Tensorflow 2 on Metal - Apple Silicon M4 Max 12/4 40 core GPU #32

obriensystems · 2024-11-29T21:27:34Z

Batch Size variations among GPUs (shorter time per iteration is better)

(venv-metal) michaelobrien@mbp8 src % cat tflow.py                  
import tensorflow as tf
#import keras
#from keras.utils import multi_gpu_model
#import keras.backend as k
#https://github.com/microsoft/tensorflow-directml/issues/352

# https://www.tensorflow.org/guide/distributed_training
#
# https://www.tensorflow.org/tutorials/distribute/keras
# https://keras.io/guides/distributed_training/
#strategy = tf.distribute.MirroredStrategy()
#print('Number of devices: {}'.format(strategy.num_replicas_in_sync))

#NUM_GPUS = 2
#strategy = tf.contrib.distribute.MirroredStrategy()#num_gpus=NUM_GPUS)
# working on dual RTX-4090
#strategy = tf.distribute.MirroredStrategy(devices=["/gpu:0", "/gpu:1"])
#WARNING:tensorflow:Some requested devices in `tf.distribute.Strategy` are not visible to TensorFlow: /replica:0/task:0/device:GPU:1,/replica:0/task:0/device:GPU:0
#Number of devices: 2


#central_storage_strategy = tf.distribute.experimental.CentralStorageStrategy()
#strategy = tf.distribute.MultiWorkerMirroredStrategy() # not in tf 1.5
#print("mirrored_strategy: ",mirrored_strategy)
strategy = tf.distribute.OneDeviceStrategy(device="/gpu:0")
#mirrored_strategy = tf.distribute.MirroredStrategy(devices=["/gpu:0","/gpu:1"],cross_device_ops=tf.contrib.distribute.AllReduceCrossDeviceOps(all_reduce_alg="hierarchical_copy"))
#mirrored_strategy = tf.distribute.MirroredStrategy(devices= ["/gpu:0","/gpu:1"],cross_device_ops=tf.distribute.HierarchicalCopyAllReduce())

#print('Number of devices: {}'.format(strategy.num_replicas_in_sync))

# https://learn.microsoft.com/en-us/windows/ai/directml/gpu-faq
#a = tf.constant([1.])
#b = tf.constant([2.])
#c = tf.add(a, b)

#gpu_config = tf.GPUOptions()
#gpu_config.visible_device_list = "1"#"0,1"
#gpu_config.visible_device_list = "0,1"
#gpu_config.allow_growth=True

#session = tf.Session(config=tf.ConfigProto(gpu_options=gpu_config))
#print(session.run(c))
#tensorflow.python.framework.errors_impl.AlreadyExistsError: TensorFlow device (DML:0) is being mapped to multiple DML devices (0 now, and 1 previously), which is not supported. This may be the result of providing different GPU configurations (ConfigProto.gpu_options, for example different visible_device_list) when creating multiple Sessions in the same process. This is not  currently supported, see https://github.com/tensorflow/tensorflow/issues/19083
#from keras import backend as K
#K.set_session(session)

cifar = tf.keras.datasets.cifar100
(x_train, y_train), (x_test, y_test) = cifar.load_data()

with strategy.scope():
# https://www.tensorflow.org/api_docs/python/tf/keras/applications/resnet50/ResNet50
# https://keras.io/api/models/model/
  parallel_model = tf.keras.applications.ResNet50(
#model = tf.keras.applications.ResNet50(
    include_top=True,
    weights=None,
    input_shape=(32, 32, 3),
    classes=100,)
# https://saturncloud.io/blog/how-to-do-multigpu-training-with-keras/  
  #parallel_model = multi_gpu_model(model, gpus=2)
  loss_fn = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=False)
# https://keras.io/api/models/model_training_apis/
  parallel_model.compile(optimizer="adam", loss=loss_fn, metrics=["accuracy"])
parallel_model.fit(x_train, y_train, epochs=25, batch_size=512)#5120)#7168)#7168)

By typing 'agree' you are agreeing to the terms of the software license agreements. Any other response will cancel. [agree, cancel]
agree
You can review the license in Xcode’s About window, or at: /Applications/Xcode.app/Contents/Resources/en.lproj/License.rtf
michaelobrien@Michaels-MBP ObrienlabsDev % git clone [email protected]:ObrienlabsDev/machine-learning.git
Cloning into 'machine-learning'...
The authenticity of host 'github.com (140.82.113.4)' can't be established.
ED25519 key fingerprint is SHA256:+DiY3wvvV6TuJJhbpZisF/zLDA0zPMSvHdkr4UvCOqU.
This key is not known by any other names.
Are you sure you want to continue connecting (yes/no/[fingerprint])? yes
Warning: Permanently added 'github.com' (ED25519) to the list of known hosts.
remote: Enumerating objects: 193, done.
remote: Counting objects: 100% (3/3), done.
remote: Compressing objects: 100% (3/3), done.
remote: Total 193 (delta 0), reused 0 (delta 0), pack-reused 190 (from 1)
Receiving objects: 100% (193/193), 57.62 KiB | 1.20 MiB/s, done.
Resolving deltas: 100% (91/91), done.
michaelobrien@Michaels-MBP ObrienlabsDev % cd machine-learning 
michaelobrien@Michaels-MBP machine-learning % ls
LICENSE		README.md	build.sh	cuda		environments	gpus.md		links.md	stats.md
michaelobrien@Michaels-MBP machine-learning % /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"
==> Checking for `sudo` access (which may request your password)...
Password:
==> This script will install:
/opt/homebrew/bin/brew
/opt/homebrew/share/doc/homebrew
/opt/homebrew/share/man/man1/brew.1
/opt/homebrew/share/zsh/site-functions/_brew
/opt/homebrew/etc/bash_completion.d/brew
/opt/homebrew
==> The following new directories will be created:
/opt/homebrew/bin
/opt/homebrew/etc
/opt/homebrew/include
/opt/homebrew/lib
/opt/homebrew/sbin
/opt/homebrew/share
/opt/homebrew/var
/opt/homebrew/opt
/opt/homebrew/share/zsh
/opt/homebrew/share/zsh/site-functions
/opt/homebrew/var/homebrew
/opt/homebrew/var/homebrew/linked
/opt/homebrew/Cellar
/opt/homebrew/Caskroom
/opt/homebrew/Frameworks

Press RETURN/ENTER to continue or any other key to abort:
==> /usr/bin/sudo /usr/bin/install -d -o root -g wheel -m 0755 /opt/homebrew
==> /usr/bin/sudo /bin/mkdir -p /opt/homebrew/bin /opt/homebrew/etc /opt/homebrew/include /opt/homebrew/lib /opt/homebrew/sbin /opt/homebrew/share /opt/homebrew/var /opt/homebrew/opt /opt/homebrew/share/zsh /opt/homebrew/share/zsh/site-functions /opt/homebrew/var/homebrew /opt/homebrew/var/homebrew/linked /opt/homebrew/Cellar /opt/homebrew/Caskroom /opt/homebrew/Frameworks
==> /usr/bin/sudo /bin/chmod ug=rwx /opt/homebrew/bin /opt/homebrew/etc /opt/homebrew/include /opt/homebrew/lib /opt/homebrew/sbin /opt/homebrew/share /opt/homebrew/var /opt/homebrew/opt /opt/homebrew/share/zsh /opt/homebrew/share/zsh/site-functions /opt/homebrew/var/homebrew /opt/homebrew/var/homebrew/linked /opt/homebrew/Cellar /opt/homebrew/Caskroom /opt/homebrew/Frameworks
==> /usr/bin/sudo /bin/chmod go-w /opt/homebrew/share/zsh /opt/homebrew/share/zsh/site-functions
==> /usr/bin/sudo /usr/sbin/chown michaelobrien /opt/homebrew/bin /opt/homebrew/etc /opt/homebrew/include /opt/homebrew/lib /opt/homebrew/sbin /opt/homebrew/share /opt/homebrew/var /opt/homebrew/opt /opt/homebrew/share/zsh /opt/homebrew/share/zsh/site-functions /opt/homebrew/var/homebrew /opt/homebrew/var/homebrew/linked /opt/homebrew/Cellar /opt/homebrew/Caskroom /opt/homebrew/Frameworks
==> /usr/bin/sudo /usr/bin/chgrp admin /opt/homebrew/bin /opt/homebrew/etc /opt/homebrew/include /opt/homebrew/lib /opt/homebrew/sbin /opt/homebrew/share /opt/homebrew/var /opt/homebrew/opt /opt/homebrew/share/zsh /opt/homebrew/share/zsh/site-functions /opt/homebrew/var/homebrew /opt/homebrew/var/homebrew/linked /opt/homebrew/Cellar /opt/homebrew/Caskroom /opt/homebrew/Frameworks
==> /usr/bin/sudo /usr/sbin/chown -R michaelobrien:admin /opt/homebrew
==> /usr/bin/sudo /bin/mkdir -p /Users/michaelobrien/Library/Caches/Homebrew
==> /usr/bin/sudo /bin/chmod g+rwx /Users/michaelobrien/Library/Caches/Homebrew
==> /usr/bin/sudo /usr/sbin/chown -R michaelobrien /Users/michaelobrien/Library/Caches/Homebrew
==> Downloading and installing Homebrew...
remote: Enumerating objects: 286588, done.
remote: Counting objects: 100% (161/161), done.
remote: Compressing objects: 100% (126/126), done.
remote: Total 286588 (delta 35), reused 152 (delta 29), pack-reused 286427 (from 1)
remote: Enumerating objects: 55, done.
remote: Counting objects: 100% (33/33), done.
remote: Total 55 (delta 33), reused 33 (delta 33), pack-reused 22 (from 1)
==> Updating Homebrew...
==> Downloading https://ghcr.io/v2/homebrew/portable-ruby/portable-ruby/blobs/sha256:303bed4c7fc431a685db3c3c151d873740114adbdccd23762ea2d1e39ea78f47
######################################################################################################################################################################################################### 100.0%
==> Pouring portable-ruby-3.3.6.arm64_big_sur.bottle.tar.gz
Warning: /opt/homebrew/bin is not in your PATH.
  Instructions on how to configure your shell for Homebrew
  can be found in the 'Next steps' section below.
==> Installation successful!

==> Homebrew has enabled anonymous aggregate formulae and cask analytics.
Read the analytics documentation (and how to opt-out) here:
  https://docs.brew.sh/Analytics
No analytics data has been sent yet (nor will any be during this install run).

==> Homebrew is run entirely by unpaid volunteers. Please consider donating:
  https://github.com/Homebrew/brew#donations

==> Next steps:
- Run these commands in your terminal to add Homebrew to your PATH:
    echo >> /Users/michaelobrien/.zprofile
    echo 'eval "$(/opt/homebrew/bin/brew shellenv)"' >> /Users/michaelobrien/.zprofile
    eval "$(/opt/homebrew/bin/brew shellenv)"
- Run brew help to get started
- Further documentation:
    https://docs.brew.sh

michaelobrien@Michaels-MBP machine-learning % echo >> /Users/michaelobrien/.zprofile
michaelobrien@Michaels-MBP machine-learning % echo 'eval "$(/opt/homebrew/bin/brew shellenv)"' >> /Users/michaelobrien/.zprofile
michaelobrien@Michaels-MBP machine-learning % eval "$(/opt/homebrew/bin/brew shellenv)"
michaelobrien@Michaels-MBP machine-learning % brew install go
==> Downloading https://formulae.brew.sh/api/formula.jws.json
######################################################################################################################################################################################################### 100.0%
==> Downloading https://formulae.brew.sh/api/cask.jws.json
######################################################################################################################################################################################################### 100.0%
==> Downloading https://ghcr.io/v2/homebrew/core/go/manifests/1.23.3
######################################################################################################################################################################################################### 100.0%
==> Fetching go
==> Downloading https://ghcr.io/v2/homebrew/core/go/blobs/sha256:1bbc1e16a0048f6d42a0522361eded589d4efeda3e2bc7527a3ca5bc65e8d7e7
######################################################################################################################################################################################################### 100.0%
==> Pouring go--1.23.3.arm64_sequoia.bottle.tar.gz
==> Caveats
Homebrew's Go toolchain is configured with
  GOTOOLCHAIN=local
per Homebrew policy on tools that update themselves.
==> Summary
🍺  /opt/homebrew/Cellar/go/1.23.3: 13,235 files, 268.2MB
==> Running `brew cleanup go`...
Disable this behaviour by setting HOMEBREW_NO_INSTALL_CLEANUP.
Hide these hints with HOMEBREW_NO_ENV_HINTS (see `man brew`).
michaelobrien@Michaels-MBP machine-learning % brew install jq
==> Downloading https://ghcr.io/v2/homebrew/core/jq/manifests/1.7.1-1
######################################################################################################################################################################################################### 100.0%
==> Fetching dependencies for jq: oniguruma
==> Downloading https://ghcr.io/v2/homebrew/core/oniguruma/manifests/6.9.9
######################################################################################################################################################################################################### 100.0%
==> Fetching oniguruma
==> Downloading https://ghcr.io/v2/homebrew/core/oniguruma/blobs/sha256:378cec2d71c33ca25809db43e9fdd46957b3459fcd5b337c519fc74cc2cfdc20
######################################################################################################################################################################################################### 100.0%
==> Fetching jq
==> Downloading https://ghcr.io/v2/homebrew/core/jq/blobs/sha256:a10c82b07e393869d4467ad3e8ba26346d026b1ad3533d31dbb5e72abe9a7968
######################################################################################################################################################################################################### 100.0%
==> Installing dependencies for jq: oniguruma
==> Installing jq dependency: oniguruma
==> Downloading https://ghcr.io/v2/homebrew/core/oniguruma/manifests/6.9.9
Already downloaded: /Users/michaelobrien/Library/Caches/Homebrew/downloads/35140c4d3995b75388bed026ef6d0acbb4d6076047cdcd895bfd996c0c8d6487--oniguruma-6.9.9.bottle_manifest.json
==> Pouring oniguruma--6.9.9.arm64_sequoia.bottle.tar.gz
🍺  /opt/homebrew/Cellar/oniguruma/6.9.9: 15 files, 1.4MB
==> Installing jq
==> Pouring jq--1.7.1.arm64_sequoia.bottle.1.tar.gz
🍺  /opt/homebrew/Cellar/jq/1.7.1: 20 files, 1.4MB
==> Running `brew cleanup jq`...
Disable this behaviour by setting HOMEBREW_NO_INSTALL_CLEANUP.
Hide these hints with HOMEBREW_NO_ENV_HINTS (see `man brew`).
michaelobrien@Michaels-MBP machine-learning % brew install kubernetes-cli
==> Downloading https://ghcr.io/v2/homebrew/core/kubernetes-cli/manifests/1.31.3
######################################################################################################################################################################################################### 100.0%
==> Fetching kubernetes-cli
==> Downloading https://ghcr.io/v2/homebrew/core/kubernetes-cli/blobs/sha256:302d2ecf90f6c962e96b6345d4031551ec0198c4db1c1cfb0ea83b87510ec6cd
######################################################################################################################################################################################################### 100.0%
==> Pouring kubernetes-cli--1.31.3.arm64_sequoia.bottle.tar.gz
==> Caveats
zsh completions have been installed to:
  /opt/homebrew/share/zsh/site-functions
==> Summary
🍺  /opt/homebrew/Cellar/kubernetes-cli/1.31.3: 237 files, 60.2MB
==> Running `brew cleanup kubernetes-cli`...
Disable this behaviour by setting HOMEBREW_NO_INSTALL_CLEANUP.
Hide these hints with HOMEBREW_NO_ENV_HINTS (see `man brew`).
michaelobrien@Michaels-MBP machine-learning % kubectl version
Client Version: v1.31.3
Kustomize Version: v5.4.2
The connection to the server localhost:8080 was refused - did you specify the right host or port?
michaelobrien@Michaels-MBP machine-learning % brew install helm
==> Downloading https://ghcr.io/v2/homebrew/core/helm/manifests/3.16.3
######################################################################################################################################################################################################### 100.0%
==> Fetching helm
==> Downloading https://ghcr.io/v2/homebrew/core/helm/blobs/sha256:c41036a65f0eb30059a23392f183f30d9b911a235b3fd2f9f6f705933ce582f8
######################################################################################################################################################################################################### 100.0%
==> Pouring helm--3.16.3.arm64_sequoia.bottle.tar.gz
==> Caveats
zsh completions have been installed to:
  /opt/homebrew/share/zsh/site-functions
==> Summary
🍺  /opt/homebrew/Cellar/helm/3.16.3: 66 files, 55.4MB
==> Running `brew cleanup helm`...
Disable this behaviour by setting HOMEBREW_NO_INSTALL_CLEANUP.
Hide these hints with HOMEBREW_NO_ENV_HINTS (see `man brew`).
michaelobrien@Michaels-MBP machine-learning % brew install awscli
==> Downloading https://ghcr.io/v2/homebrew/core/awscli/manifests/2.22.7
######################################################################################################################################################################################################### 100.0%
==> Fetching dependencies for awscli: pycparser, cffi, ca-certificates, openssl@3, cryptography, mpdecimal, readline, sqlite, xz and [email protected]
==> Downloading https://ghcr.io/v2/homebrew/core/pycparser/manifests/2.22_1
######################################################################################################################################################################################################### 100.0%
==> Fetching pycparser
==> Downloading https://ghcr.io/v2/homebrew/core/pycparser/blobs/sha256:96eddd22a812be4f919562d6525ab8fa19b03b700426d2fbf892da0f4fd02e73
######################################################################################################################################################################################################### 100.0%
==> Downloading https://ghcr.io/v2/homebrew/core/cffi/manifests/1.17.1_1
######################################################################################################################################################################################################### 100.0%
==> Fetching cffi
==> Downloading https://ghcr.io/v2/homebrew/core/cffi/blobs/sha256:908333c6c31b4da4876ccb707cdf2b1ec52aee3d83fc3c4a1b8b52f148883512
######################################################################################################################################################################################################### 100.0%
==> Downloading https://ghcr.io/v2/homebrew/core/ca-certificates/manifests/2024-11-26
######################################################################################################################################################################################################### 100.0%
==> Fetching ca-certificates
==> Downloading https://ghcr.io/v2/homebrew/core/ca-certificates/blobs/sha256:7a3b5f75ca44d330e0f37432af09f58e37bfa873f25d340dece3c3e6c7927657
######################################################################################################################################################################################################### 100.0%
==> Downloading https://ghcr.io/v2/homebrew/core/openssl/3/manifests/3.4.0
######################################################################################################################################################################################################### 100.0%
==> Fetching openssl@3
==> Downloading https://ghcr.io/v2/homebrew/core/openssl/3/blobs/sha256:bf2e6c5cb39433b00fa2aaf4a6db681814149f62c01f04d381c3c64cd32c1bd3
######################################################################################################################################################################################################### 100.0%
==> Downloading https://ghcr.io/v2/homebrew/core/cryptography/manifests/44.0.0
######################################################################################################################################################################################################### 100.0%
==> Fetching cryptography
==> Downloading https://ghcr.io/v2/homebrew/core/cryptography/blobs/sha256:12226755e51fb0906b84524b8e05f888d934746b1b1789e4bf75d5b90f528c86
######################################################################################################################################################################################################### 100.0%
==> Downloading https://ghcr.io/v2/homebrew/core/mpdecimal/manifests/4.0.0-1
######################################################################################################################################################################################################### 100.0%
==> Fetching mpdecimal
==> Downloading https://ghcr.io/v2/homebrew/core/mpdecimal/blobs/sha256:0f5f269bed0e6be2de3edfc4b52867e656f993e5bcff40717f26ee94dd0d2211
######################################################################################################################################################################################################### 100.0%
==> Downloading https://ghcr.io/v2/homebrew/core/readline/manifests/8.2.13
######################################################################################################################################################################################################### 100.0%
==> Fetching readline
==> Downloading https://ghcr.io/v2/homebrew/core/readline/blobs/sha256:738c27ceee9a8b198f98438477ef7a513a96a965e3a434ac3aa8fb4ed76494b1
######################################################################################################################################################################################################### 100.0%
==> Downloading https://ghcr.io/v2/homebrew/core/sqlite/manifests/3.47.1
######################################################################################################################################################################################################### 100.0%
==> Fetching sqlite
==> Downloading https://ghcr.io/v2/homebrew/core/sqlite/blobs/sha256:22aa891d3868b2ff3cf5b2fff0b7c3bdc43fa9cf0c2f7047b213710e2710fbdd
######################################################################################################################################################################################################### 100.0%
==> Downloading https://ghcr.io/v2/homebrew/core/xz/manifests/5.6.3
######################################################################################################################################################################################################### 100.0%
==> Fetching xz
==> Downloading https://ghcr.io/v2/homebrew/core/xz/blobs/sha256:743c3d366f11b139445c5a7c923ac937d825cd172e316e138f021a9156145fb0
######################################################################################################################################################################################################### 100.0%
==> Downloading https://ghcr.io/v2/homebrew/core/python/3.12/manifests/3.12.7_1
######################################################################################################################################################################################################### 100.0%
==> Fetching [email protected]
==> Downloading https://ghcr.io/v2/homebrew/core/python/3.12/blobs/sha256:c1488710f182d429f03b04294806abe8f6557ec13fe9f2c2560d46738b050dfa
######################################################################################################################################################################################################### 100.0%
==> Fetching awscli
==> Downloading https://ghcr.io/v2/homebrew/core/awscli/blobs/sha256:56ba53f8f786a0ddf52c6491f6f468d3e972d2b69ffa0265fd400e6478fd4cf5
######################################################################################################################################################################################################### 100.0%
==> Installing dependencies for awscli: pycparser, cffi, ca-certificates, openssl@3, cryptography, mpdecimal, readline, sqlite, xz and [email protected]
==> Installing awscli dependency: pycparser
==> Downloading https://ghcr.io/v2/homebrew/core/pycparser/manifests/2.22_1
Already downloaded: /Users/michaelobrien/Library/Caches/Homebrew/downloads/bcc371a4c6cfaae40014a9277121028f0f532091988cdacb4d8c23556d3e5b96--pycparser-2.22_1.bottle_manifest.json
==> Pouring pycparser--2.22_1.all.bottle.tar.gz
🍺  /opt/homebrew/Cellar/pycparser/2.22_1: 98 files, 1.8MB
==> Installing awscli dependency: cffi
==> Downloading https://ghcr.io/v2/homebrew/core/cffi/manifests/1.17.1_1
Already downloaded: /Users/michaelobrien/Library/Caches/Homebrew/downloads/b99b4f0faeabe37a71351274b07e8ee17f52d483de3a33efa3e00bc1d76401d2--cffi-1.17.1_1.bottle_manifest.json
==> Pouring cffi--1.17.1_1.arm64_sequoia.bottle.tar.gz
🍺  /opt/homebrew/Cellar/cffi/1.17.1_1: 96 files, 1.7MB
==> Installing awscli dependency: ca-certificates
==> Downloading https://ghcr.io/v2/homebrew/core/ca-certificates/manifests/2024-11-26
Already downloaded: /Users/michaelobrien/Library/Caches/Homebrew/downloads/e16b55434e8bc1472ffb41e1a71c1b853417578c631fa1b69e8730f8cafca76c--ca-certificates-2024-11-26.bottle_manifest.json
==> Pouring ca-certificates--2024-11-26.all.bottle.tar.gz
==> Regenerating CA certificate bundle from keychain, this may take a while...
🍺  /opt/homebrew/Cellar/ca-certificates/2024-11-26: 4 files, 239.4KB
==> Installing awscli dependency: openssl@3
==> Downloading https://ghcr.io/v2/homebrew/core/openssl/3/manifests/3.4.0
Already downloaded: /Users/michaelobrien/Library/Caches/Homebrew/downloads/a8a169d38329b14f967d497d0ea77eee6dc5444175093fc7b26026b357e81736--openssl@3-3.4.0.bottle_manifest.json
==> Pouring [email protected]_sequoia.bottle.tar.gz
🍺  /opt/homebrew/Cellar/openssl@3/3.4.0: 7,236 files, 33.4MB
==> Installing awscli dependency: cryptography
==> Downloading https://ghcr.io/v2/homebrew/core/cryptography/manifests/44.0.0
Already downloaded: /Users/michaelobrien/Library/Caches/Homebrew/downloads/c1ee6ca8b5d122a783fb18505b8fc5873d31f8d282c96daa9b23a07ba194f877--cryptography-44.0.0.bottle_manifest.json
==> Pouring cryptography--44.0.0.arm64_sequoia.bottle.tar.gz
🍺  /opt/homebrew/Cellar/cryptography/44.0.0: 232 files, 8.9MB
==> Installing awscli dependency: mpdecimal
==> Downloading https://ghcr.io/v2/homebrew/core/mpdecimal/manifests/4.0.0-1
Already downloaded: /Users/michaelobrien/Library/Caches/Homebrew/downloads/7b63c3b34bee402290af49fac829a6682ab45ea5c9258b6fe03b590a03a4c4a9--mpdecimal-4.0.0-1.bottle_manifest.json
==> Pouring mpdecimal--4.0.0.arm64_sequoia.bottle.1.tar.gz
🍺  /opt/homebrew/Cellar/mpdecimal/4.0.0: 22 files, 641.3KB
==> Installing awscli dependency: readline
==> Downloading https://ghcr.io/v2/homebrew/core/readline/manifests/8.2.13
Already downloaded: /Users/michaelobrien/Library/Caches/Homebrew/downloads/aa1afa38997a2866d91c81fdda8b36d436cd4ea7a82aed07d13c83c56eb3460e--readline-8.2.13.bottle_manifest.json
==> Pouring readline--8.2.13.arm64_sequoia.bottle.tar.gz
🍺  /opt/homebrew/Cellar/readline/8.2.13: 51 files, 1.7MB
==> Installing awscli dependency: sqlite
==> Downloading https://ghcr.io/v2/homebrew/core/sqlite/manifests/3.47.1
Already downloaded: /Users/michaelobrien/Library/Caches/Homebrew/downloads/80056eb80c4e40ece338a9418f0e34bf7af301f5bcd6714ebd9e3f603ef415ad--sqlite-3.47.1.bottle_manifest.json
==> Pouring sqlite--3.47.1.arm64_sequoia.bottle.tar.gz
🍺  /opt/homebrew/Cellar/sqlite/3.47.1: 12 files, 4.9MB
==> Installing awscli dependency: xz
==> Downloading https://ghcr.io/v2/homebrew/core/xz/manifests/5.6.3
Already downloaded: /Users/michaelobrien/Library/Caches/Homebrew/downloads/e68799185cc17334108ba286ba8c3df0587c9fd8375647d85f7157b62a797599--xz-5.6.3.bottle_manifest.json
==> Pouring xz--5.6.3.arm64_sequoia.bottle.tar.gz
🍺  /opt/homebrew/Cellar/xz/5.6.3: 96 files, 2.0MB
==> Installing awscli dependency: [email protected]
==> Downloading https://ghcr.io/v2/homebrew/core/python/3.12/manifests/3.12.7_1
Already downloaded: /Users/michaelobrien/Library/Caches/Homebrew/downloads/c34a42771b51eed6eb00535b257d60273bcda16532cbbf3288367ae739d2b4f9--python@3.12-3.12.7_1.bottle_manifest.json
==> Pouring [email protected]_1.arm64_sequoia.bottle.tar.gz
==> /opt/homebrew/Cellar/[email protected]/3.12.7_1/bin/python3.12 -Im ensurepip
==> /opt/homebrew/Cellar/[email protected]/3.12.7_1/bin/python3.12 -Im pip install -v --no-index --upgrade --isolated --target=/opt/homebrew/lib/python3.12/site-packages /opt/homebrew/Cellar/[email protected]/3.12.7_1/Fr
🍺  /opt/homebrew/Cellar/[email protected]/3.12.7_1: 3,252 files, 65.3MB
==> Installing awscli
==> Pouring awscli--2.22.7.arm64_sequoia.bottle.tar.gz
==> Caveats
The "examples" directory has been installed to:
  /opt/homebrew/share/awscli/examples

zsh completions and functions have been installed to:
  /opt/homebrew/share/zsh/site-functions
==> Summary
🍺  /opt/homebrew/Cellar/awscli/2.22.7: 14,071 files, 126.8MB
==> Running `brew cleanup awscli`...
Disable this behaviour by setting HOMEBREW_NO_INSTALL_CLEANUP.
Hide these hints with HOMEBREW_NO_ENV_HINTS (see `man brew`).
==> Caveats
==> awscli
The "examples" directory has been installed to:
  /opt/homebrew/share/awscli/examples

zsh completions and functions have been installed to:
  /opt/homebrew/share/zsh/site-functions
michaelobrien@Michaels-MBP machine-learning % python --version
zsh: command not found: python
michaelobrien@Michaels-MBP machine-learning % python3 --version
Python 3.9.6
michaelobrien@Michaels-MBP machine-learning % ls
LICENSE		README.md	build.sh	cuda		environments	gpus.md		links.md	stats.md
michaelobrien@Michaels-MBP machine-learning % cd environments/windows/src 
michaelobrien@Michaels-MBP src % ls
google-gemma	tflow.py
michaelobrien@Michaels-MBP src % vi tflow.py 
michaelobrien@Michaels-MBP src % python3 -m venv /venv-metal
Error: [Errno 30] Read-only file system: '/venv-metal'
michaelobrien@Michaels-MBP src % python3 -m venv venv-metal 
michaelobrien@Michaels-MBP src % source venv-metal/bin/activate 
(venv-metal) michaelobrien@Michaels-MBP src % python3 -m pip install tensorflow==2.14.0
Collecting tensorflow==2.14.0
  Downloading tensorflow-2.14.0-cp39-cp39-macosx_12_0_arm64.whl (2.1 kB)
Collecting tensorflow-macos==2.14.0
  Downloading tensorflow_macos-2.14.0-cp39-cp39-macosx_12_0_arm64.whl (199.7 MB)
     |████████████████████████████████| 199.7 MB 4.7 MB/s 
Collecting wrapt<1.15,>=1.11.0
  Downloading wrapt-1.14.1-cp39-cp39-macosx_11_0_arm64.whl (35 kB)
Collecting tensorflow-io-gcs-filesystem>=0.23.1
  Downloading tensorflow_io_gcs_filesystem-0.37.1-cp39-cp39-macosx_12_0_arm64.whl (3.5 MB)
     |████████████████████████████████| 3.5 MB 8.9 MB/s 
Collecting tensorboard<2.15,>=2.14
  Downloading tensorboard-2.14.1-py3-none-any.whl (5.5 MB)
     |████████████████████████████████| 5.5 MB 8.9 MB/s 
Collecting keras<2.15,>=2.14.0
  Downloading keras-2.14.0-py3-none-any.whl (1.7 MB)
     |████████████████████████████████| 1.7 MB 8.7 MB/s 
Collecting packaging
  Downloading packaging-24.2-py3-none-any.whl (65 kB)
     |████████████████████████████████| 65 kB 10.2 MB/s 
Collecting protobuf!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.20.3
  Downloading protobuf-4.25.5-cp37-abi3-macosx_10_9_universal2.whl (394 kB)
     |████████████████████████████████| 394 kB 10.5 MB/s 
Collecting libclang>=13.0.0
  Downloading libclang-18.1.1-1-py2.py3-none-macosx_11_0_arm64.whl (25.8 MB)
     |████████████████████████████████| 25.8 MB 12.3 MB/s 
Collecting astunparse>=1.6.0
  Downloading astunparse-1.6.3-py2.py3-none-any.whl (12 kB)
Collecting grpcio<2.0,>=1.24.3
  Downloading grpcio-1.68.0-cp39-cp39-macosx_10_9_universal2.whl (11.1 MB)
     |████████████████████████████████| 11.1 MB 8.9 MB/s 
Collecting six>=1.12.0
  Downloading six-1.16.0-py2.py3-none-any.whl (11 kB)
Collecting numpy>=1.23.5
  Downloading numpy-2.0.2-cp39-cp39-macosx_14_0_arm64.whl (5.3 MB)
     |████████████████████████████████| 5.3 MB 10.8 MB/s 
Collecting flatbuffers>=23.5.26
  Downloading flatbuffers-24.3.25-py2.py3-none-any.whl (26 kB)
Collecting typing-extensions>=3.6.6
  Downloading typing_extensions-4.12.2-py3-none-any.whl (37 kB)
Collecting termcolor>=1.1.0
  Downloading termcolor-2.5.0-py3-none-any.whl (7.8 kB)
Collecting opt-einsum>=2.3.2
  Downloading opt_einsum-3.4.0-py3-none-any.whl (71 kB)
     |████████████████████████████████| 71 kB 2.5 MB/s 
Collecting tensorflow-estimator<2.15,>=2.14.0
  Downloading tensorflow_estimator-2.14.0-py2.py3-none-any.whl (440 kB)
     |████████████████████████████████| 440 kB 10.1 MB/s 
Collecting gast!=0.5.0,!=0.5.1,!=0.5.2,>=0.2.1
  Downloading gast-0.6.0-py3-none-any.whl (21 kB)
Collecting h5py>=2.9.0
  Downloading h5py-3.12.1-cp39-cp39-macosx_11_0_arm64.whl (2.9 MB)
     |████████████████████████████████| 2.9 MB 8.7 MB/s 
Requirement already satisfied: setuptools in ./venv-metal/lib/python3.9/site-packages (from tensorflow-macos==2.14.0->tensorflow==2.14.0) (58.0.4)
Collecting google-pasta>=0.1.1
  Downloading google_pasta-0.2.0-py3-none-any.whl (57 kB)
     |████████████████████████████████| 57 kB 10.7 MB/s 
Collecting absl-py>=1.0.0
  Downloading absl_py-2.1.0-py3-none-any.whl (133 kB)
     |████████████████████████████████| 133 kB 11.1 MB/s 
Collecting ml-dtypes==0.2.0
  Downloading ml_dtypes-0.2.0-cp39-cp39-macosx_10_9_universal2.whl (1.2 MB)
     |████████████████████████████████| 1.2 MB 10.6 MB/s 
Collecting wheel<1.0,>=0.23.0
  Downloading wheel-0.45.1-py3-none-any.whl (72 kB)
     |████████████████████████████████| 72 kB 4.9 MB/s 
Collecting requests<3,>=2.21.0
  Downloading requests-2.32.3-py3-none-any.whl (64 kB)
     |████████████████████████████████| 64 kB 9.4 MB/s 
Collecting tensorboard-data-server<0.8.0,>=0.7.0
  Downloading tensorboard_data_server-0.7.2-py3-none-any.whl (2.4 kB)
Collecting google-auth-oauthlib<1.1,>=0.5
  Downloading google_auth_oauthlib-1.0.0-py2.py3-none-any.whl (18 kB)
Collecting markdown>=2.6.8
  Downloading Markdown-3.7-py3-none-any.whl (106 kB)
     |████████████████████████████████| 106 kB 11.6 MB/s 
Collecting werkzeug>=1.0.1
  Downloading werkzeug-3.1.3-py3-none-any.whl (224 kB)
     |████████████████████████████████| 224 kB 9.2 MB/s 
Collecting google-auth<3,>=1.6.3
  Downloading google_auth-2.36.0-py2.py3-none-any.whl (209 kB)
     |████████████████████████████████| 209 kB 10.5 MB/s 
Collecting cachetools<6.0,>=2.0.0
  Downloading cachetools-5.5.0-py3-none-any.whl (9.5 kB)
Collecting pyasn1-modules>=0.2.1
  Downloading pyasn1_modules-0.4.1-py3-none-any.whl (181 kB)
     |████████████████████████████████| 181 kB 11.6 MB/s 
Collecting rsa<5,>=3.1.4
  Downloading rsa-4.9-py3-none-any.whl (34 kB)
Collecting requests-oauthlib>=0.7.0
  Downloading requests_oauthlib-2.0.0-py2.py3-none-any.whl (24 kB)
Collecting importlib-metadata>=4.4
  Downloading importlib_metadata-8.5.0-py3-none-any.whl (26 kB)
Collecting zipp>=3.20
  Downloading zipp-3.21.0-py3-none-any.whl (9.6 kB)
Collecting pyasn1<0.7.0,>=0.4.6
  Downloading pyasn1-0.6.1-py3-none-any.whl (83 kB)
     |████████████████████████████████| 83 kB 7.4 MB/s 
Collecting urllib3<3,>=1.21.1
  Downloading urllib3-2.2.3-py3-none-any.whl (126 kB)
     |████████████████████████████████| 126 kB 10.9 MB/s 
Collecting idna<4,>=2.5
  Downloading idna-3.10-py3-none-any.whl (70 kB)
     |████████████████████████████████| 70 kB 10.8 MB/s 
Collecting certifi>=2017.4.17
  Downloading certifi-2024.8.30-py3-none-any.whl (167 kB)
     |████████████████████████████████| 167 kB 11.0 MB/s 
Collecting charset-normalizer<4,>=2
  Downloading charset_normalizer-3.4.0-cp39-cp39-macosx_11_0_arm64.whl (120 kB)
     |████████████████████████████████| 120 kB 10.9 MB/s 
Collecting oauthlib>=3.0.0
  Downloading oauthlib-3.2.2-py3-none-any.whl (151 kB)
     |████████████████████████████████| 151 kB 11.1 MB/s 
Collecting MarkupSafe>=2.1.1
  Downloading MarkupSafe-3.0.2-cp39-cp39-macosx_11_0_arm64.whl (12 kB)
Installing collected packages: urllib3, pyasn1, idna, charset-normalizer, certifi, zipp, rsa, requests, pyasn1-modules, oauthlib, cachetools, requests-oauthlib, MarkupSafe, importlib-metadata, google-auth, wheel, werkzeug, tensorboard-data-server, six, protobuf, numpy, markdown, grpcio, google-auth-oauthlib, absl-py, wrapt, typing-extensions, termcolor, tensorflow-io-gcs-filesystem, tensorflow-estimator, tensorboard, packaging, opt-einsum, ml-dtypes, libclang, keras, h5py, google-pasta, gast, flatbuffers, astunparse, tensorflow-macos, tensorflow
Successfully installed MarkupSafe-3.0.2 absl-py-2.1.0 astunparse-1.6.3 cachetools-5.5.0 certifi-2024.8.30 charset-normalizer-3.4.0 flatbuffers-24.3.25 gast-0.6.0 google-auth-2.36.0 google-auth-oauthlib-1.0.0 google-pasta-0.2.0 grpcio-1.68.0 h5py-3.12.1 idna-3.10 importlib-metadata-8.5.0 keras-2.14.0 libclang-18.1.1 markdown-3.7 ml-dtypes-0.2.0 numpy-2.0.2 oauthlib-3.2.2 opt-einsum-3.4.0 packaging-24.2 protobuf-4.25.5 pyasn1-0.6.1 pyasn1-modules-0.4.1 requests-2.32.3 requests-oauthlib-2.0.0 rsa-4.9 six-1.16.0 tensorboard-2.14.1 tensorboard-data-server-0.7.2 tensorflow-2.14.0 tensorflow-estimator-2.14.0 tensorflow-io-gcs-filesystem-0.37.1 tensorflow-macos-2.14.0 termcolor-2.5.0 typing-extensions-4.12.2 urllib3-2.2.3 werkzeug-3.1.3 wheel-0.45.1 wrapt-1.14.1 zipp-3.21.0
WARNING: You are using pip version 21.2.4; however, version 24.3.1 is available.
You should consider upgrading via the '/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/bin/python3 -m pip install --upgrade pip' command.
(venv-metal) michaelobrien@Michaels-MBP src % python3 -m pip install tensorflow-metal==1.1.0
Collecting tensorflow-metal==1.1.0
  Downloading tensorflow_metal-1.1.0-cp39-cp39-macosx_12_0_arm64.whl (1.4 MB)
     |████████████████████████████████| 1.4 MB 4.0 MB/s 
Requirement already satisfied: wheel~=0.35 in ./venv-metal/lib/python3.9/site-packages (from tensorflow-metal==1.1.0) (0.45.1)
Requirement already satisfied: six>=1.15.0 in ./venv-metal/lib/python3.9/site-packages (from tensorflow-metal==1.1.0) (1.16.0)
Installing collected packages: tensorflow-metal
Successfully installed tensorflow-metal-1.1.0
WARNING: You are using pip version 21.2.4; however, version 24.3.1 is available.
You should consider upgrading via the '/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/bin/python3 -m pip install --upgrade pip' command.
(venv-metal) michaelobrien@Michaels-MBP src % python3 tflow.py 

A module that was compiled using NumPy 1.x cannot be run in
NumPy 2.0.2 as it may crash. To support both 1.x and 2.x
versions of NumPy, modules must be compiled with NumPy 2.0.
Some module may need to rebuild instead e.g. with 'pybind11>=2.12'.

If you are a user of the module, the easiest solution will be to
downgrade to 'numpy<2' or try to upgrade the affected module.
We expect that some modules will need time to support NumPy 2.

Traceback (most recent call last):  File "/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/tflow.py", line 1, in <module>
    import tensorflow as tf
  File "/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/tensorflow/__init__.py", line 38, in <module>
    from tensorflow.python.tools import module_util as _module_util
  File "/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/tensorflow/python/__init__.py", line 42, in <module>
    from tensorflow.python.saved_model import saved_model
  File "/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/tensorflow/python/saved_model/saved_model.py", line 20, in <module>
    from tensorflow.python.saved_model import builder
  File "/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/tensorflow/python/saved_model/builder.py", line 23, in <module>
    from tensorflow.python.saved_model.builder_impl import _SavedModelBuilder
  File "/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/tensorflow/python/saved_model/builder_impl.py", line 26, in <module>
    from tensorflow.python.framework import dtypes
  File "/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/tensorflow/python/framework/dtypes.py", line 35, in <module>
    from tensorflow.tsl.python.lib.core import pywrap_ml_dtypes
AttributeError: _ARRAY_API not found
ImportError: numpy.core._multiarray_umath failed to import
ImportError: numpy.core.umath failed to import
Traceback (most recent call last):
  File "/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/tflow.py", line 1, in <module>
    import tensorflow as tf
  File "/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/tensorflow/__init__.py", line 38, in <module>
    from tensorflow.python.tools import module_util as _module_util
  File "/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/tensorflow/python/__init__.py", line 42, in <module>
    from tensorflow.python.saved_model import saved_model
  File "/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/tensorflow/python/saved_model/saved_model.py", line 20, in <module>
    from tensorflow.python.saved_model import builder
  File "/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/tensorflow/python/saved_model/builder.py", line 23, in <module>
    from tensorflow.python.saved_model.builder_impl import _SavedModelBuilder
  File "/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/tensorflow/python/saved_model/builder_impl.py", line 26, in <module>
    from tensorflow.python.framework import dtypes
  File "/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/tensorflow/python/framework/dtypes.py", line 37, in <module>
    _np_bfloat16 = pywrap_ml_dtypes.bfloat16()
TypeError: Unable to convert function return value to a Python type! The signature was
	() -> handle
(venv-metal) michaelobrien@Michaels-MBP src % python3 -m pip uninstall numpy
Found existing installation: numpy 2.0.2
Uninstalling numpy-2.0.2:
  Would remove:
    /Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/bin/f2py
 ...
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/tests/test_public_api.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/tests/test_reloading.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/tests/test_scripts.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/tests/test_warnings.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/__init__.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/mypy_plugin.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/__init__.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/arithmetic.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/array_constructors.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/array_like.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/arrayprint.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/arrayterator.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/bitwise_ops.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/comparisons.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/dtype.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/einsumfunc.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/flatiter.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/fromnumeric.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/index_tricks.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/lib_utils.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/lib_version.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/literal.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/ma.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/mod.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/modules.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/multiarray.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/ndarray_conversion.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/ndarray_misc.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/ndarray_shape_manipulation.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/numeric.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/numerictypes.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/random.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/scalars.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/simple.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/simple_py3.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/ufunc_config.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/ufunclike.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/ufuncs.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/data/pass/warnings_and_errors.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/test_isfile.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/test_runtime.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/typing/tests/test_typing.cpython-39.pyc
    /Users/michaelobrien/Library/Caches/com.apple.python/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/numpy/version.cpython-39.pyc
Proceed (Y/n)? y
  Successfully uninstalled numpy-2.0.2
(venv-metal) michaelobrien@Michaels-MBP src %  python3 -m pip install numpy==1.24.1
Collecting numpy==1.24.1
  Downloading numpy-1.24.1-cp39-cp39-macosx_11_0_arm64.whl (13.9 MB)
     |████████████████████████████████| 13.9 MB 1.9 MB/s 
Installing collected packages: numpy
Successfully installed numpy-1.24.1
WARNING: You are using pip version 21.2.4; however, version 24.3.1 is available.
You should consider upgrading via the '/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/bin/python3 -m pip install --upgrade pip' command.
(venv-metal) michaelobrien@Michaels-MBP src % python3 tflow.py                     
/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/urllib3/__init__.py:35: NotOpenSSLWarning: urllib3 v2 only supports OpenSSL 1.1.1+, currently the 'ssl' module is compiled with 'LibreSSL 2.8.3'. See: https://github.com/urllib3/urllib3/issues/3020
  warnings.warn(
2024-11-29 16:16:54.078481: I metal_plugin/src/device/metal_device.cc:1154] Metal device set to: Apple M4 Max
2024-11-29 16:16:54.078500: I metal_plugin/src/device/metal_device.cc:296] systemMemory: 48.00 GB
2024-11-29 16:16:54.078505: I metal_plugin/src/device/metal_device.cc:313] maxCacheSize: 18.00 GB
2024-11-29 16:16:54.078549: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:306] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.
2024-11-29 16:16:54.078580: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:272] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)
Downloading data from https://www.cs.toronto.edu/~kriz/cifar-100-python.tar.gz
169001437/169001437 [==============================] - 51s 0us/step
2024-11-29 16:17:47.644892: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:117] Plugin optimizer for device_type GPU is enabled.
Epoch 1/25
98/98 [==============================] - 21s 129ms/step - loss: 4.4390 - accuracy: 0.0755
Epoch 2/25
98/98 [==============================] - 11s 110ms/step - loss: 3.8977 - accuracy: 0.1361
Epoch 3/25
98/98 [==============================] - 12s 118ms/step - loss: 3.5676 - accuracy: 0.1916
Epoch 4/25
98/98 [==============================] - 12s 120ms/step - loss: 3.2979 - accuracy: 0.2426
Epoch 5/25
98/98 [==============================] - 11s 117ms/step - loss: 3.0022 - accuracy: 0.2892
Epoch 6/25
98/98 [==============================] - 11s 115ms/step - loss: 2.7999 - accuracy: 0.3198
Epoch 7/25
98/98 [==============================] - 11s 114ms/step - loss: 2.5051 - accuracy: 0.3780
Epoch 8/25
98/98 [==============================] - 12s 120ms/step - loss: 2.9318 - accuracy: 0.3063
Epoch 9/25
98/98 [==============================] - 12s 127ms/step - loss: 2.4164 - accuracy: 0.3918
Epoch 10/25
98/98 [==============================] - 13s 134ms/step - loss: 2.3208 - accuracy: 0.4186
Epoch 11/25
98/98 [==============================] - 14s 143ms/step - loss: 2.7591 - accuracy: 0.3258
Epoch 12/25
98/98 [==============================] - 15s 148ms/step - loss: 2.3814 - accuracy: 0.3930
Epoch 13/25
98/98 [==============================] - 15s 154ms/step - loss: 2.0169 - accuracy: 0.4695
Epoch 14/25
98/98 [==============================] - 15s 152ms/step - loss: 1.7882 - accuracy: 0.5274
Epoch 15/25
98/98 [==============================] - 14s 146ms/step - loss: 1.5642 - accuracy: 0.5775
Epoch 16/25
98/98 [==============================] - 15s 152ms/step - loss: 1.4134 - accuracy: 0.6154
Epoch 17/25
98/98 [==============================] - 15s 149ms/step - loss: 1.2896 - accuracy: 0.6414
Epoch 18/25
98/98 [==============================] - 15s 148ms/step - loss: 0.9938 - accuracy: 0.7226
Epoch 19/25
98/98 [==============================] - 14s 145ms/step - loss: 0.7646 - accuracy: 0.7854
Epoch 20/25
98/98 [==============================] - 14s 141ms/step - loss: 0.7736 - accuracy: 0.7753
Epoch 21/25
98/98 [==============================] - 13s 137ms/step - loss: 0.6513 - accuracy: 0.8199
Epoch 22/25
98/98 [==============================] - 13s 136ms/step - loss: 0.6704 - accuracy: 0.8139
Epoch 23/25
98/98 [==============================] - 13s 132ms/step - loss: 0.3868 - accuracy: 0.8993
Epoch 24/25
98/98 [==============================] - 13s 132ms/step - loss: 0.4294 - accuracy: 0.8914
Epoch 25/25
98/98 [==============================] - 13s 131ms/step - loss: 0.2311 - accuracy: 0.9430

The text was updated successfully, but these errors were encountered:

obriensystems · 2024-11-30T13:55:02Z

after cooldown
fan peak at epoch 10

(venv-metal) michaelobrien@mbp8 src % python3 tflow.py
/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/urllib3/__init__.py:35: NotOpenSSLWarning: urllib3 v2 only supports OpenSSL 1.1.1+, currently the 'ssl' module is compiled with 'LibreSSL 2.8.3'. See: https://github.com/urllib3/urllib3/issues/3020
  warnings.warn(
2024-11-30 08:28:00.209405: I metal_plugin/src/device/metal_device.cc:1154] Metal device set to: Apple M4 Max
2024-11-30 08:28:00.209423: I metal_plugin/src/device/metal_device.cc:296] systemMemory: 48.00 GB
2024-11-30 08:28:00.209428: I metal_plugin/src/device/metal_device.cc:313] maxCacheSize: 18.00 GB
2024-11-30 08:28:00.209470: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:306] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.
2024-11-30 08:28:00.209686: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:272] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)
2024-11-30 08:28:01.136248: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:117] Plugin optimizer for device_type GPU is enabled.
Epoch 1/25
98/98 [==============================] - 13s 112ms/step - loss: 4.3670 - accuracy: 0.0800
Epoch 2/25
98/98 [==============================] - 11s 109ms/step - loss: 3.5192 - accuracy: 0.1801
Epoch 3/25
98/98 [==============================] - 11s 109ms/step - loss: 3.1483 - accuracy: 0.2514
Epoch 4/25
98/98 [==============================] - 11s 109ms/step - loss: 3.1672 - accuracy: 0.2577
Epoch 5/25
98/98 [==============================] - 11s 110ms/step - loss: 3.0020 - accuracy: 0.2731
Epoch 6/25
98/98 [==============================] - 11s 110ms/step - loss: 3.0056 - accuracy: 0.3035
Epoch 7/25
98/98 [==============================] - 11s 109ms/step - loss: 3.3055 - accuracy: 0.2486
Epoch 8/25
98/98 [==============================] - 11s 109ms/step - loss: 2.7237 - accuracy: 0.3349
Epoch 9/25
98/98 [==============================] - 11s 109ms/step - loss: 2.6401 - accuracy: 0.3494
Epoch 10/25
98/98 [==============================] - 11s 109ms/step - loss: 2.4505 - accuracy: 0.3802
Epoch 11/25
98/98 [==============================] - 11s 109ms/step - loss: 2.1510 - accuracy: 0.4419
Epoch 12/25
98/98 [==============================] - 11s 109ms/step - loss: 2.2875 - accuracy: 0.4291
Epoch 13/25
98/98 [==============================] - 11s 109ms/step - loss: 1.9069 - accuracy: 0.4977
Epoch 14/25
98/98 [==============================] - 11s 109ms/step - loss: 1.7260 - accuracy: 0.5543
Epoch 15/25
98/98 [==============================] - 11s 109ms/step - loss: 1.4935 - accuracy: 0.6001
Epoch 16/25
98/98 [==============================] - 11s 109ms/step - loss: 1.2285 - accuracy: 0.6633
Epoch 17/25
98/98 [==============================] - 11s 109ms/step - loss: 0.9949 - accuracy: 0.7274
Epoch 18/25
98/98 [==============================] - 11s 109ms/step - loss: 0.9422 - accuracy: 0.7382
Epoch 19/25
98/98 [==============================] - 11s 109ms/step - loss: 0.6299 - accuracy: 0.8205
Epoch 20/25
98/98 [==============================] - 11s 109ms/step - loss: 0.4877 - accuracy: 0.8660
Epoch 21/25
98/98 [==============================] - 11s 109ms/step - loss: 0.4363 - accuracy: 0.8883
Epoch 22/25
98/98 [==============================] - 11s 109ms/step - loss: 0.3870 - accuracy: 0.8985
Epoch 23/25
98/98 [==============================] - 11s 109ms/step - loss: 1.6514 - accuracy: 0.6111
Epoch 24/25
98/98 [==============================] - 11s 109ms/step - loss: 1.5246 - accuracy: 0.6167
Epoch 25/25
98/98 [==============================] - 11s 109ms/step - loss: 1.3067 - accuracy: 0.6650

(venv-metal) michaelobrien@mbp8 src % python3 tflow.py
/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/urllib3/__init__.py:35: NotOpenSSLWarning: urllib3 v2 only supports OpenSSL 1.1.1+, currently the 'ssl' module is compiled with 'LibreSSL 2.8.3'. See: https://github.com/urllib3/urllib3/issues/3020
  warnings.warn(
2024-11-30 09:08:31.609379: I metal_plugin/src/device/metal_device.cc:1154] Metal device set to: Apple M4 Max
2024-11-30 09:08:31.609400: I metal_plugin/src/device/metal_device.cc:296] systemMemory: 48.00 GB
2024-11-30 09:08:31.609404: I metal_plugin/src/device/metal_device.cc:313] maxCacheSize: 18.00 GB
2024-11-30 09:08:31.609432: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:306] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.
2024-11-30 09:08:31.609452: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:272] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)
2024-11-30 09:08:32.436008: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:117] Plugin optimizer for device_type GPU is enabled.
Epoch 1/25
98/98 [==============================] - 13s 113ms/step - loss: 4.3050 - accuracy: 0.0803
Epoch 2/25
98/98 [==============================] - 11s 109ms/step - loss: 3.5179 - accuracy: 0.1863
Epoch 3/25
98/98 [==============================] - 11s 109ms/step - loss: 3.7929 - accuracy: 0.1626
Epoch 4/25
98/98 [==============================] - 11s 109ms/step - loss: 3.1797 - accuracy: 0.2402
Epoch 5/25
98/98 [==============================] - 11s 111ms/step - loss: 2.8858 - accuracy: 0.2928
Epoch 6/25
98/98 [==============================] - 11s 110ms/step - loss: 2.6869 - accuracy: 0.3399
Epoch 7/25
98/98 [==============================] - 11s 109ms/step - loss: 2.4415 - accuracy: 0.3835
Epoch 8/25
98/98 [==============================] - 11s 111ms/step - loss: 2.4758 - accuracy: 0.3876
Epoch 9/25
98/98 [==============================] - 11s 110ms/step - loss: 2.8488 - accuracy: 0.3102
Epoch 10/25
98/98 [==============================] - 11s 109ms/step - loss: 2.3142 - accuracy: 0.4082
Epoch 11/25
98/98 [==============================] - 11s 109ms/step - loss: 2.1877 - accuracy: 0.4399
Epoch 12/25
98/98 [==============================] - 11s 109ms/step - loss: 2.0834 - accuracy: 0.4722
Epoch 13/25
98/98 [==============================] - 11s 109ms/step - loss: 4.1953 - accuracy: 0.0923
Epoch 14/25
98/98 [==============================] - 11s 109ms/step - loss: 3.4182 - accuracy: 0.1844
Epoch 15/25
98/98 [==============================] - 11s 109ms/step - loss: 3.0552 - accuracy: 0.2563
Epoch 16/25
98/98 [==============================] - 11s 109ms/step - loss: 2.7534 - accuracy: 0.3087
Epoch 17/25
98/98 [==============================] - 11s 109ms/step - loss: 2.5937 - accuracy: 0.3508
Epoch 18/25
98/98 [==============================] - 11s 109ms/step - loss: 2.3531 - accuracy: 0.3908
Epoch 19/25
98/98 [==============================] - 11s 109ms/step - loss: 2.1529 - accuracy: 0.4356
Epoch 20/25
98/98 [==============================] - 11s 109ms/step - loss: 2.0110 - accuracy: 0.4726
Epoch 21/25
98/98 [==============================] - 11s 109ms/step - loss: 1.7864 - accuracy: 0.5158
Epoch 22/25
98/98 [==============================] - 11s 109ms/step - loss: 1.6686 - accuracy: 0.5551
Epoch 23/25
98/98 [==============================] - 11s 108ms/step - loss: 1.4184 - accuracy: 0.6056
Epoch 24/25
98/98 [==============================] - 11s 108ms/step - loss: 1.2265 - accuracy: 0.6594
Epoch 25/25
98/98 [==============================] - 11s 109ms/step - loss: 1.1088 - accuracy: 0.6821
(venv-metal) michaelobrien@mbp8 src % python3 tflow.py
/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/urllib3/__init__.py:35: NotOpenSSLWarning: urllib3 v2 only supports OpenSSL 1.1.1+, currently the 'ssl' module is compiled with 'LibreSSL 2.8.3'. See: https://github.com/urllib3/urllib3/issues/3020
  warnings.warn(
2024-11-30 10:35:49.321482: I metal_plugin/src/device/metal_device.cc:1154] Metal device set to: Apple M4 Max
2024-11-30 10:35:49.321507: I metal_plugin/src/device/metal_device.cc:296] systemMemory: 48.00 GB
2024-11-30 10:35:49.321512: I metal_plugin/src/device/metal_device.cc:313] maxCacheSize: 18.00 GB
2024-11-30 10:35:49.321565: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:306] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.
2024-11-30 10:35:49.321785: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:272] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)
2024-11-30 10:35:50.157551: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:117] Plugin optimizer for device_type GPU is enabled.
Epoch 1/25
98/98 [==============================] - 13s 113ms/step - loss: 4.3071 - accuracy: 0.0851
Epoch 2/25
98/98 [==============================] - 11s 109ms/step - loss: 3.4967 - accuracy: 0.1858
Epoch 3/25
98/98 [==============================] - 11s 109ms/step - loss: 3.3162 - accuracy: 0.2184
Epoch 4/25
98/98 [==============================] - 11s 110ms/step - loss: 3.1572 - accuracy: 0.2506
Epoch 5/25
98/98 [==============================] - 11s 111ms/step - loss: 2.8917 - accuracy: 0.3051
Epoch 6/25
98/98 [==============================] - 11s 109ms/step - loss: 2.9371 - accuracy: 0.2990
Epoch 7/25
98/98 [==============================] - 11s 109ms/step - loss: 2.8720 - accuracy: 0.3023
Epoch 8/25
98/98 [==============================] - 11s 109ms/step - loss: 2.9617 - accuracy: 0.3044
Epoch 9/25
98/98 [==============================] - 11s 109ms/step - loss: 2.4475 - accuracy: 0.3803
Epoch 10/25
98/98 [==============================] - 11s 109ms/step - loss: 2.2926 - accuracy: 0.4280
Epoch 11/25
98/98 [==============================] - 11s 109ms/step - loss: 2.0490 - accuracy: 0.4719
Epoch 12/25
98/98 [==============================] - 11s 109ms/step - loss: 1.7676 - accuracy: 0.5353
Epoch 13/25
98/98 [==============================] - 11s 109ms/step - loss: 1.4656 - accuracy: 0.6032
Epoch 14/25
98/98 [==============================] - 11s 109ms/step - loss: 1.4397 - accuracy: 0.6119
Epoch 15/25
98/98 [==============================] - 11s 109ms/step - loss: 1.1733 - accuracy: 0.6867
Epoch 16/25
98/98 [==============================] - 11s 109ms/step - loss: 1.1928 - accuracy: 0.6765
Epoch 17/25
98/98 [==============================] - 11s 109ms/step - loss: 0.7554 - accuracy: 0.7877
Epoch 18/25
98/98 [==============================] - 11s 109ms/step - loss: 0.5551 - accuracy: 0.8480
Epoch 19/25
98/98 [==============================] - 11s 109ms/step - loss: 0.9247 - accuracy: 0.7725
Epoch 20/25
98/98 [==============================] - 11s 109ms/step - loss: 2.6168 - accuracy: 0.3953
Epoch 21/25
98/98 [==============================] - 11s 109ms/step - loss: 1.6708 - accuracy: 0.5491
Epoch 22/25
98/98 [==============================] - 11s 109ms/step - loss: 1.0092 - accuracy: 0.7171
Epoch 23/25
98/98 [==============================] - 11s 109ms/step - loss: 0.6145 - accuracy: 0.8337
Epoch 24/25
98/98 [==============================] - 11s 109ms/step - loss: 0.6219 - accuracy: 0.8448
Epoch 25/25
98/98 [==============================] - 11s 109ms/step - loss: 0.7705 - accuracy: 0.8072
(venv-metal) michaelobrien@mbp8 src % python3 tflow.py
/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/urllib3/__init__.py:35: NotOpenSSLWarning: urllib3 v2 only supports OpenSSL 1.1.1+, currently the 'ssl' module is compiled with 'LibreSSL 2.8.3'. See: https://github.com/urllib3/urllib3/issues/3020
  warnings.warn(
2024-11-30 10:41:05.517382: I metal_plugin/src/device/metal_device.cc:1154] Metal device set to: Apple M4 Max
2024-11-30 10:41:05.517404: I metal_plugin/src/device/metal_device.cc:296] systemMemory: 48.00 GB
2024-11-30 10:41:05.517408: I metal_plugin/src/device/metal_device.cc:313] maxCacheSize: 18.00 GB
2024-11-30 10:41:05.517438: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:306] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.
2024-11-30 10:41:05.517459: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:272] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)
2024-11-30 10:41:06.305326: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:117] Plugin optimizer for device_type GPU is enabled.
Epoch 1/25
98/98 [==============================] - 13s 112ms/step - loss: 4.3656 - accuracy: 0.0755
Epoch 2/25
98/98 [==============================] - 11s 108ms/step - loss: 3.5026 - accuracy: 0.1862
Epoch 3/25
98/98 [==============================] - 11s 108ms/step - loss: 3.1130 - accuracy: 0.2524
Epoch 4/25
98/98 [==============================] - 11s 109ms/step - loss: 2.9386 - accuracy: 0.2869
Epoch 5/25
98/98 [==============================] - 11s 109ms/step - loss: 2.5929 - accuracy: 0.3523
Epoch 6/25
98/98 [==============================] - 11s 109ms/step - loss: 2.5307 - accuracy: 0.3769
Epoch 7/25
98/98 [==============================] - 11s 109ms/step - loss: 2.3148 - accuracy: 0.4052
Epoch 8/25
98/98 [==============================] - 11s 109ms/step - loss: 2.1004 - accuracy: 0.4578
Epoch 9/25
98/98 [==============================] - 11s 109ms/step - loss: 1.8950 - accuracy: 0.5044
Epoch 10/25
98/98 [==============================] - 11s 109ms/step - loss: 1.7589 - accuracy: 0.5417
Epoch 11/25
98/98 [==============================] - 11s 109ms/step - loss: 2.2731 - accuracy: 0.4491
Epoch 12/25
98/98 [==============================] - 11s 109ms/step - loss: 2.7997 - accuracy: 0.3268
Epoch 13/25
98/98 [==============================] - 11s 109ms/step - loss: 2.1761 - accuracy: 0.4382
Epoch 14/25
98/98 [==============================] - 11s 109ms/step - loss: 1.6379 - accuracy: 0.5578
Epoch 15/25
98/98 [==============================] - 11s 109ms/step - loss: 1.3593 - accuracy: 0.6270
Epoch 16/25
98/98 [==============================] - 11s 109ms/step - loss: 1.0692 - accuracy: 0.7036
Epoch 17/25
98/98 [==============================] - 11s 109ms/step - loss: 1.8630 - accuracy: 0.5141
Epoch 18/25
98/98 [==============================] - 11s 109ms/step - loss: 1.3677 - accuracy: 0.6197
Epoch 19/25
98/98 [==============================] - 11s 109ms/step - loss: 1.1256 - accuracy: 0.6974
Epoch 20/25
98/98 [==============================] - 11s 109ms/step - loss: 0.9264 - accuracy: 0.7388
Epoch 21/25
98/98 [==============================] - 11s 109ms/step - loss: 0.5641 - accuracy: 0.8499
Epoch 22/25
98/98 [==============================] - 11s 109ms/step - loss: 1.7220 - accuracy: 0.5913
Epoch 23/25
98/98 [==============================] - 11s 109ms/step - loss: 2.2638 - accuracy: 0.4121
Epoch 24/25
98/98 [==============================] - 11s 109ms/step - loss: 1.5651 - accuracy: 0.5716
Epoch 25/25
98/98 [==============================] - 11s 109ms/step - loss: 1.3241 - accuracy: 0.6348

obriensystems · 2024-12-01T00:42:14Z

256, 512,1024,2048,4096 batch_size

(venv-metal) michaelobrien@mbp8 src % vi tflow.py               
(venv-metal) michaelobrien@mbp8 src % python3 tflow.py
/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/urllib3/__init__.py:35: NotOpenSSLWarning: urllib3 v2 only supports OpenSSL 1.1.1+, currently the 'ssl' module is compiled with 'LibreSSL 2.8.3'. See: https://github.com/urllib3/urllib3/issues/3020
  warnings.warn(
2024-11-30 18:59:07.269549: I metal_plugin/src/device/metal_device.cc:1154] Metal device set to: Apple M4 Max
2024-11-30 18:59:07.269568: I metal_plugin/src/device/metal_device.cc:296] systemMemory: 48.00 GB
2024-11-30 18:59:07.269573: I metal_plugin/src/device/metal_device.cc:313] maxCacheSize: 18.00 GB
2024-11-30 18:59:07.269614: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:306] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.
2024-11-30 18:59:07.269829: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:272] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)
2024-11-30 18:59:08.139869: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:117] Plugin optimizer for device_type GPU is enabled.
Epoch 1/25
196/196 [==============================] - 14s 62ms/step - loss: 4.3166 - accuracy: 0.0849
Epoch 2/25
196/196 [==============================] - 12s 60ms/step - loss: 3.7339 - accuracy: 0.1702
Epoch 3/25
196/196 [==============================] - 12s 60ms/step - loss: 3.5242 - accuracy: 0.2091
Epoch 4/25
196/196 [==============================] - 12s 60ms/step - loss: 4.1585 - accuracy: 0.1424
Epoch 5/25
196/196 [==============================] - 12s 60ms/step - loss: 3.8899 - accuracy: 0.1484
Epoch 6/25
196/196 [==============================] - 12s 60ms/step - loss: 3.4270 - accuracy: 0.2054
Epoch 7/25
196/196 [==============================] - 12s 60ms/step - loss: 3.3056 - accuracy: 0.2240
Epoch 8/25
196/196 [==============================] - 12s 60ms/step - loss: 3.6002 - accuracy: 0.1929
Epoch 9/25
196/196 [==============================] - 12s 60ms/step - loss: 3.7808 - accuracy: 0.1387
Epoch 10/25
196/196 [==============================] - 12s 60ms/step - loss: 3.2449 - accuracy: 0.2225
Epoch 11/25
196/196 [==============================] - 12s 60ms/step - loss: 3.0649 - accuracy: 0.2551
Epoch 12/25
196/196 [==============================] - 12s 60ms/step - loss: 2.8943 - accuracy: 0.2914
Epoch 13/25
196/196 [==============================] - 12s 60ms/step - loss: 2.8075 - accuracy: 0.3021
Epoch 14/25
196/196 [==============================] - 12s 60ms/step - loss: 2.6611 - accuracy: 0.3358
Epoch 15/25
196/196 [==============================] - 12s 60ms/step - loss: 2.4845 - accuracy: 0.3687
Epoch 16/25
196/196 [==============================] - 12s 60ms/step - loss: 2.3530 - accuracy: 0.3919
Epoch 17/25
196/196 [==============================] - 12s 60ms/step - loss: 2.2040 - accuracy: 0.4248
Epoch 18/25
196/196 [==============================] - 12s 60ms/step - loss: 2.2955 - accuracy: 0.4202
Epoch 19/25
196/196 [==============================] - 12s 60ms/step - loss: 2.7651 - accuracy: 0.3448
Epoch 20/25
196/196 [==============================] - 12s 60ms/step - loss: 2.8672 - accuracy: 0.2938
Epoch 21/25
196/196 [==============================] - 12s 60ms/step - loss: 2.2576 - accuracy: 0.4118
Epoch 22/25
196/196 [==============================] - 12s 60ms/step - loss: 1.9563 - accuracy: 0.4760
Epoch 23/25
196/196 [==============================] - 12s 60ms/step - loss: 1.7246 - accuracy: 0.5284
Epoch 24/25
196/196 [==============================] - 12s 60ms/step - loss: 1.5055 - accuracy: 0.5827
Epoch 25/25
196/196 [==============================] - 12s 60ms/step - loss: 1.2927 - accuracy: 0.6363
(venv-metal) michaelobrien@mbp8 src % vi tflow.py     
(venv-metal) michaelobrien@mbp8 src % python3 tflow.py
/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/urllib3/__init__.py:35: NotOpenSSLWarning: urllib3 v2 only supports OpenSSL 1.1.1+, currently the 'ssl' module is compiled with 'LibreSSL 2.8.3'. See: https://github.com/urllib3/urllib3/issues/3020
  warnings.warn(
2024-11-30 19:04:28.558501: I metal_plugin/src/device/metal_device.cc:1154] Metal device set to: Apple M4 Max
2024-11-30 19:04:28.558523: I metal_plugin/src/device/metal_device.cc:296] systemMemory: 48.00 GB
2024-11-30 19:04:28.558527: I metal_plugin/src/device/metal_device.cc:313] maxCacheSize: 18.00 GB
2024-11-30 19:04:28.558555: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:306] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.
2024-11-30 19:04:28.558574: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:272] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)
2024-11-30 19:04:29.359890: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:117] Plugin optimizer for device_type GPU is enabled.
Epoch 1/25
98/98 [==============================] - 13s 113ms/step - loss: 4.3169 - accuracy: 0.0829
Epoch 2/25
98/98 [==============================] - 11s 109ms/step - loss: 3.5121 - accuracy: 0.1868
Epoch 3/25
98/98 [==============================] - 11s 109ms/step - loss: 3.1357 - accuracy: 0.2492
Epoch 4/25
98/98 [==============================] - 11s 109ms/step - loss: 2.9869 - accuracy: 0.2812
Epoch 5/25
98/98 [==============================] - 11s 109ms/step - loss: 3.0041 - accuracy: 0.2823
Epoch 6/25
98/98 [==============================] - 11s 109ms/step - loss: 3.2223 - accuracy: 0.2363
Epoch 7/25
98/98 [==============================] - 11s 109ms/step - loss: 2.8993 - accuracy: 0.3102
Epoch 8/25
98/98 [==============================] - 11s 109ms/step - loss: 2.7331 - accuracy: 0.3349
Epoch 9/25
98/98 [==============================] - 11s 109ms/step - loss: 3.1282 - accuracy: 0.2664
Epoch 10/25
98/98 [==============================] - 11s 109ms/step - loss: 3.0092 - accuracy: 0.2700
Epoch 11/25
98/98 [==============================] - 11s 109ms/step - loss: 2.6229 - accuracy: 0.3509
Epoch 12/25
98/98 [==============================] - 11s 109ms/step - loss: 3.6875 - accuracy: 0.1766
Epoch 13/25
98/98 [==============================] - 11s 109ms/step - loss: 2.9881 - accuracy: 0.2714
Epoch 14/25
98/98 [==============================] - 11s 109ms/step - loss: 2.6661 - accuracy: 0.3304
Epoch 15/25
98/98 [==============================] - 11s 109ms/step - loss: 2.6709 - accuracy: 0.3451
Epoch 16/25
98/98 [==============================] - 11s 109ms/step - loss: 2.3870 - accuracy: 0.3919
Epoch 17/25
98/98 [==============================] - 11s 109ms/step - loss: 2.1721 - accuracy: 0.4346
Epoch 18/25
98/98 [==============================] - 11s 109ms/step - loss: 1.9957 - accuracy: 0.4729
Epoch 19/25
98/98 [==============================] - 11s 109ms/step - loss: 1.9012 - accuracy: 0.4963
Epoch 20/25
98/98 [==============================] - 11s 109ms/step - loss: 1.6519 - accuracy: 0.5526
Epoch 21/25
98/98 [==============================] - 11s 109ms/step - loss: 1.4057 - accuracy: 0.6100
Epoch 22/25
98/98 [==============================] - 11s 109ms/step - loss: 1.2087 - accuracy: 0.6612
Epoch 23/25
98/98 [==============================] - 11s 109ms/step - loss: 0.9578 - accuracy: 0.7233
Epoch 24/25
98/98 [==============================] - 11s 109ms/step - loss: 0.8571 - accuracy: 0.7589
Epoch 25/25
98/98 [==============================] - 11s 109ms/step - loss: 0.6257 - accuracy: 0.8249
(venv-metal) michaelobrien@mbp8 src % vi tflow.py     
(venv-metal) michaelobrien@mbp8 src % python3 tflow.py
/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/urllib3/__init__.py:35: NotOpenSSLWarning: urllib3 v2 only supports OpenSSL 1.1.1+, currently the 'ssl' module is compiled with 'LibreSSL 2.8.3'. See: https://github.com/urllib3/urllib3/issues/3020
  warnings.warn(
2024-11-30 19:25:30.682030: I metal_plugin/src/device/metal_device.cc:1154] Metal device set to: Apple M4 Max
2024-11-30 19:25:30.682049: I metal_plugin/src/device/metal_device.cc:296] systemMemory: 48.00 GB
2024-11-30 19:25:30.682053: I metal_plugin/src/device/metal_device.cc:313] maxCacheSize: 18.00 GB
2024-11-30 19:25:30.682097: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:306] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.
2024-11-30 19:25:30.682308: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:272] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)
2024-11-30 19:25:31.504996: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:117] Plugin optimizer for device_type GPU is enabled.
Epoch 1/25
49/49 [==============================] - 13s 221ms/step - loss: 4.6072 - accuracy: 0.0570
Epoch 2/25
49/49 [==============================] - 11s 216ms/step - loss: 3.8497 - accuracy: 0.1316
Epoch 3/25
49/49 [==============================] - 11s 216ms/step - loss: 3.4675 - accuracy: 0.1845
Epoch 4/25
49/49 [==============================] - 11s 220ms/step - loss: 3.0766 - accuracy: 0.2551
Epoch 5/25
49/49 [==============================] - 11s 223ms/step - loss: 2.8016 - accuracy: 0.3089
Epoch 6/25
49/49 [==============================] - 11s 221ms/step - loss: 2.4840 - accuracy: 0.3695
Epoch 7/25
49/49 [==============================] - 11s 219ms/step - loss: 2.2145 - accuracy: 0.4225
Epoch 8/25
49/49 [==============================] - 11s 216ms/step - loss: 2.0393 - accuracy: 0.4656
Epoch 9/25
49/49 [==============================] - 11s 216ms/step - loss: 2.0956 - accuracy: 0.4593
Epoch 10/25
49/49 [==============================] - 11s 216ms/step - loss: 1.7062 - accuracy: 0.5415
Epoch 11/25
49/49 [==============================] - 11s 216ms/step - loss: 1.3635 - accuracy: 0.6333
Epoch 12/25
49/49 [==============================] - 11s 216ms/step - loss: 1.7514 - accuracy: 0.5307
Epoch 13/25
49/49 [==============================] - 11s 216ms/step - loss: 1.1953 - accuracy: 0.6706
Epoch 14/25
49/49 [==============================] - 11s 216ms/step - loss: 0.8726 - accuracy: 0.7557
Epoch 15/25
49/49 [==============================] - 11s 216ms/step - loss: 1.0864 - accuracy: 0.6962
Epoch 16/25
49/49 [==============================] - 11s 216ms/step - loss: 0.7288 - accuracy: 0.7970
Epoch 17/25
49/49 [==============================] - 11s 216ms/step - loss: 0.4807 - accuracy: 0.8717
Epoch 18/25
49/49 [==============================] - 11s 216ms/step - loss: 0.3632 - accuracy: 0.9041
Epoch 19/25
49/49 [==============================] - 11s 216ms/step - loss: 0.2960 - accuracy: 0.9223
Epoch 20/25
49/49 [==============================] - 11s 216ms/step - loss: 0.2486 - accuracy: 0.9354
Epoch 21/25
49/49 [==============================] - 11s 216ms/step - loss: 0.2650 - accuracy: 0.9301
Epoch 22/25
49/49 [==============================] - 11s 216ms/step - loss: 0.2625 - accuracy: 0.9299
Epoch 23/25
49/49 [==============================] - 11s 216ms/step - loss: 0.2850 - accuracy: 0.9251
Epoch 24/25
49/49 [==============================] - 11s 216ms/step - loss: 0.4320 - accuracy: 0.8790
Epoch 25/25
49/49 [==============================] - 11s 216ms/step - loss: 0.2794 - accuracy: 0.9247
(venv-metal) michaelobrien@mbp8 src % vi tflow.py     
(venv-metal) michaelobrien@mbp8 src % python3 tflow.py
/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/urllib3/__init__.py:35: NotOpenSSLWarning: urllib3 v2 only supports OpenSSL 1.1.1+, currently the 'ssl' module is compiled with 'LibreSSL 2.8.3'. See: https://github.com/urllib3/urllib3/issues/3020
  warnings.warn(
2024-11-30 19:30:34.785131: I metal_plugin/src/device/metal_device.cc:1154] Metal device set to: Apple M4 Max
2024-11-30 19:30:34.785153: I metal_plugin/src/device/metal_device.cc:296] systemMemory: 48.00 GB
2024-11-30 19:30:34.785158: I metal_plugin/src/device/metal_device.cc:313] maxCacheSize: 18.00 GB
2024-11-30 19:30:34.785188: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:306] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.
2024-11-30 19:30:34.785207: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:272] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)
2024-11-30 19:30:35.603669: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:117] Plugin optimizer for device_type GPU is enabled.
Epoch 1/25
25/25 [==============================] - 13s 425ms/step - loss: 5.1168 - accuracy: 0.0420
Epoch 2/25
25/25 [==============================] - 10s 414ms/step - loss: 3.8564 - accuracy: 0.1196
Epoch 3/25
25/25 [==============================] - 10s 414ms/step - loss: 3.4055 - accuracy: 0.1964
Epoch 4/25
25/25 [==============================] - 10s 414ms/step - loss: 2.9654 - accuracy: 0.2759
Epoch 5/25
25/25 [==============================] - 10s 414ms/step - loss: 2.6138 - accuracy: 0.3511
Epoch 6/25
25/25 [==============================] - 10s 415ms/step - loss: 2.2539 - accuracy: 0.4231
Epoch 7/25
25/25 [==============================] - 10s 415ms/step - loss: 1.9519 - accuracy: 0.4860
Epoch 8/25
25/25 [==============================] - 10s 415ms/step - loss: 1.6168 - accuracy: 0.5627
Epoch 9/25
25/25 [==============================] - 10s 415ms/step - loss: 1.7255 - accuracy: 0.5454
Epoch 10/25
25/25 [==============================] - 10s 417ms/step - loss: 1.2815 - accuracy: 0.6428
Epoch 11/25
25/25 [==============================] - 10s 415ms/step - loss: 0.9566 - accuracy: 0.7296
Epoch 12/25
25/25 [==============================] - 10s 415ms/step - loss: 0.7596 - accuracy: 0.7849
Epoch 13/25
25/25 [==============================] - 10s 414ms/step - loss: 0.6607 - accuracy: 0.8111
Epoch 14/25
25/25 [==============================] - 10s 414ms/step - loss: 0.5659 - accuracy: 0.8399
Epoch 15/25
25/25 [==============================] - 10s 414ms/step - loss: 0.4634 - accuracy: 0.8636
Epoch 16/25
25/25 [==============================] - 10s 414ms/step - loss: 0.4024 - accuracy: 0.8823
Epoch 17/25
25/25 [==============================] - 10s 414ms/step - loss: 0.3646 - accuracy: 0.8923
Epoch 18/25
25/25 [==============================] - 10s 415ms/step - loss: 0.3800 - accuracy: 0.8972
Epoch 19/25
25/25 [==============================] - 10s 415ms/step - loss: 0.4540 - accuracy: 0.8777
Epoch 20/25
25/25 [==============================] - 10s 414ms/step - loss: 0.4469 - accuracy: 0.8700
Epoch 21/25
25/25 [==============================] - 10s 414ms/step - loss: 1.5743 - accuracy: 0.5828
Epoch 22/25
25/25 [==============================] - 10s 415ms/step - loss: 0.7500 - accuracy: 0.7846
Epoch 23/25
25/25 [==============================] - 10s 414ms/step - loss: 0.3989 - accuracy: 0.8921
Epoch 24/25
25/25 [==============================] - 10s 415ms/step - loss: 0.2976 - accuracy: 0.9297
Epoch 25/25
25/25 [==============================] - 10s 414ms/step - loss: 0.2549 - accuracy: 0.9444
(venv-metal) michaelobrien@mbp8 src % vi tflow.py     
(venv-metal) michaelobrien@mbp8 src % python3 tflow.py
/Users/michaelobrien/wse_github/ObrienlabsDev/machine-learning/environments/windows/src/venv-metal/lib/python3.9/site-packages/urllib3/__init__.py:35: NotOpenSSLWarning: urllib3 v2 only supports OpenSSL 1.1.1+, currently the 'ssl' module is compiled with 'LibreSSL 2.8.3'. See: https://github.com/urllib3/urllib3/issues/3020
  warnings.warn(
2024-11-30 19:35:35.011755: I metal_plugin/src/device/metal_device.cc:1154] Metal device set to: Apple M4 Max
2024-11-30 19:35:35.011774: I metal_plugin/src/device/metal_device.cc:296] systemMemory: 48.00 GB
2024-11-30 19:35:35.011779: I metal_plugin/src/device/metal_device.cc:313] maxCacheSize: 18.00 GB
2024-11-30 19:35:35.011804: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:306] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.
2024-11-30 19:35:35.011823: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:272] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)
2024-11-30 19:35:35.831684: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:117] Plugin optimizer for device_type GPU is enabled.
Epoch 1/25
13/13 [==============================] - 13s 808ms/step - loss: 5.3305 - accuracy: 0.0285
Epoch 2/25
13/13 [==============================] - 10s 783ms/step - loss: 4.1541 - accuracy: 0.0765
Epoch 3/25
13/13 [==============================] - 10s 786ms/step - loss: 3.7416 - accuracy: 0.1293
Epoch 4/25
13/13 [==============================] - 10s 785ms/step - loss: 3.3820 - accuracy: 0.1927
Epoch 5/25
13/13 [==============================] - 10s 788ms/step - loss: 3.0269 - accuracy: 0.2626
Epoch 6/25
13/13 [==============================] - 10s 789ms/step - loss: 2.6507 - accuracy: 0.3312
Epoch 7/25
13/13 [==============================] - 10s 789ms/step - loss: 2.2682 - accuracy: 0.4159
Epoch 8/25
13/13 [==============================] - 10s 785ms/step - loss: 1.9441 - accuracy: 0.4878
Epoch 9/25
13/13 [==============================] - 10s 787ms/step - loss: 1.6875 - accuracy: 0.5436
Epoch 10/25
13/13 [==============================] - 10s 784ms/step - loss: 1.4218 - accuracy: 0.6095
Epoch 11/25
13/13 [==============================] - 10s 785ms/step - loss: 1.1683 - accuracy: 0.6699
Epoch 12/25
13/13 [==============================] - 10s 785ms/step - loss: 0.9268 - accuracy: 0.7342
Epoch 13/25
13/13 [==============================] - 10s 784ms/step - loss: 0.8257 - accuracy: 0.7571
Epoch 14/25
13/13 [==============================] - 10s 784ms/step - loss: 0.6906 - accuracy: 0.7925
Epoch 15/25
13/13 [==============================] - 10s 784ms/step - loss: 0.5851 - accuracy: 0.8239
Epoch 16/25
13/13 [==============================] - 10s 784ms/step - loss: 0.4746 - accuracy: 0.8566
Epoch 17/25
13/13 [==============================] - 10s 785ms/step - loss: 0.4323 - accuracy: 0.8680
Epoch 18/25
13/13 [==============================] - 10s 784ms/step - loss: 0.3555 - accuracy: 0.8920
Epoch 19/25
13/13 [==============================] - 10s 783ms/step - loss: 0.3420 - accuracy: 0.8962
Epoch 20/25
13/13 [==============================] - 10s 784ms/step - loss: 0.3003 - accuracy: 0.9092
Epoch 21/25
13/13 [==============================] - 10s 784ms/step - loss: 0.2264 - accuracy: 0.9317
Epoch 22/25
13/13 [==============================] - 10s 784ms/step - loss: 0.1771 - accuracy: 0.9477
Epoch 23/25
13/13 [==============================] - 10s 785ms/step - loss: 0.1598 - accuracy: 0.9514
Epoch 24/25
13/13 [==============================] - 10s 785ms/step - loss: 0.1485 - accuracy: 0.9553
Epoch 25/25
13/13 [==============================] - 10s 784ms/step - loss: 0.1351 - accuracy: 0.9595

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Tensorflow 2 on Metal - Apple Silicon M4 Max 12/4 40 core GPU #32

Tensorflow 2 on Metal - Apple Silicon M4 Max 12/4 40 core GPU #32

obriensystems commented Nov 29, 2024 •

edited

Loading

obriensystems commented Nov 30, 2024 •

edited

Loading

obriensystems commented Dec 1, 2024

Tensorflow 2 on Metal - Apple Silicon M4 Max 12/4 40 core GPU #32

Tensorflow 2 on Metal - Apple Silicon M4 Max 12/4 40 core GPU #32

Comments

obriensystems commented Nov 29, 2024 • edited Loading

obriensystems commented Nov 30, 2024 • edited Loading

obriensystems commented Dec 1, 2024

obriensystems commented Nov 29, 2024 •

edited

Loading

obriensystems commented Nov 30, 2024 •

edited

Loading