Guten Tag,
ich bitte um Hilfe für folgendes Problem:
nach folgender Anleitung habe ich in mühevoller Kleinarbeit
cHOW TO TRAIN YOUR OWN LC0 NET - THE GUIDE FOR WINDOWS 10
https://docs.google.com/document/d/1pB2MtqVVgYC2BwJbIrVvP2xWe5KIFFVteEqZroMf20E/editinstalliert. Leider habe ich keine bzw. wenige Vorkentnisse in Python , C usw., ich versuche mich step by step in dieses Thema einzuarbeiten, was mit 70 Jahren nicht immer ganz so einfach ist. Beim Start wird folgende Fehlermeldung angezeigt:
(pyvenv) C:\Users\Wolfgang\lczero-training\tf>python train.py --cfg configs/test.yaml
dataset:
allow_less_chunks: true
experimental_v5_only_dataset: false
input: /Users/wolfgang/Lczero-training/T60data/T60prepareddata/
num_chunks: 100000
test_workers: 32
train_ratio: 0.9
train_workers: 64
gpu: 0
model:
filters: 64
moves_left: none
policy: convolution
residual_blocks: 6
se_ratio: 4
value: wdl
name: ex-64x6_test
training:
batch_size: 4096
checkpoint_steps: 10000
lr_boundaries:
- 100000
- 130000
lr_values:
- 0.02
- 0.002
- 0.0005
mask_legal_moves: true
moves_left_loss_weight: 0.0
num_batch_splits: 8
num_test_positions: 10000
path: /Users/wolfgang/Lczero-training/T60data/mynetwork
policy_loss_weight: 1.0
precision: single
q_ratio: 0.0
renorm: false
shuffle_size: 200000
swa: true
swa_max_n: 10
swa_output: true
swa_steps: 20
test_steps: 2000
total_steps: 140000
train_avg_report_steps: 200
validation_steps: 2000
value_loss_weight: 1.0
warmup_steps: 250
got 1 chunks for /Users/wolfgang/Lczero-training/T60data/T60prepareddata/
sorting 1 chunks...[done]
training.284466207.gz - training.284466207.gz
Using 64 worker processes.
Using 32 worker processes.
chunk_reader didn't find any chunks.
2023-02-23 08:55:05.466576: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
2023-02-23 08:55:07.867222: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library nvcuda.dll
2023-02-23 08:55:07.907248: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1555] Found device 0 with properties:
pciBusID: 0000:01:00.0 name: NVIDIA GeForce RTX 3060 Ti computeCapability: 8.6
coreClock: 1.755GHz coreCount: 38 deviceMemorySize: 8.00GiB deviceMemoryBandwidth: 417.29GiB/s
2023-02-23 08:55:07.907411: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
2023-02-23 08:55:08.112327: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cublas64_10.dll
2023-02-23 08:55:08.364056: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cufft64_10.dll
2023-02-23 08:55:08.369247: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library curand64_10.dll
2023-02-23 08:55:08.394949: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusolver64_10.dll
2023-02-23 08:55:08.411513: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusparse64_10.dll
2023-02-23 08:55:29.757714: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudnn64_7.dll
2023-02-23 08:55:29.757903: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1697] Adding visible gpu devices: 0
[PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU')]
2023-02-23 08:55:29.759094: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
2023-02-23 08:55:29.760174: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1555] Found device 0 with properties:
pciBusID: 0000:01:00.0 name: NVIDIA GeForce RTX 3060 Ti computeCapability: 8.6
coreClock: 1.755GHz coreCount: 38 deviceMemorySize: 8.00GiB deviceMemoryBandwidth: 417.29GiB/s
2023-02-23 08:55:29.760376: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
2023-02-23 08:55:29.760499: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cublas64_10.dll
2023-02-23 08:55:29.760629: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cufft64_10.dll
2023-02-23 08:55:29.760767: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library curand64_10.dll
2023-02-23 08:55:29.760916: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusolver64_10.dll
2023-02-23 08:55:29.761051: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusparse64_10.dll
2023-02-23 08:55:29.761189: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudnn64_7.dll
2023-02-23 08:55:29.761332: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1697] Adding visible gpu devices: 0
2023-02-23 08:55:31.383928: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1096] Device interconnect StreamExecutor with strength 1 edge matrix:
2023-02-23 08:55:31.384071: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1102] 0
2023-02-23 08:55:31.384214: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] 0: N
2023-02-23 08:55:31.384415: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1241] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 6680 MB memory) -> physical GPU (device: 0, name: NVIDIA GeForce RTX 3060 Ti, pci bus id: 0000:01:00.0, compute capability: 8.6)
Using 19 evaluation batches
Traceback (most recent call last):
File "train.py", line 257, in <module>
main(argparser.parse_args())
File "train.py", line 234, in main
batch_splits=batch_splits)
File "C:\Users\Wolfgang\lczero-training\tf\tfprocess.py", line 625, in process_loop
self.process(batch_size, test_batches, batch_splits=batch_splits)
File "C:\Users\Wolfgang\lczero-training\tf\tfprocess.py", line 808, in process
with tf.profiler.experimental.Trace("Test", step_num=steps + 1):
AttributeError: module 'tensorflow' has no attribute 'profiler'
(pyvenv) C:\Users\Wolfgang\lczero-training\tf>
Würde mich sehr freuen wenn mir von den alten Hasen jemand helfen könnte , gerne auch per TeamViewer.
VG, Wolfgang Remin
Forum
CSS-Shop