Comments (4)
That worked. Thanks. Maybe I was using the mlx data that I had built locally. And the cigar example works.
...
Epoch 00 [080] | Train loss 1.609 | Train acc 0.410 | Throughput: 2588.24 images/second
Epoch 00 [090] | Train loss 1.580 | Train acc 0.359 | Throughput: 2588.09 images/second
Epoch 00 [100] | Train loss 1.635 | Train acc 0.398 | Throughput: 2557.45 images/second
Epoch 00 [110] | Train loss 1.408 | Train acc 0.488 | Throughput: 2566.57 images/second
Epoch 00 [120] | Train loss 1.410 | Train acc 0.500 | Throughput: 2580.58 images/second
Epoch 00 [130] | Train loss 1.477 | Train acc 0.438 | Throughput: 2565.19 images/second
Epoch 00 [140] | Train loss 1.396 | Train acc 0.488 | Throughput: 2588.39 images/second
Epoch 00 [150] | Train loss 1.426 | Train acc 0.449 | Throughput: 2577.05 images/second
Epoch 00 [160] | Train loss 1.430 | Train acc 0.488 | Throughput: 2562.38 images/second
Epoch 00 [170] | Train loss 1.370 | Train acc 0.484 | Throughput: 2587.62 images/second
Epoch 00 [180] | Train loss 1.362 | Train acc 0.480 | Throughput: 2595.86 images/second
Epoch 00 [190] | Train loss 1.154 | Train acc 0.590 | Throughput: 2579.88 images/second
...
If you understand what was happening, perhaps the installation from source documentation could be updated.
from mlx-examples.
Wow. mlx seems to be getting faster:
Epoch 01 [010] | Train loss 1.266 | Train acc 0.539 | Throughput: 4963.58 images/second
Epoch 01 [020] | Train loss 1.248 | Train acc 0.555 | Throughput: 4992.35 images/second
Epoch 01 [030] | Train loss 1.335 | Train acc 0.484 | Throughput: 4811.33 images/second
Epoch 01 [040] | Train loss 1.363 | Train acc 0.535 | Throughput: 4922.65 images/second
Epoch 01 [050] | Train loss 1.364 | Train acc 0.484 | Throughput: 4850.03 images/second
Epoch 01 [060] | Train loss 1.179 | Train acc 0.570 | Throughput: 4912.40 images/second
Epoch 01 [070] | Train loss 1.214 | Train acc 0.598 | Throughput: 4818.70 images/second
Epoch 01 [080] | Train loss 1.174 | Train acc 0.590 | Throughput: 4917.31 images/second
Epoch 01 [090] | Train loss 1.152 | Train acc 0.586 | Throughput: 4900.87 images/second
Epoch 01 [100] | Train loss 1.167 | Train acc 0.566 | Throughput: 4939.23 images/second
Epoch 01 [110] | Train loss 1.102 | Train acc 0.570 | Throughput: 4930.13 images/second
Epoch 01 [120] | Train loss 1.100 | Train acc 0.625 | Throughput: 4845.19 images/second
Epoch 01 [130] | Train loss 1.178 | Train acc 0.594 | Throughput: 4858.89 images/second
Epoch 01 [140] | Train loss 1.052 | Train acc 0.590 | Throughput: 4864.82 images/second
from mlx-examples.
Can you try pip install -U mlx-data
and then
python -c "from mlx.data.datasets import load_cifar10"
from mlx-examples.
More to come!
from mlx-examples.
Related Issues (20)
- GatedRepoError: 401 Client Error; "You must be authenticated to access it." HOT 1
- [Feature Request] When generating using mlx_lm, specify data format HOT 2
- how to merge lora adapter to base model HOT 1
- delete and uninstall HOT 11
- KV Cache can only process more than self.step tokens if offset % step == 0 HOT 2
- Text to Speech MLX model. HOT 1
- SLM Example Code HOT 1
- Enhance load function to support model configuration editing HOT 1
- Support for full set of output formats - e.g. vtt, json and json-full HOT 2
- Whisper stutters HOT 8
- mlx 0.13 very slow with q8 and fp16 HOT 5
- Fine tuned a Mixtral-8x7B-Instruct-v0.1 model and unable to load with AutoModelForCausalLM HOT 1
- Phi-3-mini-4k-instruct : Failing to stop at <|end|> on generating the answer. HOT 5
- PaliGemma 4bit Quantization broken and Inference issues. HOT 27
- [Feature Request] Function Calling for mlx_lm.server HOT 4
- OS system requirement for mlx HOT 1
- 01-ai/Yi-1.5-9B-Chat got ValueError: Cannot instantiate this tokenizer from a slow version. HOT 4
- CLIP Tokenizer unable to take text of unequal length (or token length) HOT 7
- Package 'mlx_whisper.assets' is absent from the `packages` configuration HOT 1
- [Feature Request] Add support for logprobs to the mlx_lm server HOT 3
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from mlx-examples.