 [56320/60000 (94%)] Loss: 0.175549 Train Epoch: 1 [56960/60000 (95%)] Loss: 0.028869 Train Epoch: 1 [57600/60000 (96%)] Loss: 0.123760 Train Epoch: 1 [58240/60000 (97%)] Loss: 0.028526 Train Epoch: 1 [58880/60000 (98%)] Loss: 0.013072 Train Epoch: 1 [59520/60000 (99%)] Loss: 0.053476 Epoch in 84.0s

Test set: Average loss: 0.0480, Accuracy: 9823/10000 (98%)

Train Epoch: 2 [0/60000 (0%)] Loss: 0.065459 Train Epoch: 2 [640/60000 (1%)] Loss: 0.039748 Train Epoch: 2 [1280/60000 (2%)] Loss: 0.026082 Train Epoch: 2 [1920/60000 (3%)] Loss: 0.040584 Train Epoch: 2 [2560/60000 (4%)] Loss: 0.112989 Train Epoch: 2 [3200/60000 (5%)] Loss: 0.052024 Train Epoch: 2 [3840/60000 (6%)] Loss: 0.014717 Train Epoch: 2 [4480/60000 (7%)] Loss: 0.080200 Train Epoch: 2 [5120/60000 (9%)] Loss: 0.008801 Train Epoch: 2 [5760/60000 (10%)] Loss: 0.021775 Train Epoch: 2 [6400/60000 (11%)] Loss: 0.033641 Train Epoch: 2 [7040/60000 (12%)] Loss: 0.038709 Train Epoch: 2 [7680/60000 (13%)] Loss: 0.023366 Train Epoch: 2 [8320/60000 (14%)] Loss: 0.010033 Train Epoch: 2 [8960/60000 (15%)] Loss: 0.007349 Train Epoch: 2 [9600/60000 (16%)] Loss: 0.007722 Train Epoch: 2 [10240/60000 (17%)] Loss: 0.034887 Train Epoch: 2 [10880/60000 (18%)] Loss: 0.108077 Train Epoch: 2 [11520/60000 (19%)] Loss: 0.025160 Train Epoch: 2 [12160/60000 (20%)] Loss: 0.054884 Train Epoch: 2 [12800/60000 (21%)] Loss: 0.036236 Train Epoch: 2 [13440/60000 (22%)] Loss: 0.004143 Train Epoch: 2 [14080/60000 (23%)] Loss: 0.006044 Train Epoch: 2 [14720/60000 (25%)] Loss: 0.005873 Train Epoch: 2 [15360/60000 (26%)] Loss: 0.056684 Train Epoch: 2 [16000/60000 (27%)] Loss: 0.097826 Train Epoch: 2 [16640/60000 (28%)] Loss: 0.048662 Train Epoch: 2 [17280/60000 (29%)] Loss: 0.014356 Train Epoch: 2 [17920/60000 (30%)] Loss: 0.043007 Train Epoch: 2 [18560/60000 (31%)] Loss: 0.002331 Train Epoch: 2 [19200/60000 (32%)] Loss: 0.017479 Train Epoch: 2 [19840/60000 (33%)] Loss: 0.009449 Train Epoch: 2 [20480/60000 (34%)] Loss: 0.006558 Train Epoch: 2 [21120/60000 (35%)] Loss: 0.005025 Train Epoch: 2 [21760/60000 (36%)] Loss: 0.009748 Train Epoch: 2 [22400/60000 (37%)] Loss: 0.122231 Train Epoch: 2 [23040/60000 (38%)] Loss: 0.004964 Train Epoch: 2 [23680/60000 (39%)] Loss: 0.043455 Train Epoch: 2 [24320/60000 (41%)] Loss: 0.019358 Train Epoch: 2 [24960/60000 (42%)] Loss: 0.015154 Train Epoch: 2 [25600/60000 (43%)] Loss: 0.048404 Train Epoch: 2 [26240/60000 (44%)] Loss: 0.025492 Train Epoch: 2 [26880/60000 (45%)] Loss: 0.097685 Train Epoch: 2 [27520/60000 (46%)] Loss: 0.012076 Train Epoch: 2 [28160/60000 (47%)] Loss: 0.002270 Train Epoch: 2 [28800/60000 (48%)] Loss: 0.006972 Train Epoch: 2 [29440/60000 (49%)] Loss: 0.012000 Train Epoch: 2 [30080/60000 (50%)] Loss: 0.000439 Train Epoch: 2 [30720/60000 (51%)] Loss: 0.202667 Train Epoch: 2 [31360/60000 (52%)] Loss: 0.053927 Train Epoch: 2 [32000/60000 (53%)] Loss: 0.025709 Train Epoch: 2 [32640/60000 (54%)] Loss: 0.049223 Train Epoch: 2 [33280/60000 (55%)] Loss: 0.018739 Train Epoch: 2 [33920/60000 (57%)] Loss: 0.038373 Train Epoch: 2 [34560/60000 (58%)] Loss: 0.005439 Train Epoch: 2 [35200/60000 (59%)] Loss: 0.046203 Train Epoch: 2 [35840/60000 (60%)] Loss: 0.028460 Train Epoch: 2 [36480/60000 (61%)] Loss: 0.171114 Train Epoch: 2 [37120/60000 (62%)] Loss: 0.026590 Train Epoch: 2 [37760/60000 (63%)] Loss: 0.041375 Train Epoch: 2 [38400/60000 (64%)] Loss: 0.009419 Train Epoch: 2 [39040/60000 (65%)] Loss: 0.002045 Train Epoch: 2 [39680/60000 (66%)] Loss: 0.002927 Train Epoch: 2 [40320/60000 (67%)] Loss: 0.003417 Train Epoch: 2 [40960/60000 (68%)] Loss: 0.027682 Train Epoch: 2 [41600/60000 (69%)] Loss: 0.081617 Train Epoch: 2 [42240/60000 (70%)] Loss: 0.001047 Train Epoch: 2 [42880/60000 (71%)] Loss: 0.002091 Train Epoch: 2 [43520/60000 (72%)] Loss: 0.037615 Train Epoch: 2 [44160/60000 (74%)] Loss: 0.019530 Train Epoch: 2 [44800/60000 (75%)] Loss: 0.068913 Train Epoch: 2 [45440/60000 (76%)] Loss: 0.037312 Train Epoch: 2 [46080/60000 (77%)] Loss: 0.006535 Train Epoch: 2 [46720/60000 (78%)] Loss: 0.006351 Train Epoch: 2 [47360/60000 (79%)] Loss: 0.115172 Train Epoch: 2 [48000/60000 (80%)] Loss: 0.018656 Train Epoch: 2 [48640/60000 (81%)] Loss: 0.042368 Train Epoch: 2 [49280/60000 (82%)] Loss: 0.043556 Train Epoch: 2 [49920/60000 (83%)] Loss: 0.056863 Train Epoch: 2 [50560/60000 (84%)] Loss: 0.032384 Train Epoch: 2 [51200/60000 (85%)] Loss: 0.115363 Train Epoch: 2 [51840/60000 (86%)] Loss: 0.007574 Train Epoch: 2 [52480/60000 (87%)] Loss: 0.009734 Train Epoch: 2 [53120/60000 (88%)] Loss: 0.007060 Train Epoch: 2 [53760/60000 (90%)] Loss: 0.003769 Train Epoch: 2 [54400/60000 (91%)] Loss: 0.032285 Train Epoch: 2 [55040/60000 (92%)] Loss: 0.014053 Train Epoch: 2 [55680/60000 (93%)] Loss: 0.021713 Train Epoch: 2 [56320/60000 (94%)] Loss: 0.014905 Train Epoch: 2 [56960/60000 (95%)] Loss: 0.002904 Train Epoch: 2 [57600/60000 (96%)] Loss: 0.047537 Train Epoch: 2 [58240/60000 (97%)] Loss: 0.015722 Train Epoch: 2 [58880/60000 (98%)] Loss: 0.023946 Train Epoch: 2 [59520/60000 (99%)] Loss: 0.107267 Epoch in 81.5s

Test set: Average loss: 0.0351, Accuracy: 9890/10000 (99%)

Train Epoch: 3 [0/60000 (0%)] Loss: 0.010208 Train Epoch: 3 [640/60000 (1%)] Loss: 0.004863 Train Epoch: 3 [1280/60000 (2%)] Loss: 0.083801 Train Epoch: 3 [1920/60000 (3%)] Loss: 0.052472 Train Epoch: 3 [2560/60000 (4%)] Loss: 0.022324 Train Epoch: 3 [3200/60000 (5%)] Loss: 0.002020 Train Epoch: 3 [3840/60000 (6%)] Loss: 0.010708 Train Epoch: 3 [4480/60000 (7%)] Loss: 0.008449 Train Epoch: 3 [5120/60000 (9%)] Loss: 0.008116 Train Epoch: 3 [5760/60000 (10%)] Loss: 0.057926 Train Epoch: 3 [6400/60000 (11%)] Loss: 0.004955 Train Epoch: 3 [7040/60000 (12%)] Loss: 0.005344 Train Epoch: 3 [7680/60000 (13%)] Loss: 0.011234 Train Epoch: 3 [8320/60000 (14%)] Loss: 0.035252 Train Epoch: 3 [8960/60000 (15%)] Loss: 0.009050 Train Epoch: 3 [9600/60000 (16%)] Loss: 0.007375 Train Epoch: 3 [10240/60000 (17%)] Loss: 0.001962 Train Epoch: 3 [10880/60000 (18%)] Loss: 0.026312 Train Epoch: 3 [11520/60000 (19%)] Loss: 0.003460 Train Epoch: 3 [12160/60000 (20%)] Loss: 0.003715 Train Epoch: 3 [12800/60000 (21%)] Loss: 0.115187 Train Epoch: 3 [13440/60000 (22%)] Loss: 0.002534 Train Epoch: 3 [14080/60000 (23%)] Loss: 0.007834 Train Epoch: 3 [14720/60000 (25%)] Loss: 0.004763 Train Epoch: 3 [15360/60000 (26%)] Loss: 0.015894 Train Epoch: 3 [16000/60000 (27%)] Loss: 0.002579 Train Epoch: 3 [16640/60000 (28%)] Loss: 0.002418 Train Epoch: 3 [17280/60000 (29%)] Loss: 0.025393 Train Epoch: 3 [17920/60000 (30%)] Loss: 0.007640 Train Epoch: 3 [18560/60000 (31%)] Loss: 0.041093 Train Epoch: 3 [19200/60000 (32%)] Loss: 0.068456 Train Epoch: 3 [19840/60000 (33%)] Loss: 0.013836 Train Epoch: 3 [20480/60000 (34%)] Loss: 0.002818 Train Epoch: 3 [21120/60000 (35%)] Loss: 0.075637 Train Epoch: 3 [21760/60000 (36%)] Loss: 0.005217 Train Epoch: 3 [22400/60000 (37%)] Loss: 0.002275 Train Epoch: 3 [23040/60000 (38%)] Loss: 0.008789 Train Epoch: 3 [23680/60000 (39%)] Loss: 0.007001 Train Epoch: 3 [24320/60000 (41%)] Loss: 0.002952 Train Epoch: 3 [24960/60000 (42%)] Loss: 0.013566 Train Epoch: 3 [25600/60000 (43%)] Loss: 0.005279 Train Epoch: 3 [26240/60000 (44%)] Loss: 0.043639 Train Epoch: 3 [26880/60000 (45%)] Loss: 0.000604 Train Epoch: 3 [27520/60000 (46%)] Loss: 0.007317 Train Epoch: 3 [28160/60000 (47%)] Loss: 0.055156 Train Epoch: 3 [28800/60000 (48%)] Loss: 0.022920 Train Epoch: 3 [29440/60000 (49%)] Loss: 0.005157 Train Epoch: 3 [30080/60000 (50%)] Loss: 0.002840 Train Epoch: 3 [30720/60000 (51%)] Loss: 0.007870 Train Epoch: 3 [31360/60000 (52%)] Loss: 0.009656 Train Epoch: 3 [32000/60000 (53%)] Loss: 0.110479 Train Epoch: 3 [32640/60000 (54%)] Loss: 0.002230 Train Epoch: 3 [33280/60000 (55%)] Loss: 0.050295 Train Epoch: 3 [33920/60000 (57%)] Loss: 0.007495 Train Epoch: 3 [34560/60000 (58%)] Loss: 0.026453 Train Epoch: 3 [35200/60000 (59%)] Loss: 0.003770 Train Epoch: 3 [35840/60000 (60%)] Loss: 0.004537 Train Epoch: 3 [36480/60000 (61%)] Loss: 0.006647 Train Epoch: 3 [37120/60000 (62%)] Loss: 0.009076 Train Epoch: 3 [37760/60000 (63%)] Loss: 0.096330 Train Epoch: 3 [38400/60000 (64%)] Loss: 0.078154 Train Epoch: 3 [39040/60000 (65%)] Loss: 0.021905 Train Epoch: 3 [39680/60000 (66%)] Loss: 0.003132 Train Epoch: 3 [40320/60000 (67%)] Loss: 0.000418 Train Epoch: 3 [40960/60000 (68%)] Loss: 0.004076 Train Epoch: 3 [41600/60000 (69%)] Loss: 0.001671 Train Epoch: 3 [42240/60000 (70%)] Loss: 0.009720 Train Epoch: 3 [42880/60000 (71%)] Loss: 0.006832 Train Epoch: 3 [43520/60000 (72%)] Loss: 0.043676

Continues without any issue; CPU is way faster that this in-built GPU.

artyom-beilis commented 2 months ago

Built-in Intel graphics card is typically weaker that CPU itself on modern processors. Typical is ~400GFLOPS while modern multi-core CPU can do much more.

So while built-in GPU supported it does not mean it gives an advantage.

skn123 commented 2 months ago

Absoluetly! and I have to intention of using Intel GPU. I merely wanted to test the process and the best I can do on WSL is the Intel GPU. I have to test it on a AMD GPU along with a Zeon processor.